• Home

  • Documents

  • Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs...
126
Precision Cancer Medicine OSLO 2016 Alexander EGGERMONT. MD, PhD Gustave Roussy Comprehensive Cancer Center Cancer Campus Grand Paris, France

Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

  • Upload
    others

  • View
    2

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Precision Cancer Medicine OSLO 2016

Alexander EGGERMONT MD PhD

Gustave Roussy Comprehensive Cancer Center

Cancer Campus Grand Paris France

Key Clinical Missions

bull Tertiary Highly Complex Medicine

bull Rare Tumors ( gt 25)

bull Care integrated with Clinical Research (~3800 ptsyr in Clinical Trials)

bull Early Drug Development ndash (900 pts included in 2015 ~25 of CR program)

KEY MISSION INNOVATE and create ACCESS TO INNOVATION

INTEGRATION of RESEARCH and CARE to create TOMORROWS MEDICINE

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

Molecular Alterations in Melanoma

FGFR

PTEN

PI3K Akt

TOR

KIT

GRB2

SOS Ras

GDP C-Raf

N-Ras

GTP B-Raf

MEK

ERK

ELK

MITF

CDK24

Cyclin D

p16

Amplified

in 30

Amplified

in 30

50-65

V600E

mutation

15 mutation

Amplified or mutated in 20-40

acral and mucosal melanoma

25-50 loss

Frequent loss

Amplified in 10-15

Adapted from Sosman Curr Oncol Rep 11 405 (2009)

Vemurafenib and Dabrafenib show similar efficacy

Dabrafenib

Hauschild et al

Lancet 2012

Chapman et al

NEJM 2011

(n=337)

(n=187)

Vemurafenib DTIC

6

OS 97-136 mts Gain 39 mts HR 070

V600E (91) 133-100 HR = 075 V600K(9) 145ndash 76 HR = 043

PFS 16- 69 mts Gain 53 mts

HR 038

SUCCESS AND FAILURE

Molecular Alterations in Melanoma

BRAF + MEK Inhibitors

FGFR

PTEN

PI3K Akt

TOR

KIT

GRB2

SOS Ras

GDP C-Raf

N-Ras

GTP B-Raf

MEK

ERK

ELK

MITF

CDK24

Cyclin D

p16

Amplified

in 30

Amplified

in 30

50-65

V600E

mutation

15 mutation

Amplified or mutated in 20-40

acral and mucosal melanoma

25-50 loss

Frequent loss

Amplified in 10-15

Adapted from Sosman Curr Oncol Rep 11 405 (2009)

Median Follow-up D + T = 11 months and Vem = 10 months

asymp 8 weeks

A SOMEWHAT SOBERING END RESULT

bull Tumor by evolution is ldquomoving targetrdquo

bull Heterogeneity and Innate resistance

bull Acquired resistanceAdditional mutations

bull Mono-Dimensional Thinking

We need to address

bull Nodes of Convergence of Pathways

bull Cross Talk Complexity between pathways

Drug Development Challenges

eIF4F Node of Convergence

RasRaf ndash PI3K- Caspase cascade

FGFR

PTEN

PI3K Akt

TOR

KIT

GRB2

SOS Ras

GDP C-Raf

N-Ras

GTP B-Raf

MEK

ERK

ELK

MITF

CDK24

Cyclin D

p16

Amplified

in 30

Amplified

in 30

50-65

V600E

mutation

15 mutation

Amplified or mutated in 20-40

acral and mucosal melanoma

25-50 loss

Frequent loss

Amplified in 10-15

Adapted from Sosman Curr Oncol Rep 11 405 (2009)

Caspase cascade

Nexus eIF4F

12

bull Nodes of Convergence of Pathways

bull Cross Talk Complexity between pathways

Drug Development Challenges

81

520

0

10

20

30

40

50

60

70

80

90

Melanoma Colon cancer

N Engl J Med 2010 363809-19

Kopetz et al ASCO 2010

Differential response of BRAF inhibition in

BRAF mutant melanoma vs colon cancer

CROSS TALK and RESISTANCE

Prahallad et alhelliphellipBernards R Nature 2012

No drug

EGFR inhibitor

BRAF inhibitor

BRAF+EGFR

inhibitor

Human colon cancer growth in mice

16

bull Targeted Agents will be effective accross different tumor types with that target

ndash importance of organ of origin

ndashAnswer is NO

bull For combination therapies one must demonstrate the antitumor effects for each individual agent

ndashAnswer is NO

PUTS AN END TO TWO PARADIGMS

Molecular profiling

Identification of the molecular alteration

Targeted therapy according to the molecular profile

Tumor Specimen

Precision Medicine identify-hit the target

GUSTAVE ROUSSY PCM PROGRAM

Rational Genomics lsquorsquoMolecular Portraitsrsquorsquo for targeted therapy allocation Rapid through Clinical Trials Logistics ndash Logistics ndash Logistics Focus time pressure culture infrastructure Examples of Current Clinical Trials

SAFIR01 Molecular Screening in

Breast Cancer

Which candidate target FGFR1

FGFR2

FGF4 amp

NOTCH amp

Genetic

instability

PAK1 ampli

PIK3CA AKT PTEN

alteration

VEGFA

amplification

Target

discovery

Primary endpoint

of patients included

in phase III trial according to the

profile

Biopsy of metastatic sites

Frozen sample

CGHhot spot mutations

(PIK3CAAKT)

eligible for phase I

N= 400

Trial A

Trial F

Trial E

Trial D

Trial C

Trial B

Trials XYhellip

Funded by INCA Sanger 30 genes

Andreacute et al ESMO 2012 Lancet Oncol 2014

High level of expectation

Andreacute ESMO 2012

Inclusions

Recruitment completed between May 2011 and August 2012

Molecular alterations

Targetable alterations

Rare alterations

0

5

10

15

20

25

o

f p

atie

nts

wit

h a

vaila

ble

gen

om

ic r

esu

lt mutations

copy number alterations

Andreacute et al Lancet Oncology 2014

29 molecular alteration

IN THE END ONLY 12

gets targeted therapy

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

1200 PATIENTS IN 4 Years

MATCH-R trial

In vitro Cell lines Mouse avatar

Patients with + biomarker tumor exposed to a targeted therapy and an initial response

Resistant Sensitive

Tumor biopy or effusion

Biopsy metastatic site NGS

Array CGH

Chemotherapy 4 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

EGFR TKI if SCC

No PD metastatic NSCLC fisrt line chemotherapy

RANDOMIZED TRIAL SAFIR 02 LUNG

All histologies

Pemetrexed if Non-SCC Molecular alteration Excluding EGFR mut and ALK trl

Genetic abnormality Gene location Squamous Cell

Carcinoma Adenocarcinoma

Therapeutic

intrevention

PIK3CA amplification[ 3q263 33 6 AZD2014

(TORC12)

FGFR1 amplification[ 8p12 22 1 AZD4547

(FGFR)

PTEN mutation 10q233 10 2 AZD8186 (PI3Kbeta)

MET amplification 7q311 3-21 3-21 Volitinib

PTEN loss 10q233 8-20 8-20 AZD8186 (PI3Kbeta)

KRAS mutation[ 12p121 6 21

AZD6244

(MEKi)

Or combo with

AZD2014

LKB1 mutation 19p133 5 23 AZD2014

(TORC12)

HER 2 amplification 17q112-q12 17q21 3-5 5-9 AZD 8931

(pan-HER)

PIK3CA mutation 3q263 3 3 AZD2014

(TORC12)

RET translocation 10q112 2 1

AZD6474

(VEGFR EGFR RET)

BRAF mutation 7p34 2 1-3 AZD6244

(MEKi)

AKT1 mutation 14q3232 1 Very rare AZD5363

(Akt)

MET mutation 7q311 1 2 Volitinib

HER2 mutation 17q112-q12 17q21 1 2 AZD 8931

(pan-HER)

Get COMPLETE PIPELINES

Biopsy metastatic site NGS

CGH 51 alterations

Chemotherapy 6-8 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

Chemotherapy continuation

No PD metastatic Breast cancer patient 1st or

2nd line

RANDOMIZED TRIAL SAFIR 02 BREAST

Molecular alteration Excluding HER2

SAFIR02

BreastAndre

520600

Since 2010 Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics)

SAFIR01

(Andre)

427427

MOSCATO

(Soria)

11501200

SAFIR02

LungSoria

480600

MOST

preSAFIR

(Andre)

108108

SHIVA

(Letourneau

Pilot study 1st Gener Trials 2nd Gener

No NGS NGS WES Randomized trials Sponsor

Gustave Roussy monocentric

Gustave RoussyUnicancer multicenter

L BeacuterardLyon

Curie

Unified Database Pick-up

the winner targets

2nd generation Algorithm for Personnalized

medicine

WINTHER

150200

Profiler

MATCH-R

(WES PDX cfDNA)

200600

FUNDING TOTAL ~50 Million Fondation GR (10) MCM Building (15) IHU (6) INCA (6) ARC (4) Philanthropia (2) WINconsort (4) EU-FP7 (3)

MSN LungMel

BesseRobert

600600

Gustave RoussyWIN multicenter

MOSKIDO

(Goerger)

80100

GETUG

Fizazi

3580 ACSE

Vassal

600

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

12001200 PATIENTS IN 35 Years

bull ATTRITION RATE

bull 100 pts with molecular portrait

bull 50 pts have target identified

bull 25 are actionable

bull 25 overall response rate

bull So in the end 6100 patients benefithellip

bull INNATE RESISTANCE + ORGAN CONTEXT

PROBLEMS with

Molecular Portraits

NEEDS

bull Higher Target Identification Rate

bull Higher of Actionable Targets

bull Higher number diversification of new drugs

bull Rational intrainterpathway combinations

bull Functional Genetics (Crosstalk) ndash Rene Bernards

bull Nodes of convergence ndash Vagner Robert

bull Simplification of models ndash Master Regulators (Andrea Califano)

31

Problems with Targeted Drugs

bull Lack of Durability of responses

ndash Improvement with comborsquos limited

ndash Comborsquos of non-active drugs can be

active

bull Lack of Transversality of impact across

tumor types

ndash Organ of origin

ndash Context 32

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

INHIBIT THE INHIBITOR

vs ACTIVATE THE ACTIVATOR

BREAKING TOLERANCE Immune-Checkpoint-Blockers

REVOLUTION IN IMMUNOTHERAPY

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 2: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Key Clinical Missions

bull Tertiary Highly Complex Medicine

bull Rare Tumors ( gt 25)

bull Care integrated with Clinical Research (~3800 ptsyr in Clinical Trials)

bull Early Drug Development ndash (900 pts included in 2015 ~25 of CR program)

KEY MISSION INNOVATE and create ACCESS TO INNOVATION

INTEGRATION of RESEARCH and CARE to create TOMORROWS MEDICINE

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

Molecular Alterations in Melanoma

FGFR

PTEN

PI3K Akt

TOR

KIT

GRB2

SOS Ras

GDP C-Raf

N-Ras

GTP B-Raf

MEK

ERK

ELK

MITF

CDK24

Cyclin D

p16

Amplified

in 30

Amplified

in 30

50-65

V600E

mutation

15 mutation

Amplified or mutated in 20-40

acral and mucosal melanoma

25-50 loss

Frequent loss

Amplified in 10-15

Adapted from Sosman Curr Oncol Rep 11 405 (2009)

Vemurafenib and Dabrafenib show similar efficacy

Dabrafenib

Hauschild et al

Lancet 2012

Chapman et al

NEJM 2011

(n=337)

(n=187)

Vemurafenib DTIC

6

OS 97-136 mts Gain 39 mts HR 070

V600E (91) 133-100 HR = 075 V600K(9) 145ndash 76 HR = 043

PFS 16- 69 mts Gain 53 mts

HR 038

SUCCESS AND FAILURE

Molecular Alterations in Melanoma

BRAF + MEK Inhibitors

FGFR

PTEN

PI3K Akt

TOR

KIT

GRB2

SOS Ras

GDP C-Raf

N-Ras

GTP B-Raf

MEK

ERK

ELK

MITF

CDK24

Cyclin D

p16

Amplified

in 30

Amplified

in 30

50-65

V600E

mutation

15 mutation

Amplified or mutated in 20-40

acral and mucosal melanoma

25-50 loss

Frequent loss

Amplified in 10-15

Adapted from Sosman Curr Oncol Rep 11 405 (2009)

Median Follow-up D + T = 11 months and Vem = 10 months

asymp 8 weeks

A SOMEWHAT SOBERING END RESULT

bull Tumor by evolution is ldquomoving targetrdquo

bull Heterogeneity and Innate resistance

bull Acquired resistanceAdditional mutations

bull Mono-Dimensional Thinking

We need to address

bull Nodes of Convergence of Pathways

bull Cross Talk Complexity between pathways

Drug Development Challenges

eIF4F Node of Convergence

RasRaf ndash PI3K- Caspase cascade

FGFR

PTEN

PI3K Akt

TOR

KIT

GRB2

SOS Ras

GDP C-Raf

N-Ras

GTP B-Raf

MEK

ERK

ELK

MITF

CDK24

Cyclin D

p16

Amplified

in 30

Amplified

in 30

50-65

V600E

mutation

15 mutation

Amplified or mutated in 20-40

acral and mucosal melanoma

25-50 loss

Frequent loss

Amplified in 10-15

Adapted from Sosman Curr Oncol Rep 11 405 (2009)

Caspase cascade

Nexus eIF4F

12

bull Nodes of Convergence of Pathways

bull Cross Talk Complexity between pathways

Drug Development Challenges

81

520

0

10

20

30

40

50

60

70

80

90

Melanoma Colon cancer

N Engl J Med 2010 363809-19

Kopetz et al ASCO 2010

Differential response of BRAF inhibition in

BRAF mutant melanoma vs colon cancer

CROSS TALK and RESISTANCE

Prahallad et alhelliphellipBernards R Nature 2012

No drug

EGFR inhibitor

BRAF inhibitor

BRAF+EGFR

inhibitor

Human colon cancer growth in mice

16

bull Targeted Agents will be effective accross different tumor types with that target

ndash importance of organ of origin

ndashAnswer is NO

bull For combination therapies one must demonstrate the antitumor effects for each individual agent

ndashAnswer is NO

PUTS AN END TO TWO PARADIGMS

Molecular profiling

Identification of the molecular alteration

Targeted therapy according to the molecular profile

Tumor Specimen

Precision Medicine identify-hit the target

GUSTAVE ROUSSY PCM PROGRAM

Rational Genomics lsquorsquoMolecular Portraitsrsquorsquo for targeted therapy allocation Rapid through Clinical Trials Logistics ndash Logistics ndash Logistics Focus time pressure culture infrastructure Examples of Current Clinical Trials

SAFIR01 Molecular Screening in

Breast Cancer

Which candidate target FGFR1

FGFR2

FGF4 amp

NOTCH amp

Genetic

instability

PAK1 ampli

PIK3CA AKT PTEN

alteration

VEGFA

amplification

Target

discovery

Primary endpoint

of patients included

in phase III trial according to the

profile

Biopsy of metastatic sites

Frozen sample

CGHhot spot mutations

(PIK3CAAKT)

eligible for phase I

N= 400

Trial A

Trial F

Trial E

Trial D

Trial C

Trial B

Trials XYhellip

Funded by INCA Sanger 30 genes

Andreacute et al ESMO 2012 Lancet Oncol 2014

High level of expectation

Andreacute ESMO 2012

Inclusions

Recruitment completed between May 2011 and August 2012

Molecular alterations

Targetable alterations

Rare alterations

0

5

10

15

20

25

o

f p

atie

nts

wit

h a

vaila

ble

gen

om

ic r

esu

lt mutations

copy number alterations

Andreacute et al Lancet Oncology 2014

29 molecular alteration

IN THE END ONLY 12

gets targeted therapy

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

1200 PATIENTS IN 4 Years

MATCH-R trial

In vitro Cell lines Mouse avatar

Patients with + biomarker tumor exposed to a targeted therapy and an initial response

Resistant Sensitive

Tumor biopy or effusion

Biopsy metastatic site NGS

Array CGH

Chemotherapy 4 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

EGFR TKI if SCC

No PD metastatic NSCLC fisrt line chemotherapy

RANDOMIZED TRIAL SAFIR 02 LUNG

All histologies

Pemetrexed if Non-SCC Molecular alteration Excluding EGFR mut and ALK trl

Genetic abnormality Gene location Squamous Cell

Carcinoma Adenocarcinoma

Therapeutic

intrevention

PIK3CA amplification[ 3q263 33 6 AZD2014

(TORC12)

FGFR1 amplification[ 8p12 22 1 AZD4547

(FGFR)

PTEN mutation 10q233 10 2 AZD8186 (PI3Kbeta)

MET amplification 7q311 3-21 3-21 Volitinib

PTEN loss 10q233 8-20 8-20 AZD8186 (PI3Kbeta)

KRAS mutation[ 12p121 6 21

AZD6244

(MEKi)

Or combo with

AZD2014

LKB1 mutation 19p133 5 23 AZD2014

(TORC12)

HER 2 amplification 17q112-q12 17q21 3-5 5-9 AZD 8931

(pan-HER)

PIK3CA mutation 3q263 3 3 AZD2014

(TORC12)

RET translocation 10q112 2 1

AZD6474

(VEGFR EGFR RET)

BRAF mutation 7p34 2 1-3 AZD6244

(MEKi)

AKT1 mutation 14q3232 1 Very rare AZD5363

(Akt)

MET mutation 7q311 1 2 Volitinib

HER2 mutation 17q112-q12 17q21 1 2 AZD 8931

(pan-HER)

Get COMPLETE PIPELINES

Biopsy metastatic site NGS

CGH 51 alterations

Chemotherapy 6-8 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

Chemotherapy continuation

No PD metastatic Breast cancer patient 1st or

2nd line

RANDOMIZED TRIAL SAFIR 02 BREAST

Molecular alteration Excluding HER2

SAFIR02

BreastAndre

520600

Since 2010 Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics)

SAFIR01

(Andre)

427427

MOSCATO

(Soria)

11501200

SAFIR02

LungSoria

480600

MOST

preSAFIR

(Andre)

108108

SHIVA

(Letourneau

Pilot study 1st Gener Trials 2nd Gener

No NGS NGS WES Randomized trials Sponsor

Gustave Roussy monocentric

Gustave RoussyUnicancer multicenter

L BeacuterardLyon

Curie

Unified Database Pick-up

the winner targets

2nd generation Algorithm for Personnalized

medicine

WINTHER

150200

Profiler

MATCH-R

(WES PDX cfDNA)

200600

FUNDING TOTAL ~50 Million Fondation GR (10) MCM Building (15) IHU (6) INCA (6) ARC (4) Philanthropia (2) WINconsort (4) EU-FP7 (3)

MSN LungMel

BesseRobert

600600

Gustave RoussyWIN multicenter

MOSKIDO

(Goerger)

80100

GETUG

Fizazi

3580 ACSE

Vassal

600

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

12001200 PATIENTS IN 35 Years

bull ATTRITION RATE

bull 100 pts with molecular portrait

bull 50 pts have target identified

bull 25 are actionable

bull 25 overall response rate

bull So in the end 6100 patients benefithellip

bull INNATE RESISTANCE + ORGAN CONTEXT

PROBLEMS with

Molecular Portraits

NEEDS

bull Higher Target Identification Rate

bull Higher of Actionable Targets

bull Higher number diversification of new drugs

bull Rational intrainterpathway combinations

bull Functional Genetics (Crosstalk) ndash Rene Bernards

bull Nodes of convergence ndash Vagner Robert

bull Simplification of models ndash Master Regulators (Andrea Califano)

31

Problems with Targeted Drugs

bull Lack of Durability of responses

ndash Improvement with comborsquos limited

ndash Comborsquos of non-active drugs can be

active

bull Lack of Transversality of impact across

tumor types

ndash Organ of origin

ndash Context 32

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

INHIBIT THE INHIBITOR

vs ACTIVATE THE ACTIVATOR

BREAKING TOLERANCE Immune-Checkpoint-Blockers

REVOLUTION IN IMMUNOTHERAPY

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 3: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

Molecular Alterations in Melanoma

FGFR

PTEN

PI3K Akt

TOR

KIT

GRB2

SOS Ras

GDP C-Raf

N-Ras

GTP B-Raf

MEK

ERK

ELK

MITF

CDK24

Cyclin D

p16

Amplified

in 30

Amplified

in 30

50-65

V600E

mutation

15 mutation

Amplified or mutated in 20-40

acral and mucosal melanoma

25-50 loss

Frequent loss

Amplified in 10-15

Adapted from Sosman Curr Oncol Rep 11 405 (2009)

Vemurafenib and Dabrafenib show similar efficacy

Dabrafenib

Hauschild et al

Lancet 2012

Chapman et al

NEJM 2011

(n=337)

(n=187)

Vemurafenib DTIC

6

OS 97-136 mts Gain 39 mts HR 070

V600E (91) 133-100 HR = 075 V600K(9) 145ndash 76 HR = 043

PFS 16- 69 mts Gain 53 mts

HR 038

SUCCESS AND FAILURE

Molecular Alterations in Melanoma

BRAF + MEK Inhibitors

FGFR

PTEN

PI3K Akt

TOR

KIT

GRB2

SOS Ras

GDP C-Raf

N-Ras

GTP B-Raf

MEK

ERK

ELK

MITF

CDK24

Cyclin D

p16

Amplified

in 30

Amplified

in 30

50-65

V600E

mutation

15 mutation

Amplified or mutated in 20-40

acral and mucosal melanoma

25-50 loss

Frequent loss

Amplified in 10-15

Adapted from Sosman Curr Oncol Rep 11 405 (2009)

Median Follow-up D + T = 11 months and Vem = 10 months

asymp 8 weeks

A SOMEWHAT SOBERING END RESULT

bull Tumor by evolution is ldquomoving targetrdquo

bull Heterogeneity and Innate resistance

bull Acquired resistanceAdditional mutations

bull Mono-Dimensional Thinking

We need to address

bull Nodes of Convergence of Pathways

bull Cross Talk Complexity between pathways

Drug Development Challenges

eIF4F Node of Convergence

RasRaf ndash PI3K- Caspase cascade

FGFR

PTEN

PI3K Akt

TOR

KIT

GRB2

SOS Ras

GDP C-Raf

N-Ras

GTP B-Raf

MEK

ERK

ELK

MITF

CDK24

Cyclin D

p16

Amplified

in 30

Amplified

in 30

50-65

V600E

mutation

15 mutation

Amplified or mutated in 20-40

acral and mucosal melanoma

25-50 loss

Frequent loss

Amplified in 10-15

Adapted from Sosman Curr Oncol Rep 11 405 (2009)

Caspase cascade

Nexus eIF4F

12

bull Nodes of Convergence of Pathways

bull Cross Talk Complexity between pathways

Drug Development Challenges

81

520

0

10

20

30

40

50

60

70

80

90

Melanoma Colon cancer

N Engl J Med 2010 363809-19

Kopetz et al ASCO 2010

Differential response of BRAF inhibition in

BRAF mutant melanoma vs colon cancer

CROSS TALK and RESISTANCE

Prahallad et alhelliphellipBernards R Nature 2012

No drug

EGFR inhibitor

BRAF inhibitor

BRAF+EGFR

inhibitor

Human colon cancer growth in mice

16

bull Targeted Agents will be effective accross different tumor types with that target

ndash importance of organ of origin

ndashAnswer is NO

bull For combination therapies one must demonstrate the antitumor effects for each individual agent

ndashAnswer is NO

PUTS AN END TO TWO PARADIGMS

Molecular profiling

Identification of the molecular alteration

Targeted therapy according to the molecular profile

Tumor Specimen

Precision Medicine identify-hit the target

GUSTAVE ROUSSY PCM PROGRAM

Rational Genomics lsquorsquoMolecular Portraitsrsquorsquo for targeted therapy allocation Rapid through Clinical Trials Logistics ndash Logistics ndash Logistics Focus time pressure culture infrastructure Examples of Current Clinical Trials

SAFIR01 Molecular Screening in

Breast Cancer

Which candidate target FGFR1

FGFR2

FGF4 amp

NOTCH amp

Genetic

instability

PAK1 ampli

PIK3CA AKT PTEN

alteration

VEGFA

amplification

Target

discovery

Primary endpoint

of patients included

in phase III trial according to the

profile

Biopsy of metastatic sites

Frozen sample

CGHhot spot mutations

(PIK3CAAKT)

eligible for phase I

N= 400

Trial A

Trial F

Trial E

Trial D

Trial C

Trial B

Trials XYhellip

Funded by INCA Sanger 30 genes

Andreacute et al ESMO 2012 Lancet Oncol 2014

High level of expectation

Andreacute ESMO 2012

Inclusions

Recruitment completed between May 2011 and August 2012

Molecular alterations

Targetable alterations

Rare alterations

0

5

10

15

20

25

o

f p

atie

nts

wit

h a

vaila

ble

gen

om

ic r

esu

lt mutations

copy number alterations

Andreacute et al Lancet Oncology 2014

29 molecular alteration

IN THE END ONLY 12

gets targeted therapy

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

1200 PATIENTS IN 4 Years

MATCH-R trial

In vitro Cell lines Mouse avatar

Patients with + biomarker tumor exposed to a targeted therapy and an initial response

Resistant Sensitive

Tumor biopy or effusion

Biopsy metastatic site NGS

Array CGH

Chemotherapy 4 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

EGFR TKI if SCC

No PD metastatic NSCLC fisrt line chemotherapy

RANDOMIZED TRIAL SAFIR 02 LUNG

All histologies

Pemetrexed if Non-SCC Molecular alteration Excluding EGFR mut and ALK trl

Genetic abnormality Gene location Squamous Cell

Carcinoma Adenocarcinoma

Therapeutic

intrevention

PIK3CA amplification[ 3q263 33 6 AZD2014

(TORC12)

FGFR1 amplification[ 8p12 22 1 AZD4547

(FGFR)

PTEN mutation 10q233 10 2 AZD8186 (PI3Kbeta)

MET amplification 7q311 3-21 3-21 Volitinib

PTEN loss 10q233 8-20 8-20 AZD8186 (PI3Kbeta)

KRAS mutation[ 12p121 6 21

AZD6244

(MEKi)

Or combo with

AZD2014

LKB1 mutation 19p133 5 23 AZD2014

(TORC12)

HER 2 amplification 17q112-q12 17q21 3-5 5-9 AZD 8931

(pan-HER)

PIK3CA mutation 3q263 3 3 AZD2014

(TORC12)

RET translocation 10q112 2 1

AZD6474

(VEGFR EGFR RET)

BRAF mutation 7p34 2 1-3 AZD6244

(MEKi)

AKT1 mutation 14q3232 1 Very rare AZD5363

(Akt)

MET mutation 7q311 1 2 Volitinib

HER2 mutation 17q112-q12 17q21 1 2 AZD 8931

(pan-HER)

Get COMPLETE PIPELINES

Biopsy metastatic site NGS

CGH 51 alterations

Chemotherapy 6-8 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

Chemotherapy continuation

No PD metastatic Breast cancer patient 1st or

2nd line

RANDOMIZED TRIAL SAFIR 02 BREAST

Molecular alteration Excluding HER2

SAFIR02

BreastAndre

520600

Since 2010 Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics)

SAFIR01

(Andre)

427427

MOSCATO

(Soria)

11501200

SAFIR02

LungSoria

480600

MOST

preSAFIR

(Andre)

108108

SHIVA

(Letourneau

Pilot study 1st Gener Trials 2nd Gener

No NGS NGS WES Randomized trials Sponsor

Gustave Roussy monocentric

Gustave RoussyUnicancer multicenter

L BeacuterardLyon

Curie

Unified Database Pick-up

the winner targets

2nd generation Algorithm for Personnalized

medicine

WINTHER

150200

Profiler

MATCH-R

(WES PDX cfDNA)

200600

FUNDING TOTAL ~50 Million Fondation GR (10) MCM Building (15) IHU (6) INCA (6) ARC (4) Philanthropia (2) WINconsort (4) EU-FP7 (3)

MSN LungMel

BesseRobert

600600

Gustave RoussyWIN multicenter

MOSKIDO

(Goerger)

80100

GETUG

Fizazi

3580 ACSE

Vassal

600

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

12001200 PATIENTS IN 35 Years

bull ATTRITION RATE

bull 100 pts with molecular portrait

bull 50 pts have target identified

bull 25 are actionable

bull 25 overall response rate

bull So in the end 6100 patients benefithellip

bull INNATE RESISTANCE + ORGAN CONTEXT

PROBLEMS with

Molecular Portraits

NEEDS

bull Higher Target Identification Rate

bull Higher of Actionable Targets

bull Higher number diversification of new drugs

bull Rational intrainterpathway combinations

bull Functional Genetics (Crosstalk) ndash Rene Bernards

bull Nodes of convergence ndash Vagner Robert

bull Simplification of models ndash Master Regulators (Andrea Califano)

31

Problems with Targeted Drugs

bull Lack of Durability of responses

ndash Improvement with comborsquos limited

ndash Comborsquos of non-active drugs can be

active

bull Lack of Transversality of impact across

tumor types

ndash Organ of origin

ndash Context 32

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

INHIBIT THE INHIBITOR

vs ACTIVATE THE ACTIVATOR

BREAKING TOLERANCE Immune-Checkpoint-Blockers

REVOLUTION IN IMMUNOTHERAPY

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 4: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Molecular Alterations in Melanoma

FGFR

PTEN

PI3K Akt

TOR

KIT

GRB2

SOS Ras

GDP C-Raf

N-Ras

GTP B-Raf

MEK

ERK

ELK

MITF

CDK24

Cyclin D

p16

Amplified

in 30

Amplified

in 30

50-65

V600E

mutation

15 mutation

Amplified or mutated in 20-40

acral and mucosal melanoma

25-50 loss

Frequent loss

Amplified in 10-15

Adapted from Sosman Curr Oncol Rep 11 405 (2009)

Vemurafenib and Dabrafenib show similar efficacy

Dabrafenib

Hauschild et al

Lancet 2012

Chapman et al

NEJM 2011

(n=337)

(n=187)

Vemurafenib DTIC

6

OS 97-136 mts Gain 39 mts HR 070

V600E (91) 133-100 HR = 075 V600K(9) 145ndash 76 HR = 043

PFS 16- 69 mts Gain 53 mts

HR 038

SUCCESS AND FAILURE

Molecular Alterations in Melanoma

BRAF + MEK Inhibitors

FGFR

PTEN

PI3K Akt

TOR

KIT

GRB2

SOS Ras

GDP C-Raf

N-Ras

GTP B-Raf

MEK

ERK

ELK

MITF

CDK24

Cyclin D

p16

Amplified

in 30

Amplified

in 30

50-65

V600E

mutation

15 mutation

Amplified or mutated in 20-40

acral and mucosal melanoma

25-50 loss

Frequent loss

Amplified in 10-15

Adapted from Sosman Curr Oncol Rep 11 405 (2009)

Median Follow-up D + T = 11 months and Vem = 10 months

asymp 8 weeks

A SOMEWHAT SOBERING END RESULT

bull Tumor by evolution is ldquomoving targetrdquo

bull Heterogeneity and Innate resistance

bull Acquired resistanceAdditional mutations

bull Mono-Dimensional Thinking

We need to address

bull Nodes of Convergence of Pathways

bull Cross Talk Complexity between pathways

Drug Development Challenges

eIF4F Node of Convergence

RasRaf ndash PI3K- Caspase cascade

FGFR

PTEN

PI3K Akt

TOR

KIT

GRB2

SOS Ras

GDP C-Raf

N-Ras

GTP B-Raf

MEK

ERK

ELK

MITF

CDK24

Cyclin D

p16

Amplified

in 30

Amplified

in 30

50-65

V600E

mutation

15 mutation

Amplified or mutated in 20-40

acral and mucosal melanoma

25-50 loss

Frequent loss

Amplified in 10-15

Adapted from Sosman Curr Oncol Rep 11 405 (2009)

Caspase cascade

Nexus eIF4F

12

bull Nodes of Convergence of Pathways

bull Cross Talk Complexity between pathways

Drug Development Challenges

81

520

0

10

20

30

40

50

60

70

80

90

Melanoma Colon cancer

N Engl J Med 2010 363809-19

Kopetz et al ASCO 2010

Differential response of BRAF inhibition in

BRAF mutant melanoma vs colon cancer

CROSS TALK and RESISTANCE

Prahallad et alhelliphellipBernards R Nature 2012

No drug

EGFR inhibitor

BRAF inhibitor

BRAF+EGFR

inhibitor

Human colon cancer growth in mice

16

bull Targeted Agents will be effective accross different tumor types with that target

ndash importance of organ of origin

ndashAnswer is NO

bull For combination therapies one must demonstrate the antitumor effects for each individual agent

ndashAnswer is NO

PUTS AN END TO TWO PARADIGMS

Molecular profiling

Identification of the molecular alteration

Targeted therapy according to the molecular profile

Tumor Specimen

Precision Medicine identify-hit the target

GUSTAVE ROUSSY PCM PROGRAM

Rational Genomics lsquorsquoMolecular Portraitsrsquorsquo for targeted therapy allocation Rapid through Clinical Trials Logistics ndash Logistics ndash Logistics Focus time pressure culture infrastructure Examples of Current Clinical Trials

SAFIR01 Molecular Screening in

Breast Cancer

Which candidate target FGFR1

FGFR2

FGF4 amp

NOTCH amp

Genetic

instability

PAK1 ampli

PIK3CA AKT PTEN

alteration

VEGFA

amplification

Target

discovery

Primary endpoint

of patients included

in phase III trial according to the

profile

Biopsy of metastatic sites

Frozen sample

CGHhot spot mutations

(PIK3CAAKT)

eligible for phase I

N= 400

Trial A

Trial F

Trial E

Trial D

Trial C

Trial B

Trials XYhellip

Funded by INCA Sanger 30 genes

Andreacute et al ESMO 2012 Lancet Oncol 2014

High level of expectation

Andreacute ESMO 2012

Inclusions

Recruitment completed between May 2011 and August 2012

Molecular alterations

Targetable alterations

Rare alterations

0

5

10

15

20

25

o

f p

atie

nts

wit

h a

vaila

ble

gen

om

ic r

esu

lt mutations

copy number alterations

Andreacute et al Lancet Oncology 2014

29 molecular alteration

IN THE END ONLY 12

gets targeted therapy

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

1200 PATIENTS IN 4 Years

MATCH-R trial

In vitro Cell lines Mouse avatar

Patients with + biomarker tumor exposed to a targeted therapy and an initial response

Resistant Sensitive

Tumor biopy or effusion

Biopsy metastatic site NGS

Array CGH

Chemotherapy 4 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

EGFR TKI if SCC

No PD metastatic NSCLC fisrt line chemotherapy

RANDOMIZED TRIAL SAFIR 02 LUNG

All histologies

Pemetrexed if Non-SCC Molecular alteration Excluding EGFR mut and ALK trl

Genetic abnormality Gene location Squamous Cell

Carcinoma Adenocarcinoma

Therapeutic

intrevention

PIK3CA amplification[ 3q263 33 6 AZD2014

(TORC12)

FGFR1 amplification[ 8p12 22 1 AZD4547

(FGFR)

PTEN mutation 10q233 10 2 AZD8186 (PI3Kbeta)

MET amplification 7q311 3-21 3-21 Volitinib

PTEN loss 10q233 8-20 8-20 AZD8186 (PI3Kbeta)

KRAS mutation[ 12p121 6 21

AZD6244

(MEKi)

Or combo with

AZD2014

LKB1 mutation 19p133 5 23 AZD2014

(TORC12)

HER 2 amplification 17q112-q12 17q21 3-5 5-9 AZD 8931

(pan-HER)

PIK3CA mutation 3q263 3 3 AZD2014

(TORC12)

RET translocation 10q112 2 1

AZD6474

(VEGFR EGFR RET)

BRAF mutation 7p34 2 1-3 AZD6244

(MEKi)

AKT1 mutation 14q3232 1 Very rare AZD5363

(Akt)

MET mutation 7q311 1 2 Volitinib

HER2 mutation 17q112-q12 17q21 1 2 AZD 8931

(pan-HER)

Get COMPLETE PIPELINES

Biopsy metastatic site NGS

CGH 51 alterations

Chemotherapy 6-8 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

Chemotherapy continuation

No PD metastatic Breast cancer patient 1st or

2nd line

RANDOMIZED TRIAL SAFIR 02 BREAST

Molecular alteration Excluding HER2

SAFIR02

BreastAndre

520600

Since 2010 Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics)

SAFIR01

(Andre)

427427

MOSCATO

(Soria)

11501200

SAFIR02

LungSoria

480600

MOST

preSAFIR

(Andre)

108108

SHIVA

(Letourneau

Pilot study 1st Gener Trials 2nd Gener

No NGS NGS WES Randomized trials Sponsor

Gustave Roussy monocentric

Gustave RoussyUnicancer multicenter

L BeacuterardLyon

Curie

Unified Database Pick-up

the winner targets

2nd generation Algorithm for Personnalized

medicine

WINTHER

150200

Profiler

MATCH-R

(WES PDX cfDNA)

200600

FUNDING TOTAL ~50 Million Fondation GR (10) MCM Building (15) IHU (6) INCA (6) ARC (4) Philanthropia (2) WINconsort (4) EU-FP7 (3)

MSN LungMel

BesseRobert

600600

Gustave RoussyWIN multicenter

MOSKIDO

(Goerger)

80100

GETUG

Fizazi

3580 ACSE

Vassal

600

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

12001200 PATIENTS IN 35 Years

bull ATTRITION RATE

bull 100 pts with molecular portrait

bull 50 pts have target identified

bull 25 are actionable

bull 25 overall response rate

bull So in the end 6100 patients benefithellip

bull INNATE RESISTANCE + ORGAN CONTEXT

PROBLEMS with

Molecular Portraits

NEEDS

bull Higher Target Identification Rate

bull Higher of Actionable Targets

bull Higher number diversification of new drugs

bull Rational intrainterpathway combinations

bull Functional Genetics (Crosstalk) ndash Rene Bernards

bull Nodes of convergence ndash Vagner Robert

bull Simplification of models ndash Master Regulators (Andrea Califano)

31

Problems with Targeted Drugs

bull Lack of Durability of responses

ndash Improvement with comborsquos limited

ndash Comborsquos of non-active drugs can be

active

bull Lack of Transversality of impact across

tumor types

ndash Organ of origin

ndash Context 32

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

INHIBIT THE INHIBITOR

vs ACTIVATE THE ACTIVATOR

BREAKING TOLERANCE Immune-Checkpoint-Blockers

REVOLUTION IN IMMUNOTHERAPY

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 5: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Vemurafenib and Dabrafenib show similar efficacy

Dabrafenib

Hauschild et al

Lancet 2012

Chapman et al

NEJM 2011

(n=337)

(n=187)

Vemurafenib DTIC

6

OS 97-136 mts Gain 39 mts HR 070

V600E (91) 133-100 HR = 075 V600K(9) 145ndash 76 HR = 043

PFS 16- 69 mts Gain 53 mts

HR 038

SUCCESS AND FAILURE

Molecular Alterations in Melanoma

BRAF + MEK Inhibitors

FGFR

PTEN

PI3K Akt

TOR

KIT

GRB2

SOS Ras

GDP C-Raf

N-Ras

GTP B-Raf

MEK

ERK

ELK

MITF

CDK24

Cyclin D

p16

Amplified

in 30

Amplified

in 30

50-65

V600E

mutation

15 mutation

Amplified or mutated in 20-40

acral and mucosal melanoma

25-50 loss

Frequent loss

Amplified in 10-15

Adapted from Sosman Curr Oncol Rep 11 405 (2009)

Median Follow-up D + T = 11 months and Vem = 10 months

asymp 8 weeks

A SOMEWHAT SOBERING END RESULT

bull Tumor by evolution is ldquomoving targetrdquo

bull Heterogeneity and Innate resistance

bull Acquired resistanceAdditional mutations

bull Mono-Dimensional Thinking

We need to address

bull Nodes of Convergence of Pathways

bull Cross Talk Complexity between pathways

Drug Development Challenges

eIF4F Node of Convergence

RasRaf ndash PI3K- Caspase cascade

FGFR

PTEN

PI3K Akt

TOR

KIT

GRB2

SOS Ras

GDP C-Raf

N-Ras

GTP B-Raf

MEK

ERK

ELK

MITF

CDK24

Cyclin D

p16

Amplified

in 30

Amplified

in 30

50-65

V600E

mutation

15 mutation

Amplified or mutated in 20-40

acral and mucosal melanoma

25-50 loss

Frequent loss

Amplified in 10-15

Adapted from Sosman Curr Oncol Rep 11 405 (2009)

Caspase cascade

Nexus eIF4F

12

bull Nodes of Convergence of Pathways

bull Cross Talk Complexity between pathways

Drug Development Challenges

81

520

0

10

20

30

40

50

60

70

80

90

Melanoma Colon cancer

N Engl J Med 2010 363809-19

Kopetz et al ASCO 2010

Differential response of BRAF inhibition in

BRAF mutant melanoma vs colon cancer

CROSS TALK and RESISTANCE

Prahallad et alhelliphellipBernards R Nature 2012

No drug

EGFR inhibitor

BRAF inhibitor

BRAF+EGFR

inhibitor

Human colon cancer growth in mice

16

bull Targeted Agents will be effective accross different tumor types with that target

ndash importance of organ of origin

ndashAnswer is NO

bull For combination therapies one must demonstrate the antitumor effects for each individual agent

ndashAnswer is NO

PUTS AN END TO TWO PARADIGMS

Molecular profiling

Identification of the molecular alteration

Targeted therapy according to the molecular profile

Tumor Specimen

Precision Medicine identify-hit the target

GUSTAVE ROUSSY PCM PROGRAM

Rational Genomics lsquorsquoMolecular Portraitsrsquorsquo for targeted therapy allocation Rapid through Clinical Trials Logistics ndash Logistics ndash Logistics Focus time pressure culture infrastructure Examples of Current Clinical Trials

SAFIR01 Molecular Screening in

Breast Cancer

Which candidate target FGFR1

FGFR2

FGF4 amp

NOTCH amp

Genetic

instability

PAK1 ampli

PIK3CA AKT PTEN

alteration

VEGFA

amplification

Target

discovery

Primary endpoint

of patients included

in phase III trial according to the

profile

Biopsy of metastatic sites

Frozen sample

CGHhot spot mutations

(PIK3CAAKT)

eligible for phase I

N= 400

Trial A

Trial F

Trial E

Trial D

Trial C

Trial B

Trials XYhellip

Funded by INCA Sanger 30 genes

Andreacute et al ESMO 2012 Lancet Oncol 2014

High level of expectation

Andreacute ESMO 2012

Inclusions

Recruitment completed between May 2011 and August 2012

Molecular alterations

Targetable alterations

Rare alterations

0

5

10

15

20

25

o

f p

atie

nts

wit

h a

vaila

ble

gen

om

ic r

esu

lt mutations

copy number alterations

Andreacute et al Lancet Oncology 2014

29 molecular alteration

IN THE END ONLY 12

gets targeted therapy

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

1200 PATIENTS IN 4 Years

MATCH-R trial

In vitro Cell lines Mouse avatar

Patients with + biomarker tumor exposed to a targeted therapy and an initial response

Resistant Sensitive

Tumor biopy or effusion

Biopsy metastatic site NGS

Array CGH

Chemotherapy 4 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

EGFR TKI if SCC

No PD metastatic NSCLC fisrt line chemotherapy

RANDOMIZED TRIAL SAFIR 02 LUNG

All histologies

Pemetrexed if Non-SCC Molecular alteration Excluding EGFR mut and ALK trl

Genetic abnormality Gene location Squamous Cell

Carcinoma Adenocarcinoma

Therapeutic

intrevention

PIK3CA amplification[ 3q263 33 6 AZD2014

(TORC12)

FGFR1 amplification[ 8p12 22 1 AZD4547

(FGFR)

PTEN mutation 10q233 10 2 AZD8186 (PI3Kbeta)

MET amplification 7q311 3-21 3-21 Volitinib

PTEN loss 10q233 8-20 8-20 AZD8186 (PI3Kbeta)

KRAS mutation[ 12p121 6 21

AZD6244

(MEKi)

Or combo with

AZD2014

LKB1 mutation 19p133 5 23 AZD2014

(TORC12)

HER 2 amplification 17q112-q12 17q21 3-5 5-9 AZD 8931

(pan-HER)

PIK3CA mutation 3q263 3 3 AZD2014

(TORC12)

RET translocation 10q112 2 1

AZD6474

(VEGFR EGFR RET)

BRAF mutation 7p34 2 1-3 AZD6244

(MEKi)

AKT1 mutation 14q3232 1 Very rare AZD5363

(Akt)

MET mutation 7q311 1 2 Volitinib

HER2 mutation 17q112-q12 17q21 1 2 AZD 8931

(pan-HER)

Get COMPLETE PIPELINES

Biopsy metastatic site NGS

CGH 51 alterations

Chemotherapy 6-8 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

Chemotherapy continuation

No PD metastatic Breast cancer patient 1st or

2nd line

RANDOMIZED TRIAL SAFIR 02 BREAST

Molecular alteration Excluding HER2

SAFIR02

BreastAndre

520600

Since 2010 Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics)

SAFIR01

(Andre)

427427

MOSCATO

(Soria)

11501200

SAFIR02

LungSoria

480600

MOST

preSAFIR

(Andre)

108108

SHIVA

(Letourneau

Pilot study 1st Gener Trials 2nd Gener

No NGS NGS WES Randomized trials Sponsor

Gustave Roussy monocentric

Gustave RoussyUnicancer multicenter

L BeacuterardLyon

Curie

Unified Database Pick-up

the winner targets

2nd generation Algorithm for Personnalized

medicine

WINTHER

150200

Profiler

MATCH-R

(WES PDX cfDNA)

200600

FUNDING TOTAL ~50 Million Fondation GR (10) MCM Building (15) IHU (6) INCA (6) ARC (4) Philanthropia (2) WINconsort (4) EU-FP7 (3)

MSN LungMel

BesseRobert

600600

Gustave RoussyWIN multicenter

MOSKIDO

(Goerger)

80100

GETUG

Fizazi

3580 ACSE

Vassal

600

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

12001200 PATIENTS IN 35 Years

bull ATTRITION RATE

bull 100 pts with molecular portrait

bull 50 pts have target identified

bull 25 are actionable

bull 25 overall response rate

bull So in the end 6100 patients benefithellip

bull INNATE RESISTANCE + ORGAN CONTEXT

PROBLEMS with

Molecular Portraits

NEEDS

bull Higher Target Identification Rate

bull Higher of Actionable Targets

bull Higher number diversification of new drugs

bull Rational intrainterpathway combinations

bull Functional Genetics (Crosstalk) ndash Rene Bernards

bull Nodes of convergence ndash Vagner Robert

bull Simplification of models ndash Master Regulators (Andrea Califano)

31

Problems with Targeted Drugs

bull Lack of Durability of responses

ndash Improvement with comborsquos limited

ndash Comborsquos of non-active drugs can be

active

bull Lack of Transversality of impact across

tumor types

ndash Organ of origin

ndash Context 32

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

INHIBIT THE INHIBITOR

vs ACTIVATE THE ACTIVATOR

BREAKING TOLERANCE Immune-Checkpoint-Blockers

REVOLUTION IN IMMUNOTHERAPY

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 6: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

6

OS 97-136 mts Gain 39 mts HR 070

V600E (91) 133-100 HR = 075 V600K(9) 145ndash 76 HR = 043

PFS 16- 69 mts Gain 53 mts

HR 038

SUCCESS AND FAILURE

Molecular Alterations in Melanoma

BRAF + MEK Inhibitors

FGFR

PTEN

PI3K Akt

TOR

KIT

GRB2

SOS Ras

GDP C-Raf

N-Ras

GTP B-Raf

MEK

ERK

ELK

MITF

CDK24

Cyclin D

p16

Amplified

in 30

Amplified

in 30

50-65

V600E

mutation

15 mutation

Amplified or mutated in 20-40

acral and mucosal melanoma

25-50 loss

Frequent loss

Amplified in 10-15

Adapted from Sosman Curr Oncol Rep 11 405 (2009)

Median Follow-up D + T = 11 months and Vem = 10 months

asymp 8 weeks

A SOMEWHAT SOBERING END RESULT

bull Tumor by evolution is ldquomoving targetrdquo

bull Heterogeneity and Innate resistance

bull Acquired resistanceAdditional mutations

bull Mono-Dimensional Thinking

We need to address

bull Nodes of Convergence of Pathways

bull Cross Talk Complexity between pathways

Drug Development Challenges

eIF4F Node of Convergence

RasRaf ndash PI3K- Caspase cascade

FGFR

PTEN

PI3K Akt

TOR

KIT

GRB2

SOS Ras

GDP C-Raf

N-Ras

GTP B-Raf

MEK

ERK

ELK

MITF

CDK24

Cyclin D

p16

Amplified

in 30

Amplified

in 30

50-65

V600E

mutation

15 mutation

Amplified or mutated in 20-40

acral and mucosal melanoma

25-50 loss

Frequent loss

Amplified in 10-15

Adapted from Sosman Curr Oncol Rep 11 405 (2009)

Caspase cascade

Nexus eIF4F

12

bull Nodes of Convergence of Pathways

bull Cross Talk Complexity between pathways

Drug Development Challenges

81

520

0

10

20

30

40

50

60

70

80

90

Melanoma Colon cancer

N Engl J Med 2010 363809-19

Kopetz et al ASCO 2010

Differential response of BRAF inhibition in

BRAF mutant melanoma vs colon cancer

CROSS TALK and RESISTANCE

Prahallad et alhelliphellipBernards R Nature 2012

No drug

EGFR inhibitor

BRAF inhibitor

BRAF+EGFR

inhibitor

Human colon cancer growth in mice

16

bull Targeted Agents will be effective accross different tumor types with that target

ndash importance of organ of origin

ndashAnswer is NO

bull For combination therapies one must demonstrate the antitumor effects for each individual agent

ndashAnswer is NO

PUTS AN END TO TWO PARADIGMS

Molecular profiling

Identification of the molecular alteration

Targeted therapy according to the molecular profile

Tumor Specimen

Precision Medicine identify-hit the target

GUSTAVE ROUSSY PCM PROGRAM

Rational Genomics lsquorsquoMolecular Portraitsrsquorsquo for targeted therapy allocation Rapid through Clinical Trials Logistics ndash Logistics ndash Logistics Focus time pressure culture infrastructure Examples of Current Clinical Trials

SAFIR01 Molecular Screening in

Breast Cancer

Which candidate target FGFR1

FGFR2

FGF4 amp

NOTCH amp

Genetic

instability

PAK1 ampli

PIK3CA AKT PTEN

alteration

VEGFA

amplification

Target

discovery

Primary endpoint

of patients included

in phase III trial according to the

profile

Biopsy of metastatic sites

Frozen sample

CGHhot spot mutations

(PIK3CAAKT)

eligible for phase I

N= 400

Trial A

Trial F

Trial E

Trial D

Trial C

Trial B

Trials XYhellip

Funded by INCA Sanger 30 genes

Andreacute et al ESMO 2012 Lancet Oncol 2014

High level of expectation

Andreacute ESMO 2012

Inclusions

Recruitment completed between May 2011 and August 2012

Molecular alterations

Targetable alterations

Rare alterations

0

5

10

15

20

25

o

f p

atie

nts

wit

h a

vaila

ble

gen

om

ic r

esu

lt mutations

copy number alterations

Andreacute et al Lancet Oncology 2014

29 molecular alteration

IN THE END ONLY 12

gets targeted therapy

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

1200 PATIENTS IN 4 Years

MATCH-R trial

In vitro Cell lines Mouse avatar

Patients with + biomarker tumor exposed to a targeted therapy and an initial response

Resistant Sensitive

Tumor biopy or effusion

Biopsy metastatic site NGS

Array CGH

Chemotherapy 4 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

EGFR TKI if SCC

No PD metastatic NSCLC fisrt line chemotherapy

RANDOMIZED TRIAL SAFIR 02 LUNG

All histologies

Pemetrexed if Non-SCC Molecular alteration Excluding EGFR mut and ALK trl

Genetic abnormality Gene location Squamous Cell

Carcinoma Adenocarcinoma

Therapeutic

intrevention

PIK3CA amplification[ 3q263 33 6 AZD2014

(TORC12)

FGFR1 amplification[ 8p12 22 1 AZD4547

(FGFR)

PTEN mutation 10q233 10 2 AZD8186 (PI3Kbeta)

MET amplification 7q311 3-21 3-21 Volitinib

PTEN loss 10q233 8-20 8-20 AZD8186 (PI3Kbeta)

KRAS mutation[ 12p121 6 21

AZD6244

(MEKi)

Or combo with

AZD2014

LKB1 mutation 19p133 5 23 AZD2014

(TORC12)

HER 2 amplification 17q112-q12 17q21 3-5 5-9 AZD 8931

(pan-HER)

PIK3CA mutation 3q263 3 3 AZD2014

(TORC12)

RET translocation 10q112 2 1

AZD6474

(VEGFR EGFR RET)

BRAF mutation 7p34 2 1-3 AZD6244

(MEKi)

AKT1 mutation 14q3232 1 Very rare AZD5363

(Akt)

MET mutation 7q311 1 2 Volitinib

HER2 mutation 17q112-q12 17q21 1 2 AZD 8931

(pan-HER)

Get COMPLETE PIPELINES

Biopsy metastatic site NGS

CGH 51 alterations

Chemotherapy 6-8 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

Chemotherapy continuation

No PD metastatic Breast cancer patient 1st or

2nd line

RANDOMIZED TRIAL SAFIR 02 BREAST

Molecular alteration Excluding HER2

SAFIR02

BreastAndre

520600

Since 2010 Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics)

SAFIR01

(Andre)

427427

MOSCATO

(Soria)

11501200

SAFIR02

LungSoria

480600

MOST

preSAFIR

(Andre)

108108

SHIVA

(Letourneau

Pilot study 1st Gener Trials 2nd Gener

No NGS NGS WES Randomized trials Sponsor

Gustave Roussy monocentric

Gustave RoussyUnicancer multicenter

L BeacuterardLyon

Curie

Unified Database Pick-up

the winner targets

2nd generation Algorithm for Personnalized

medicine

WINTHER

150200

Profiler

MATCH-R

(WES PDX cfDNA)

200600

FUNDING TOTAL ~50 Million Fondation GR (10) MCM Building (15) IHU (6) INCA (6) ARC (4) Philanthropia (2) WINconsort (4) EU-FP7 (3)

MSN LungMel

BesseRobert

600600

Gustave RoussyWIN multicenter

MOSKIDO

(Goerger)

80100

GETUG

Fizazi

3580 ACSE

Vassal

600

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

12001200 PATIENTS IN 35 Years

bull ATTRITION RATE

bull 100 pts with molecular portrait

bull 50 pts have target identified

bull 25 are actionable

bull 25 overall response rate

bull So in the end 6100 patients benefithellip

bull INNATE RESISTANCE + ORGAN CONTEXT

PROBLEMS with

Molecular Portraits

NEEDS

bull Higher Target Identification Rate

bull Higher of Actionable Targets

bull Higher number diversification of new drugs

bull Rational intrainterpathway combinations

bull Functional Genetics (Crosstalk) ndash Rene Bernards

bull Nodes of convergence ndash Vagner Robert

bull Simplification of models ndash Master Regulators (Andrea Califano)

31

Problems with Targeted Drugs

bull Lack of Durability of responses

ndash Improvement with comborsquos limited

ndash Comborsquos of non-active drugs can be

active

bull Lack of Transversality of impact across

tumor types

ndash Organ of origin

ndash Context 32

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

INHIBIT THE INHIBITOR

vs ACTIVATE THE ACTIVATOR

BREAKING TOLERANCE Immune-Checkpoint-Blockers

REVOLUTION IN IMMUNOTHERAPY

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 7: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

SUCCESS AND FAILURE

Molecular Alterations in Melanoma

BRAF + MEK Inhibitors

FGFR

PTEN

PI3K Akt

TOR

KIT

GRB2

SOS Ras

GDP C-Raf

N-Ras

GTP B-Raf

MEK

ERK

ELK

MITF

CDK24

Cyclin D

p16

Amplified

in 30

Amplified

in 30

50-65

V600E

mutation

15 mutation

Amplified or mutated in 20-40

acral and mucosal melanoma

25-50 loss

Frequent loss

Amplified in 10-15

Adapted from Sosman Curr Oncol Rep 11 405 (2009)

Median Follow-up D + T = 11 months and Vem = 10 months

asymp 8 weeks

A SOMEWHAT SOBERING END RESULT

bull Tumor by evolution is ldquomoving targetrdquo

bull Heterogeneity and Innate resistance

bull Acquired resistanceAdditional mutations

bull Mono-Dimensional Thinking

We need to address

bull Nodes of Convergence of Pathways

bull Cross Talk Complexity between pathways

Drug Development Challenges

eIF4F Node of Convergence

RasRaf ndash PI3K- Caspase cascade

FGFR

PTEN

PI3K Akt

TOR

KIT

GRB2

SOS Ras

GDP C-Raf

N-Ras

GTP B-Raf

MEK

ERK

ELK

MITF

CDK24

Cyclin D

p16

Amplified

in 30

Amplified

in 30

50-65

V600E

mutation

15 mutation

Amplified or mutated in 20-40

acral and mucosal melanoma

25-50 loss

Frequent loss

Amplified in 10-15

Adapted from Sosman Curr Oncol Rep 11 405 (2009)

Caspase cascade

Nexus eIF4F

12

bull Nodes of Convergence of Pathways

bull Cross Talk Complexity between pathways

Drug Development Challenges

81

520

0

10

20

30

40

50

60

70

80

90

Melanoma Colon cancer

N Engl J Med 2010 363809-19

Kopetz et al ASCO 2010

Differential response of BRAF inhibition in

BRAF mutant melanoma vs colon cancer

CROSS TALK and RESISTANCE

Prahallad et alhelliphellipBernards R Nature 2012

No drug

EGFR inhibitor

BRAF inhibitor

BRAF+EGFR

inhibitor

Human colon cancer growth in mice

16

bull Targeted Agents will be effective accross different tumor types with that target

ndash importance of organ of origin

ndashAnswer is NO

bull For combination therapies one must demonstrate the antitumor effects for each individual agent

ndashAnswer is NO

PUTS AN END TO TWO PARADIGMS

Molecular profiling

Identification of the molecular alteration

Targeted therapy according to the molecular profile

Tumor Specimen

Precision Medicine identify-hit the target

GUSTAVE ROUSSY PCM PROGRAM

Rational Genomics lsquorsquoMolecular Portraitsrsquorsquo for targeted therapy allocation Rapid through Clinical Trials Logistics ndash Logistics ndash Logistics Focus time pressure culture infrastructure Examples of Current Clinical Trials

SAFIR01 Molecular Screening in

Breast Cancer

Which candidate target FGFR1

FGFR2

FGF4 amp

NOTCH amp

Genetic

instability

PAK1 ampli

PIK3CA AKT PTEN

alteration

VEGFA

amplification

Target

discovery

Primary endpoint

of patients included

in phase III trial according to the

profile

Biopsy of metastatic sites

Frozen sample

CGHhot spot mutations

(PIK3CAAKT)

eligible for phase I

N= 400

Trial A

Trial F

Trial E

Trial D

Trial C

Trial B

Trials XYhellip

Funded by INCA Sanger 30 genes

Andreacute et al ESMO 2012 Lancet Oncol 2014

High level of expectation

Andreacute ESMO 2012

Inclusions

Recruitment completed between May 2011 and August 2012

Molecular alterations

Targetable alterations

Rare alterations

0

5

10

15

20

25

o

f p

atie

nts

wit

h a

vaila

ble

gen

om

ic r

esu

lt mutations

copy number alterations

Andreacute et al Lancet Oncology 2014

29 molecular alteration

IN THE END ONLY 12

gets targeted therapy

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

1200 PATIENTS IN 4 Years

MATCH-R trial

In vitro Cell lines Mouse avatar

Patients with + biomarker tumor exposed to a targeted therapy and an initial response

Resistant Sensitive

Tumor biopy or effusion

Biopsy metastatic site NGS

Array CGH

Chemotherapy 4 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

EGFR TKI if SCC

No PD metastatic NSCLC fisrt line chemotherapy

RANDOMIZED TRIAL SAFIR 02 LUNG

All histologies

Pemetrexed if Non-SCC Molecular alteration Excluding EGFR mut and ALK trl

Genetic abnormality Gene location Squamous Cell

Carcinoma Adenocarcinoma

Therapeutic

intrevention

PIK3CA amplification[ 3q263 33 6 AZD2014

(TORC12)

FGFR1 amplification[ 8p12 22 1 AZD4547

(FGFR)

PTEN mutation 10q233 10 2 AZD8186 (PI3Kbeta)

MET amplification 7q311 3-21 3-21 Volitinib

PTEN loss 10q233 8-20 8-20 AZD8186 (PI3Kbeta)

KRAS mutation[ 12p121 6 21

AZD6244

(MEKi)

Or combo with

AZD2014

LKB1 mutation 19p133 5 23 AZD2014

(TORC12)

HER 2 amplification 17q112-q12 17q21 3-5 5-9 AZD 8931

(pan-HER)

PIK3CA mutation 3q263 3 3 AZD2014

(TORC12)

RET translocation 10q112 2 1

AZD6474

(VEGFR EGFR RET)

BRAF mutation 7p34 2 1-3 AZD6244

(MEKi)

AKT1 mutation 14q3232 1 Very rare AZD5363

(Akt)

MET mutation 7q311 1 2 Volitinib

HER2 mutation 17q112-q12 17q21 1 2 AZD 8931

(pan-HER)

Get COMPLETE PIPELINES

Biopsy metastatic site NGS

CGH 51 alterations

Chemotherapy 6-8 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

Chemotherapy continuation

No PD metastatic Breast cancer patient 1st or

2nd line

RANDOMIZED TRIAL SAFIR 02 BREAST

Molecular alteration Excluding HER2

SAFIR02

BreastAndre

520600

Since 2010 Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics)

SAFIR01

(Andre)

427427

MOSCATO

(Soria)

11501200

SAFIR02

LungSoria

480600

MOST

preSAFIR

(Andre)

108108

SHIVA

(Letourneau

Pilot study 1st Gener Trials 2nd Gener

No NGS NGS WES Randomized trials Sponsor

Gustave Roussy monocentric

Gustave RoussyUnicancer multicenter

L BeacuterardLyon

Curie

Unified Database Pick-up

the winner targets

2nd generation Algorithm for Personnalized

medicine

WINTHER

150200

Profiler

MATCH-R

(WES PDX cfDNA)

200600

FUNDING TOTAL ~50 Million Fondation GR (10) MCM Building (15) IHU (6) INCA (6) ARC (4) Philanthropia (2) WINconsort (4) EU-FP7 (3)

MSN LungMel

BesseRobert

600600

Gustave RoussyWIN multicenter

MOSKIDO

(Goerger)

80100

GETUG

Fizazi

3580 ACSE

Vassal

600

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

12001200 PATIENTS IN 35 Years

bull ATTRITION RATE

bull 100 pts with molecular portrait

bull 50 pts have target identified

bull 25 are actionable

bull 25 overall response rate

bull So in the end 6100 patients benefithellip

bull INNATE RESISTANCE + ORGAN CONTEXT

PROBLEMS with

Molecular Portraits

NEEDS

bull Higher Target Identification Rate

bull Higher of Actionable Targets

bull Higher number diversification of new drugs

bull Rational intrainterpathway combinations

bull Functional Genetics (Crosstalk) ndash Rene Bernards

bull Nodes of convergence ndash Vagner Robert

bull Simplification of models ndash Master Regulators (Andrea Califano)

31

Problems with Targeted Drugs

bull Lack of Durability of responses

ndash Improvement with comborsquos limited

ndash Comborsquos of non-active drugs can be

active

bull Lack of Transversality of impact across

tumor types

ndash Organ of origin

ndash Context 32

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

INHIBIT THE INHIBITOR

vs ACTIVATE THE ACTIVATOR

BREAKING TOLERANCE Immune-Checkpoint-Blockers

REVOLUTION IN IMMUNOTHERAPY

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 8: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Molecular Alterations in Melanoma

BRAF + MEK Inhibitors

FGFR

PTEN

PI3K Akt

TOR

KIT

GRB2

SOS Ras

GDP C-Raf

N-Ras

GTP B-Raf

MEK

ERK

ELK

MITF

CDK24

Cyclin D

p16

Amplified

in 30

Amplified

in 30

50-65

V600E

mutation

15 mutation

Amplified or mutated in 20-40

acral and mucosal melanoma

25-50 loss

Frequent loss

Amplified in 10-15

Adapted from Sosman Curr Oncol Rep 11 405 (2009)

Median Follow-up D + T = 11 months and Vem = 10 months

asymp 8 weeks

A SOMEWHAT SOBERING END RESULT

bull Tumor by evolution is ldquomoving targetrdquo

bull Heterogeneity and Innate resistance

bull Acquired resistanceAdditional mutations

bull Mono-Dimensional Thinking

We need to address

bull Nodes of Convergence of Pathways

bull Cross Talk Complexity between pathways

Drug Development Challenges

eIF4F Node of Convergence

RasRaf ndash PI3K- Caspase cascade

FGFR

PTEN

PI3K Akt

TOR

KIT

GRB2

SOS Ras

GDP C-Raf

N-Ras

GTP B-Raf

MEK

ERK

ELK

MITF

CDK24

Cyclin D

p16

Amplified

in 30

Amplified

in 30

50-65

V600E

mutation

15 mutation

Amplified or mutated in 20-40

acral and mucosal melanoma

25-50 loss

Frequent loss

Amplified in 10-15

Adapted from Sosman Curr Oncol Rep 11 405 (2009)

Caspase cascade

Nexus eIF4F

12

bull Nodes of Convergence of Pathways

bull Cross Talk Complexity between pathways

Drug Development Challenges

81

520

0

10

20

30

40

50

60

70

80

90

Melanoma Colon cancer

N Engl J Med 2010 363809-19

Kopetz et al ASCO 2010

Differential response of BRAF inhibition in

BRAF mutant melanoma vs colon cancer

CROSS TALK and RESISTANCE

Prahallad et alhelliphellipBernards R Nature 2012

No drug

EGFR inhibitor

BRAF inhibitor

BRAF+EGFR

inhibitor

Human colon cancer growth in mice

16

bull Targeted Agents will be effective accross different tumor types with that target

ndash importance of organ of origin

ndashAnswer is NO

bull For combination therapies one must demonstrate the antitumor effects for each individual agent

ndashAnswer is NO

PUTS AN END TO TWO PARADIGMS

Molecular profiling

Identification of the molecular alteration

Targeted therapy according to the molecular profile

Tumor Specimen

Precision Medicine identify-hit the target

GUSTAVE ROUSSY PCM PROGRAM

Rational Genomics lsquorsquoMolecular Portraitsrsquorsquo for targeted therapy allocation Rapid through Clinical Trials Logistics ndash Logistics ndash Logistics Focus time pressure culture infrastructure Examples of Current Clinical Trials

SAFIR01 Molecular Screening in

Breast Cancer

Which candidate target FGFR1

FGFR2

FGF4 amp

NOTCH amp

Genetic

instability

PAK1 ampli

PIK3CA AKT PTEN

alteration

VEGFA

amplification

Target

discovery

Primary endpoint

of patients included

in phase III trial according to the

profile

Biopsy of metastatic sites

Frozen sample

CGHhot spot mutations

(PIK3CAAKT)

eligible for phase I

N= 400

Trial A

Trial F

Trial E

Trial D

Trial C

Trial B

Trials XYhellip

Funded by INCA Sanger 30 genes

Andreacute et al ESMO 2012 Lancet Oncol 2014

High level of expectation

Andreacute ESMO 2012

Inclusions

Recruitment completed between May 2011 and August 2012

Molecular alterations

Targetable alterations

Rare alterations

0

5

10

15

20

25

o

f p

atie

nts

wit

h a

vaila

ble

gen

om

ic r

esu

lt mutations

copy number alterations

Andreacute et al Lancet Oncology 2014

29 molecular alteration

IN THE END ONLY 12

gets targeted therapy

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

1200 PATIENTS IN 4 Years

MATCH-R trial

In vitro Cell lines Mouse avatar

Patients with + biomarker tumor exposed to a targeted therapy and an initial response

Resistant Sensitive

Tumor biopy or effusion

Biopsy metastatic site NGS

Array CGH

Chemotherapy 4 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

EGFR TKI if SCC

No PD metastatic NSCLC fisrt line chemotherapy

RANDOMIZED TRIAL SAFIR 02 LUNG

All histologies

Pemetrexed if Non-SCC Molecular alteration Excluding EGFR mut and ALK trl

Genetic abnormality Gene location Squamous Cell

Carcinoma Adenocarcinoma

Therapeutic

intrevention

PIK3CA amplification[ 3q263 33 6 AZD2014

(TORC12)

FGFR1 amplification[ 8p12 22 1 AZD4547

(FGFR)

PTEN mutation 10q233 10 2 AZD8186 (PI3Kbeta)

MET amplification 7q311 3-21 3-21 Volitinib

PTEN loss 10q233 8-20 8-20 AZD8186 (PI3Kbeta)

KRAS mutation[ 12p121 6 21

AZD6244

(MEKi)

Or combo with

AZD2014

LKB1 mutation 19p133 5 23 AZD2014

(TORC12)

HER 2 amplification 17q112-q12 17q21 3-5 5-9 AZD 8931

(pan-HER)

PIK3CA mutation 3q263 3 3 AZD2014

(TORC12)

RET translocation 10q112 2 1

AZD6474

(VEGFR EGFR RET)

BRAF mutation 7p34 2 1-3 AZD6244

(MEKi)

AKT1 mutation 14q3232 1 Very rare AZD5363

(Akt)

MET mutation 7q311 1 2 Volitinib

HER2 mutation 17q112-q12 17q21 1 2 AZD 8931

(pan-HER)

Get COMPLETE PIPELINES

Biopsy metastatic site NGS

CGH 51 alterations

Chemotherapy 6-8 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

Chemotherapy continuation

No PD metastatic Breast cancer patient 1st or

2nd line

RANDOMIZED TRIAL SAFIR 02 BREAST

Molecular alteration Excluding HER2

SAFIR02

BreastAndre

520600

Since 2010 Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics)

SAFIR01

(Andre)

427427

MOSCATO

(Soria)

11501200

SAFIR02

LungSoria

480600

MOST

preSAFIR

(Andre)

108108

SHIVA

(Letourneau

Pilot study 1st Gener Trials 2nd Gener

No NGS NGS WES Randomized trials Sponsor

Gustave Roussy monocentric

Gustave RoussyUnicancer multicenter

L BeacuterardLyon

Curie

Unified Database Pick-up

the winner targets

2nd generation Algorithm for Personnalized

medicine

WINTHER

150200

Profiler

MATCH-R

(WES PDX cfDNA)

200600

FUNDING TOTAL ~50 Million Fondation GR (10) MCM Building (15) IHU (6) INCA (6) ARC (4) Philanthropia (2) WINconsort (4) EU-FP7 (3)

MSN LungMel

BesseRobert

600600

Gustave RoussyWIN multicenter

MOSKIDO

(Goerger)

80100

GETUG

Fizazi

3580 ACSE

Vassal

600

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

12001200 PATIENTS IN 35 Years

bull ATTRITION RATE

bull 100 pts with molecular portrait

bull 50 pts have target identified

bull 25 are actionable

bull 25 overall response rate

bull So in the end 6100 patients benefithellip

bull INNATE RESISTANCE + ORGAN CONTEXT

PROBLEMS with

Molecular Portraits

NEEDS

bull Higher Target Identification Rate

bull Higher of Actionable Targets

bull Higher number diversification of new drugs

bull Rational intrainterpathway combinations

bull Functional Genetics (Crosstalk) ndash Rene Bernards

bull Nodes of convergence ndash Vagner Robert

bull Simplification of models ndash Master Regulators (Andrea Califano)

31

Problems with Targeted Drugs

bull Lack of Durability of responses

ndash Improvement with comborsquos limited

ndash Comborsquos of non-active drugs can be

active

bull Lack of Transversality of impact across

tumor types

ndash Organ of origin

ndash Context 32

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

INHIBIT THE INHIBITOR

vs ACTIVATE THE ACTIVATOR

BREAKING TOLERANCE Immune-Checkpoint-Blockers

REVOLUTION IN IMMUNOTHERAPY

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 9: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Median Follow-up D + T = 11 months and Vem = 10 months

asymp 8 weeks

A SOMEWHAT SOBERING END RESULT

bull Tumor by evolution is ldquomoving targetrdquo

bull Heterogeneity and Innate resistance

bull Acquired resistanceAdditional mutations

bull Mono-Dimensional Thinking

We need to address

bull Nodes of Convergence of Pathways

bull Cross Talk Complexity between pathways

Drug Development Challenges

eIF4F Node of Convergence

RasRaf ndash PI3K- Caspase cascade

FGFR

PTEN

PI3K Akt

TOR

KIT

GRB2

SOS Ras

GDP C-Raf

N-Ras

GTP B-Raf

MEK

ERK

ELK

MITF

CDK24

Cyclin D

p16

Amplified

in 30

Amplified

in 30

50-65

V600E

mutation

15 mutation

Amplified or mutated in 20-40

acral and mucosal melanoma

25-50 loss

Frequent loss

Amplified in 10-15

Adapted from Sosman Curr Oncol Rep 11 405 (2009)

Caspase cascade

Nexus eIF4F

12

bull Nodes of Convergence of Pathways

bull Cross Talk Complexity between pathways

Drug Development Challenges

81

520

0

10

20

30

40

50

60

70

80

90

Melanoma Colon cancer

N Engl J Med 2010 363809-19

Kopetz et al ASCO 2010

Differential response of BRAF inhibition in

BRAF mutant melanoma vs colon cancer

CROSS TALK and RESISTANCE

Prahallad et alhelliphellipBernards R Nature 2012

No drug

EGFR inhibitor

BRAF inhibitor

BRAF+EGFR

inhibitor

Human colon cancer growth in mice

16

bull Targeted Agents will be effective accross different tumor types with that target

ndash importance of organ of origin

ndashAnswer is NO

bull For combination therapies one must demonstrate the antitumor effects for each individual agent

ndashAnswer is NO

PUTS AN END TO TWO PARADIGMS

Molecular profiling

Identification of the molecular alteration

Targeted therapy according to the molecular profile

Tumor Specimen

Precision Medicine identify-hit the target

GUSTAVE ROUSSY PCM PROGRAM

Rational Genomics lsquorsquoMolecular Portraitsrsquorsquo for targeted therapy allocation Rapid through Clinical Trials Logistics ndash Logistics ndash Logistics Focus time pressure culture infrastructure Examples of Current Clinical Trials

SAFIR01 Molecular Screening in

Breast Cancer

Which candidate target FGFR1

FGFR2

FGF4 amp

NOTCH amp

Genetic

instability

PAK1 ampli

PIK3CA AKT PTEN

alteration

VEGFA

amplification

Target

discovery

Primary endpoint

of patients included

in phase III trial according to the

profile

Biopsy of metastatic sites

Frozen sample

CGHhot spot mutations

(PIK3CAAKT)

eligible for phase I

N= 400

Trial A

Trial F

Trial E

Trial D

Trial C

Trial B

Trials XYhellip

Funded by INCA Sanger 30 genes

Andreacute et al ESMO 2012 Lancet Oncol 2014

High level of expectation

Andreacute ESMO 2012

Inclusions

Recruitment completed between May 2011 and August 2012

Molecular alterations

Targetable alterations

Rare alterations

0

5

10

15

20

25

o

f p

atie

nts

wit

h a

vaila

ble

gen

om

ic r

esu

lt mutations

copy number alterations

Andreacute et al Lancet Oncology 2014

29 molecular alteration

IN THE END ONLY 12

gets targeted therapy

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

1200 PATIENTS IN 4 Years

MATCH-R trial

In vitro Cell lines Mouse avatar

Patients with + biomarker tumor exposed to a targeted therapy and an initial response

Resistant Sensitive

Tumor biopy or effusion

Biopsy metastatic site NGS

Array CGH

Chemotherapy 4 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

EGFR TKI if SCC

No PD metastatic NSCLC fisrt line chemotherapy

RANDOMIZED TRIAL SAFIR 02 LUNG

All histologies

Pemetrexed if Non-SCC Molecular alteration Excluding EGFR mut and ALK trl

Genetic abnormality Gene location Squamous Cell

Carcinoma Adenocarcinoma

Therapeutic

intrevention

PIK3CA amplification[ 3q263 33 6 AZD2014

(TORC12)

FGFR1 amplification[ 8p12 22 1 AZD4547

(FGFR)

PTEN mutation 10q233 10 2 AZD8186 (PI3Kbeta)

MET amplification 7q311 3-21 3-21 Volitinib

PTEN loss 10q233 8-20 8-20 AZD8186 (PI3Kbeta)

KRAS mutation[ 12p121 6 21

AZD6244

(MEKi)

Or combo with

AZD2014

LKB1 mutation 19p133 5 23 AZD2014

(TORC12)

HER 2 amplification 17q112-q12 17q21 3-5 5-9 AZD 8931

(pan-HER)

PIK3CA mutation 3q263 3 3 AZD2014

(TORC12)

RET translocation 10q112 2 1

AZD6474

(VEGFR EGFR RET)

BRAF mutation 7p34 2 1-3 AZD6244

(MEKi)

AKT1 mutation 14q3232 1 Very rare AZD5363

(Akt)

MET mutation 7q311 1 2 Volitinib

HER2 mutation 17q112-q12 17q21 1 2 AZD 8931

(pan-HER)

Get COMPLETE PIPELINES

Biopsy metastatic site NGS

CGH 51 alterations

Chemotherapy 6-8 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

Chemotherapy continuation

No PD metastatic Breast cancer patient 1st or

2nd line

RANDOMIZED TRIAL SAFIR 02 BREAST

Molecular alteration Excluding HER2

SAFIR02

BreastAndre

520600

Since 2010 Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics)

SAFIR01

(Andre)

427427

MOSCATO

(Soria)

11501200

SAFIR02

LungSoria

480600

MOST

preSAFIR

(Andre)

108108

SHIVA

(Letourneau

Pilot study 1st Gener Trials 2nd Gener

No NGS NGS WES Randomized trials Sponsor

Gustave Roussy monocentric

Gustave RoussyUnicancer multicenter

L BeacuterardLyon

Curie

Unified Database Pick-up

the winner targets

2nd generation Algorithm for Personnalized

medicine

WINTHER

150200

Profiler

MATCH-R

(WES PDX cfDNA)

200600

FUNDING TOTAL ~50 Million Fondation GR (10) MCM Building (15) IHU (6) INCA (6) ARC (4) Philanthropia (2) WINconsort (4) EU-FP7 (3)

MSN LungMel

BesseRobert

600600

Gustave RoussyWIN multicenter

MOSKIDO

(Goerger)

80100

GETUG

Fizazi

3580 ACSE

Vassal

600

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

12001200 PATIENTS IN 35 Years

bull ATTRITION RATE

bull 100 pts with molecular portrait

bull 50 pts have target identified

bull 25 are actionable

bull 25 overall response rate

bull So in the end 6100 patients benefithellip

bull INNATE RESISTANCE + ORGAN CONTEXT

PROBLEMS with

Molecular Portraits

NEEDS

bull Higher Target Identification Rate

bull Higher of Actionable Targets

bull Higher number diversification of new drugs

bull Rational intrainterpathway combinations

bull Functional Genetics (Crosstalk) ndash Rene Bernards

bull Nodes of convergence ndash Vagner Robert

bull Simplification of models ndash Master Regulators (Andrea Califano)

31

Problems with Targeted Drugs

bull Lack of Durability of responses

ndash Improvement with comborsquos limited

ndash Comborsquos of non-active drugs can be

active

bull Lack of Transversality of impact across

tumor types

ndash Organ of origin

ndash Context 32

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

INHIBIT THE INHIBITOR

vs ACTIVATE THE ACTIVATOR

BREAKING TOLERANCE Immune-Checkpoint-Blockers

REVOLUTION IN IMMUNOTHERAPY

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 10: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

bull Tumor by evolution is ldquomoving targetrdquo

bull Heterogeneity and Innate resistance

bull Acquired resistanceAdditional mutations

bull Mono-Dimensional Thinking

We need to address

bull Nodes of Convergence of Pathways

bull Cross Talk Complexity between pathways

Drug Development Challenges

eIF4F Node of Convergence

RasRaf ndash PI3K- Caspase cascade

FGFR

PTEN

PI3K Akt

TOR

KIT

GRB2

SOS Ras

GDP C-Raf

N-Ras

GTP B-Raf

MEK

ERK

ELK

MITF

CDK24

Cyclin D

p16

Amplified

in 30

Amplified

in 30

50-65

V600E

mutation

15 mutation

Amplified or mutated in 20-40

acral and mucosal melanoma

25-50 loss

Frequent loss

Amplified in 10-15

Adapted from Sosman Curr Oncol Rep 11 405 (2009)

Caspase cascade

Nexus eIF4F

12

bull Nodes of Convergence of Pathways

bull Cross Talk Complexity between pathways

Drug Development Challenges

81

520

0

10

20

30

40

50

60

70

80

90

Melanoma Colon cancer

N Engl J Med 2010 363809-19

Kopetz et al ASCO 2010

Differential response of BRAF inhibition in

BRAF mutant melanoma vs colon cancer

CROSS TALK and RESISTANCE

Prahallad et alhelliphellipBernards R Nature 2012

No drug

EGFR inhibitor

BRAF inhibitor

BRAF+EGFR

inhibitor

Human colon cancer growth in mice

16

bull Targeted Agents will be effective accross different tumor types with that target

ndash importance of organ of origin

ndashAnswer is NO

bull For combination therapies one must demonstrate the antitumor effects for each individual agent

ndashAnswer is NO

PUTS AN END TO TWO PARADIGMS

Molecular profiling

Identification of the molecular alteration

Targeted therapy according to the molecular profile

Tumor Specimen

Precision Medicine identify-hit the target

GUSTAVE ROUSSY PCM PROGRAM

Rational Genomics lsquorsquoMolecular Portraitsrsquorsquo for targeted therapy allocation Rapid through Clinical Trials Logistics ndash Logistics ndash Logistics Focus time pressure culture infrastructure Examples of Current Clinical Trials

SAFIR01 Molecular Screening in

Breast Cancer

Which candidate target FGFR1

FGFR2

FGF4 amp

NOTCH amp

Genetic

instability

PAK1 ampli

PIK3CA AKT PTEN

alteration

VEGFA

amplification

Target

discovery

Primary endpoint

of patients included

in phase III trial according to the

profile

Biopsy of metastatic sites

Frozen sample

CGHhot spot mutations

(PIK3CAAKT)

eligible for phase I

N= 400

Trial A

Trial F

Trial E

Trial D

Trial C

Trial B

Trials XYhellip

Funded by INCA Sanger 30 genes

Andreacute et al ESMO 2012 Lancet Oncol 2014

High level of expectation

Andreacute ESMO 2012

Inclusions

Recruitment completed between May 2011 and August 2012

Molecular alterations

Targetable alterations

Rare alterations

0

5

10

15

20

25

o

f p

atie

nts

wit

h a

vaila

ble

gen

om

ic r

esu

lt mutations

copy number alterations

Andreacute et al Lancet Oncology 2014

29 molecular alteration

IN THE END ONLY 12

gets targeted therapy

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

1200 PATIENTS IN 4 Years

MATCH-R trial

In vitro Cell lines Mouse avatar

Patients with + biomarker tumor exposed to a targeted therapy and an initial response

Resistant Sensitive

Tumor biopy or effusion

Biopsy metastatic site NGS

Array CGH

Chemotherapy 4 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

EGFR TKI if SCC

No PD metastatic NSCLC fisrt line chemotherapy

RANDOMIZED TRIAL SAFIR 02 LUNG

All histologies

Pemetrexed if Non-SCC Molecular alteration Excluding EGFR mut and ALK trl

Genetic abnormality Gene location Squamous Cell

Carcinoma Adenocarcinoma

Therapeutic

intrevention

PIK3CA amplification[ 3q263 33 6 AZD2014

(TORC12)

FGFR1 amplification[ 8p12 22 1 AZD4547

(FGFR)

PTEN mutation 10q233 10 2 AZD8186 (PI3Kbeta)

MET amplification 7q311 3-21 3-21 Volitinib

PTEN loss 10q233 8-20 8-20 AZD8186 (PI3Kbeta)

KRAS mutation[ 12p121 6 21

AZD6244

(MEKi)

Or combo with

AZD2014

LKB1 mutation 19p133 5 23 AZD2014

(TORC12)

HER 2 amplification 17q112-q12 17q21 3-5 5-9 AZD 8931

(pan-HER)

PIK3CA mutation 3q263 3 3 AZD2014

(TORC12)

RET translocation 10q112 2 1

AZD6474

(VEGFR EGFR RET)

BRAF mutation 7p34 2 1-3 AZD6244

(MEKi)

AKT1 mutation 14q3232 1 Very rare AZD5363

(Akt)

MET mutation 7q311 1 2 Volitinib

HER2 mutation 17q112-q12 17q21 1 2 AZD 8931

(pan-HER)

Get COMPLETE PIPELINES

Biopsy metastatic site NGS

CGH 51 alterations

Chemotherapy 6-8 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

Chemotherapy continuation

No PD metastatic Breast cancer patient 1st or

2nd line

RANDOMIZED TRIAL SAFIR 02 BREAST

Molecular alteration Excluding HER2

SAFIR02

BreastAndre

520600

Since 2010 Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics)

SAFIR01

(Andre)

427427

MOSCATO

(Soria)

11501200

SAFIR02

LungSoria

480600

MOST

preSAFIR

(Andre)

108108

SHIVA

(Letourneau

Pilot study 1st Gener Trials 2nd Gener

No NGS NGS WES Randomized trials Sponsor

Gustave Roussy monocentric

Gustave RoussyUnicancer multicenter

L BeacuterardLyon

Curie

Unified Database Pick-up

the winner targets

2nd generation Algorithm for Personnalized

medicine

WINTHER

150200

Profiler

MATCH-R

(WES PDX cfDNA)

200600

FUNDING TOTAL ~50 Million Fondation GR (10) MCM Building (15) IHU (6) INCA (6) ARC (4) Philanthropia (2) WINconsort (4) EU-FP7 (3)

MSN LungMel

BesseRobert

600600

Gustave RoussyWIN multicenter

MOSKIDO

(Goerger)

80100

GETUG

Fizazi

3580 ACSE

Vassal

600

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

12001200 PATIENTS IN 35 Years

bull ATTRITION RATE

bull 100 pts with molecular portrait

bull 50 pts have target identified

bull 25 are actionable

bull 25 overall response rate

bull So in the end 6100 patients benefithellip

bull INNATE RESISTANCE + ORGAN CONTEXT

PROBLEMS with

Molecular Portraits

NEEDS

bull Higher Target Identification Rate

bull Higher of Actionable Targets

bull Higher number diversification of new drugs

bull Rational intrainterpathway combinations

bull Functional Genetics (Crosstalk) ndash Rene Bernards

bull Nodes of convergence ndash Vagner Robert

bull Simplification of models ndash Master Regulators (Andrea Califano)

31

Problems with Targeted Drugs

bull Lack of Durability of responses

ndash Improvement with comborsquos limited

ndash Comborsquos of non-active drugs can be

active

bull Lack of Transversality of impact across

tumor types

ndash Organ of origin

ndash Context 32

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

INHIBIT THE INHIBITOR

vs ACTIVATE THE ACTIVATOR

BREAKING TOLERANCE Immune-Checkpoint-Blockers

REVOLUTION IN IMMUNOTHERAPY

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 11: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

eIF4F Node of Convergence

RasRaf ndash PI3K- Caspase cascade

FGFR

PTEN

PI3K Akt

TOR

KIT

GRB2

SOS Ras

GDP C-Raf

N-Ras

GTP B-Raf

MEK

ERK

ELK

MITF

CDK24

Cyclin D

p16

Amplified

in 30

Amplified

in 30

50-65

V600E

mutation

15 mutation

Amplified or mutated in 20-40

acral and mucosal melanoma

25-50 loss

Frequent loss

Amplified in 10-15

Adapted from Sosman Curr Oncol Rep 11 405 (2009)

Caspase cascade

Nexus eIF4F

12

bull Nodes of Convergence of Pathways

bull Cross Talk Complexity between pathways

Drug Development Challenges

81

520

0

10

20

30

40

50

60

70

80

90

Melanoma Colon cancer

N Engl J Med 2010 363809-19

Kopetz et al ASCO 2010

Differential response of BRAF inhibition in

BRAF mutant melanoma vs colon cancer

CROSS TALK and RESISTANCE

Prahallad et alhelliphellipBernards R Nature 2012

No drug

EGFR inhibitor

BRAF inhibitor

BRAF+EGFR

inhibitor

Human colon cancer growth in mice

16

bull Targeted Agents will be effective accross different tumor types with that target

ndash importance of organ of origin

ndashAnswer is NO

bull For combination therapies one must demonstrate the antitumor effects for each individual agent

ndashAnswer is NO

PUTS AN END TO TWO PARADIGMS

Molecular profiling

Identification of the molecular alteration

Targeted therapy according to the molecular profile

Tumor Specimen

Precision Medicine identify-hit the target

GUSTAVE ROUSSY PCM PROGRAM

Rational Genomics lsquorsquoMolecular Portraitsrsquorsquo for targeted therapy allocation Rapid through Clinical Trials Logistics ndash Logistics ndash Logistics Focus time pressure culture infrastructure Examples of Current Clinical Trials

SAFIR01 Molecular Screening in

Breast Cancer

Which candidate target FGFR1

FGFR2

FGF4 amp

NOTCH amp

Genetic

instability

PAK1 ampli

PIK3CA AKT PTEN

alteration

VEGFA

amplification

Target

discovery

Primary endpoint

of patients included

in phase III trial according to the

profile

Biopsy of metastatic sites

Frozen sample

CGHhot spot mutations

(PIK3CAAKT)

eligible for phase I

N= 400

Trial A

Trial F

Trial E

Trial D

Trial C

Trial B

Trials XYhellip

Funded by INCA Sanger 30 genes

Andreacute et al ESMO 2012 Lancet Oncol 2014

High level of expectation

Andreacute ESMO 2012

Inclusions

Recruitment completed between May 2011 and August 2012

Molecular alterations

Targetable alterations

Rare alterations

0

5

10

15

20

25

o

f p

atie

nts

wit

h a

vaila

ble

gen

om

ic r

esu

lt mutations

copy number alterations

Andreacute et al Lancet Oncology 2014

29 molecular alteration

IN THE END ONLY 12

gets targeted therapy

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

1200 PATIENTS IN 4 Years

MATCH-R trial

In vitro Cell lines Mouse avatar

Patients with + biomarker tumor exposed to a targeted therapy and an initial response

Resistant Sensitive

Tumor biopy or effusion

Biopsy metastatic site NGS

Array CGH

Chemotherapy 4 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

EGFR TKI if SCC

No PD metastatic NSCLC fisrt line chemotherapy

RANDOMIZED TRIAL SAFIR 02 LUNG

All histologies

Pemetrexed if Non-SCC Molecular alteration Excluding EGFR mut and ALK trl

Genetic abnormality Gene location Squamous Cell

Carcinoma Adenocarcinoma

Therapeutic

intrevention

PIK3CA amplification[ 3q263 33 6 AZD2014

(TORC12)

FGFR1 amplification[ 8p12 22 1 AZD4547

(FGFR)

PTEN mutation 10q233 10 2 AZD8186 (PI3Kbeta)

MET amplification 7q311 3-21 3-21 Volitinib

PTEN loss 10q233 8-20 8-20 AZD8186 (PI3Kbeta)

KRAS mutation[ 12p121 6 21

AZD6244

(MEKi)

Or combo with

AZD2014

LKB1 mutation 19p133 5 23 AZD2014

(TORC12)

HER 2 amplification 17q112-q12 17q21 3-5 5-9 AZD 8931

(pan-HER)

PIK3CA mutation 3q263 3 3 AZD2014

(TORC12)

RET translocation 10q112 2 1

AZD6474

(VEGFR EGFR RET)

BRAF mutation 7p34 2 1-3 AZD6244

(MEKi)

AKT1 mutation 14q3232 1 Very rare AZD5363

(Akt)

MET mutation 7q311 1 2 Volitinib

HER2 mutation 17q112-q12 17q21 1 2 AZD 8931

(pan-HER)

Get COMPLETE PIPELINES

Biopsy metastatic site NGS

CGH 51 alterations

Chemotherapy 6-8 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

Chemotherapy continuation

No PD metastatic Breast cancer patient 1st or

2nd line

RANDOMIZED TRIAL SAFIR 02 BREAST

Molecular alteration Excluding HER2

SAFIR02

BreastAndre

520600

Since 2010 Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics)

SAFIR01

(Andre)

427427

MOSCATO

(Soria)

11501200

SAFIR02

LungSoria

480600

MOST

preSAFIR

(Andre)

108108

SHIVA

(Letourneau

Pilot study 1st Gener Trials 2nd Gener

No NGS NGS WES Randomized trials Sponsor

Gustave Roussy monocentric

Gustave RoussyUnicancer multicenter

L BeacuterardLyon

Curie

Unified Database Pick-up

the winner targets

2nd generation Algorithm for Personnalized

medicine

WINTHER

150200

Profiler

MATCH-R

(WES PDX cfDNA)

200600

FUNDING TOTAL ~50 Million Fondation GR (10) MCM Building (15) IHU (6) INCA (6) ARC (4) Philanthropia (2) WINconsort (4) EU-FP7 (3)

MSN LungMel

BesseRobert

600600

Gustave RoussyWIN multicenter

MOSKIDO

(Goerger)

80100

GETUG

Fizazi

3580 ACSE

Vassal

600

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

12001200 PATIENTS IN 35 Years

bull ATTRITION RATE

bull 100 pts with molecular portrait

bull 50 pts have target identified

bull 25 are actionable

bull 25 overall response rate

bull So in the end 6100 patients benefithellip

bull INNATE RESISTANCE + ORGAN CONTEXT

PROBLEMS with

Molecular Portraits

NEEDS

bull Higher Target Identification Rate

bull Higher of Actionable Targets

bull Higher number diversification of new drugs

bull Rational intrainterpathway combinations

bull Functional Genetics (Crosstalk) ndash Rene Bernards

bull Nodes of convergence ndash Vagner Robert

bull Simplification of models ndash Master Regulators (Andrea Califano)

31

Problems with Targeted Drugs

bull Lack of Durability of responses

ndash Improvement with comborsquos limited

ndash Comborsquos of non-active drugs can be

active

bull Lack of Transversality of impact across

tumor types

ndash Organ of origin

ndash Context 32

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

INHIBIT THE INHIBITOR

vs ACTIVATE THE ACTIVATOR

BREAKING TOLERANCE Immune-Checkpoint-Blockers

REVOLUTION IN IMMUNOTHERAPY

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 12: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

12

bull Nodes of Convergence of Pathways

bull Cross Talk Complexity between pathways

Drug Development Challenges

81

520

0

10

20

30

40

50

60

70

80

90

Melanoma Colon cancer

N Engl J Med 2010 363809-19

Kopetz et al ASCO 2010

Differential response of BRAF inhibition in

BRAF mutant melanoma vs colon cancer

CROSS TALK and RESISTANCE

Prahallad et alhelliphellipBernards R Nature 2012

No drug

EGFR inhibitor

BRAF inhibitor

BRAF+EGFR

inhibitor

Human colon cancer growth in mice

16

bull Targeted Agents will be effective accross different tumor types with that target

ndash importance of organ of origin

ndashAnswer is NO

bull For combination therapies one must demonstrate the antitumor effects for each individual agent

ndashAnswer is NO

PUTS AN END TO TWO PARADIGMS

Molecular profiling

Identification of the molecular alteration

Targeted therapy according to the molecular profile

Tumor Specimen

Precision Medicine identify-hit the target

GUSTAVE ROUSSY PCM PROGRAM

Rational Genomics lsquorsquoMolecular Portraitsrsquorsquo for targeted therapy allocation Rapid through Clinical Trials Logistics ndash Logistics ndash Logistics Focus time pressure culture infrastructure Examples of Current Clinical Trials

SAFIR01 Molecular Screening in

Breast Cancer

Which candidate target FGFR1

FGFR2

FGF4 amp

NOTCH amp

Genetic

instability

PAK1 ampli

PIK3CA AKT PTEN

alteration

VEGFA

amplification

Target

discovery

Primary endpoint

of patients included

in phase III trial according to the

profile

Biopsy of metastatic sites

Frozen sample

CGHhot spot mutations

(PIK3CAAKT)

eligible for phase I

N= 400

Trial A

Trial F

Trial E

Trial D

Trial C

Trial B

Trials XYhellip

Funded by INCA Sanger 30 genes

Andreacute et al ESMO 2012 Lancet Oncol 2014

High level of expectation

Andreacute ESMO 2012

Inclusions

Recruitment completed between May 2011 and August 2012

Molecular alterations

Targetable alterations

Rare alterations

0

5

10

15

20

25

o

f p

atie

nts

wit

h a

vaila

ble

gen

om

ic r

esu

lt mutations

copy number alterations

Andreacute et al Lancet Oncology 2014

29 molecular alteration

IN THE END ONLY 12

gets targeted therapy

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

1200 PATIENTS IN 4 Years

MATCH-R trial

In vitro Cell lines Mouse avatar

Patients with + biomarker tumor exposed to a targeted therapy and an initial response

Resistant Sensitive

Tumor biopy or effusion

Biopsy metastatic site NGS

Array CGH

Chemotherapy 4 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

EGFR TKI if SCC

No PD metastatic NSCLC fisrt line chemotherapy

RANDOMIZED TRIAL SAFIR 02 LUNG

All histologies

Pemetrexed if Non-SCC Molecular alteration Excluding EGFR mut and ALK trl

Genetic abnormality Gene location Squamous Cell

Carcinoma Adenocarcinoma

Therapeutic

intrevention

PIK3CA amplification[ 3q263 33 6 AZD2014

(TORC12)

FGFR1 amplification[ 8p12 22 1 AZD4547

(FGFR)

PTEN mutation 10q233 10 2 AZD8186 (PI3Kbeta)

MET amplification 7q311 3-21 3-21 Volitinib

PTEN loss 10q233 8-20 8-20 AZD8186 (PI3Kbeta)

KRAS mutation[ 12p121 6 21

AZD6244

(MEKi)

Or combo with

AZD2014

LKB1 mutation 19p133 5 23 AZD2014

(TORC12)

HER 2 amplification 17q112-q12 17q21 3-5 5-9 AZD 8931

(pan-HER)

PIK3CA mutation 3q263 3 3 AZD2014

(TORC12)

RET translocation 10q112 2 1

AZD6474

(VEGFR EGFR RET)

BRAF mutation 7p34 2 1-3 AZD6244

(MEKi)

AKT1 mutation 14q3232 1 Very rare AZD5363

(Akt)

MET mutation 7q311 1 2 Volitinib

HER2 mutation 17q112-q12 17q21 1 2 AZD 8931

(pan-HER)

Get COMPLETE PIPELINES

Biopsy metastatic site NGS

CGH 51 alterations

Chemotherapy 6-8 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

Chemotherapy continuation

No PD metastatic Breast cancer patient 1st or

2nd line

RANDOMIZED TRIAL SAFIR 02 BREAST

Molecular alteration Excluding HER2

SAFIR02

BreastAndre

520600

Since 2010 Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics)

SAFIR01

(Andre)

427427

MOSCATO

(Soria)

11501200

SAFIR02

LungSoria

480600

MOST

preSAFIR

(Andre)

108108

SHIVA

(Letourneau

Pilot study 1st Gener Trials 2nd Gener

No NGS NGS WES Randomized trials Sponsor

Gustave Roussy monocentric

Gustave RoussyUnicancer multicenter

L BeacuterardLyon

Curie

Unified Database Pick-up

the winner targets

2nd generation Algorithm for Personnalized

medicine

WINTHER

150200

Profiler

MATCH-R

(WES PDX cfDNA)

200600

FUNDING TOTAL ~50 Million Fondation GR (10) MCM Building (15) IHU (6) INCA (6) ARC (4) Philanthropia (2) WINconsort (4) EU-FP7 (3)

MSN LungMel

BesseRobert

600600

Gustave RoussyWIN multicenter

MOSKIDO

(Goerger)

80100

GETUG

Fizazi

3580 ACSE

Vassal

600

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

12001200 PATIENTS IN 35 Years

bull ATTRITION RATE

bull 100 pts with molecular portrait

bull 50 pts have target identified

bull 25 are actionable

bull 25 overall response rate

bull So in the end 6100 patients benefithellip

bull INNATE RESISTANCE + ORGAN CONTEXT

PROBLEMS with

Molecular Portraits

NEEDS

bull Higher Target Identification Rate

bull Higher of Actionable Targets

bull Higher number diversification of new drugs

bull Rational intrainterpathway combinations

bull Functional Genetics (Crosstalk) ndash Rene Bernards

bull Nodes of convergence ndash Vagner Robert

bull Simplification of models ndash Master Regulators (Andrea Califano)

31

Problems with Targeted Drugs

bull Lack of Durability of responses

ndash Improvement with comborsquos limited

ndash Comborsquos of non-active drugs can be

active

bull Lack of Transversality of impact across

tumor types

ndash Organ of origin

ndash Context 32

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

INHIBIT THE INHIBITOR

vs ACTIVATE THE ACTIVATOR

BREAKING TOLERANCE Immune-Checkpoint-Blockers

REVOLUTION IN IMMUNOTHERAPY

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 13: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

bull Nodes of Convergence of Pathways

bull Cross Talk Complexity between pathways

Drug Development Challenges

81

520

0

10

20

30

40

50

60

70

80

90

Melanoma Colon cancer

N Engl J Med 2010 363809-19

Kopetz et al ASCO 2010

Differential response of BRAF inhibition in

BRAF mutant melanoma vs colon cancer

CROSS TALK and RESISTANCE

Prahallad et alhelliphellipBernards R Nature 2012

No drug

EGFR inhibitor

BRAF inhibitor

BRAF+EGFR

inhibitor

Human colon cancer growth in mice

16

bull Targeted Agents will be effective accross different tumor types with that target

ndash importance of organ of origin

ndashAnswer is NO

bull For combination therapies one must demonstrate the antitumor effects for each individual agent

ndashAnswer is NO

PUTS AN END TO TWO PARADIGMS

Molecular profiling

Identification of the molecular alteration

Targeted therapy according to the molecular profile

Tumor Specimen

Precision Medicine identify-hit the target

GUSTAVE ROUSSY PCM PROGRAM

Rational Genomics lsquorsquoMolecular Portraitsrsquorsquo for targeted therapy allocation Rapid through Clinical Trials Logistics ndash Logistics ndash Logistics Focus time pressure culture infrastructure Examples of Current Clinical Trials

SAFIR01 Molecular Screening in

Breast Cancer

Which candidate target FGFR1

FGFR2

FGF4 amp

NOTCH amp

Genetic

instability

PAK1 ampli

PIK3CA AKT PTEN

alteration

VEGFA

amplification

Target

discovery

Primary endpoint

of patients included

in phase III trial according to the

profile

Biopsy of metastatic sites

Frozen sample

CGHhot spot mutations

(PIK3CAAKT)

eligible for phase I

N= 400

Trial A

Trial F

Trial E

Trial D

Trial C

Trial B

Trials XYhellip

Funded by INCA Sanger 30 genes

Andreacute et al ESMO 2012 Lancet Oncol 2014

High level of expectation

Andreacute ESMO 2012

Inclusions

Recruitment completed between May 2011 and August 2012

Molecular alterations

Targetable alterations

Rare alterations

0

5

10

15

20

25

o

f p

atie

nts

wit

h a

vaila

ble

gen

om

ic r

esu

lt mutations

copy number alterations

Andreacute et al Lancet Oncology 2014

29 molecular alteration

IN THE END ONLY 12

gets targeted therapy

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

1200 PATIENTS IN 4 Years

MATCH-R trial

In vitro Cell lines Mouse avatar

Patients with + biomarker tumor exposed to a targeted therapy and an initial response

Resistant Sensitive

Tumor biopy or effusion

Biopsy metastatic site NGS

Array CGH

Chemotherapy 4 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

EGFR TKI if SCC

No PD metastatic NSCLC fisrt line chemotherapy

RANDOMIZED TRIAL SAFIR 02 LUNG

All histologies

Pemetrexed if Non-SCC Molecular alteration Excluding EGFR mut and ALK trl

Genetic abnormality Gene location Squamous Cell

Carcinoma Adenocarcinoma

Therapeutic

intrevention

PIK3CA amplification[ 3q263 33 6 AZD2014

(TORC12)

FGFR1 amplification[ 8p12 22 1 AZD4547

(FGFR)

PTEN mutation 10q233 10 2 AZD8186 (PI3Kbeta)

MET amplification 7q311 3-21 3-21 Volitinib

PTEN loss 10q233 8-20 8-20 AZD8186 (PI3Kbeta)

KRAS mutation[ 12p121 6 21

AZD6244

(MEKi)

Or combo with

AZD2014

LKB1 mutation 19p133 5 23 AZD2014

(TORC12)

HER 2 amplification 17q112-q12 17q21 3-5 5-9 AZD 8931

(pan-HER)

PIK3CA mutation 3q263 3 3 AZD2014

(TORC12)

RET translocation 10q112 2 1

AZD6474

(VEGFR EGFR RET)

BRAF mutation 7p34 2 1-3 AZD6244

(MEKi)

AKT1 mutation 14q3232 1 Very rare AZD5363

(Akt)

MET mutation 7q311 1 2 Volitinib

HER2 mutation 17q112-q12 17q21 1 2 AZD 8931

(pan-HER)

Get COMPLETE PIPELINES

Biopsy metastatic site NGS

CGH 51 alterations

Chemotherapy 6-8 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

Chemotherapy continuation

No PD metastatic Breast cancer patient 1st or

2nd line

RANDOMIZED TRIAL SAFIR 02 BREAST

Molecular alteration Excluding HER2

SAFIR02

BreastAndre

520600

Since 2010 Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics)

SAFIR01

(Andre)

427427

MOSCATO

(Soria)

11501200

SAFIR02

LungSoria

480600

MOST

preSAFIR

(Andre)

108108

SHIVA

(Letourneau

Pilot study 1st Gener Trials 2nd Gener

No NGS NGS WES Randomized trials Sponsor

Gustave Roussy monocentric

Gustave RoussyUnicancer multicenter

L BeacuterardLyon

Curie

Unified Database Pick-up

the winner targets

2nd generation Algorithm for Personnalized

medicine

WINTHER

150200

Profiler

MATCH-R

(WES PDX cfDNA)

200600

FUNDING TOTAL ~50 Million Fondation GR (10) MCM Building (15) IHU (6) INCA (6) ARC (4) Philanthropia (2) WINconsort (4) EU-FP7 (3)

MSN LungMel

BesseRobert

600600

Gustave RoussyWIN multicenter

MOSKIDO

(Goerger)

80100

GETUG

Fizazi

3580 ACSE

Vassal

600

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

12001200 PATIENTS IN 35 Years

bull ATTRITION RATE

bull 100 pts with molecular portrait

bull 50 pts have target identified

bull 25 are actionable

bull 25 overall response rate

bull So in the end 6100 patients benefithellip

bull INNATE RESISTANCE + ORGAN CONTEXT

PROBLEMS with

Molecular Portraits

NEEDS

bull Higher Target Identification Rate

bull Higher of Actionable Targets

bull Higher number diversification of new drugs

bull Rational intrainterpathway combinations

bull Functional Genetics (Crosstalk) ndash Rene Bernards

bull Nodes of convergence ndash Vagner Robert

bull Simplification of models ndash Master Regulators (Andrea Califano)

31

Problems with Targeted Drugs

bull Lack of Durability of responses

ndash Improvement with comborsquos limited

ndash Comborsquos of non-active drugs can be

active

bull Lack of Transversality of impact across

tumor types

ndash Organ of origin

ndash Context 32

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

INHIBIT THE INHIBITOR

vs ACTIVATE THE ACTIVATOR

BREAKING TOLERANCE Immune-Checkpoint-Blockers

REVOLUTION IN IMMUNOTHERAPY

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 14: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

81

520

0

10

20

30

40

50

60

70

80

90

Melanoma Colon cancer

N Engl J Med 2010 363809-19

Kopetz et al ASCO 2010

Differential response of BRAF inhibition in

BRAF mutant melanoma vs colon cancer

CROSS TALK and RESISTANCE

Prahallad et alhelliphellipBernards R Nature 2012

No drug

EGFR inhibitor

BRAF inhibitor

BRAF+EGFR

inhibitor

Human colon cancer growth in mice

16

bull Targeted Agents will be effective accross different tumor types with that target

ndash importance of organ of origin

ndashAnswer is NO

bull For combination therapies one must demonstrate the antitumor effects for each individual agent

ndashAnswer is NO

PUTS AN END TO TWO PARADIGMS

Molecular profiling

Identification of the molecular alteration

Targeted therapy according to the molecular profile

Tumor Specimen

Precision Medicine identify-hit the target

GUSTAVE ROUSSY PCM PROGRAM

Rational Genomics lsquorsquoMolecular Portraitsrsquorsquo for targeted therapy allocation Rapid through Clinical Trials Logistics ndash Logistics ndash Logistics Focus time pressure culture infrastructure Examples of Current Clinical Trials

SAFIR01 Molecular Screening in

Breast Cancer

Which candidate target FGFR1

FGFR2

FGF4 amp

NOTCH amp

Genetic

instability

PAK1 ampli

PIK3CA AKT PTEN

alteration

VEGFA

amplification

Target

discovery

Primary endpoint

of patients included

in phase III trial according to the

profile

Biopsy of metastatic sites

Frozen sample

CGHhot spot mutations

(PIK3CAAKT)

eligible for phase I

N= 400

Trial A

Trial F

Trial E

Trial D

Trial C

Trial B

Trials XYhellip

Funded by INCA Sanger 30 genes

Andreacute et al ESMO 2012 Lancet Oncol 2014

High level of expectation

Andreacute ESMO 2012

Inclusions

Recruitment completed between May 2011 and August 2012

Molecular alterations

Targetable alterations

Rare alterations

0

5

10

15

20

25

o

f p

atie

nts

wit

h a

vaila

ble

gen

om

ic r

esu

lt mutations

copy number alterations

Andreacute et al Lancet Oncology 2014

29 molecular alteration

IN THE END ONLY 12

gets targeted therapy

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

1200 PATIENTS IN 4 Years

MATCH-R trial

In vitro Cell lines Mouse avatar

Patients with + biomarker tumor exposed to a targeted therapy and an initial response

Resistant Sensitive

Tumor biopy or effusion

Biopsy metastatic site NGS

Array CGH

Chemotherapy 4 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

EGFR TKI if SCC

No PD metastatic NSCLC fisrt line chemotherapy

RANDOMIZED TRIAL SAFIR 02 LUNG

All histologies

Pemetrexed if Non-SCC Molecular alteration Excluding EGFR mut and ALK trl

Genetic abnormality Gene location Squamous Cell

Carcinoma Adenocarcinoma

Therapeutic

intrevention

PIK3CA amplification[ 3q263 33 6 AZD2014

(TORC12)

FGFR1 amplification[ 8p12 22 1 AZD4547

(FGFR)

PTEN mutation 10q233 10 2 AZD8186 (PI3Kbeta)

MET amplification 7q311 3-21 3-21 Volitinib

PTEN loss 10q233 8-20 8-20 AZD8186 (PI3Kbeta)

KRAS mutation[ 12p121 6 21

AZD6244

(MEKi)

Or combo with

AZD2014

LKB1 mutation 19p133 5 23 AZD2014

(TORC12)

HER 2 amplification 17q112-q12 17q21 3-5 5-9 AZD 8931

(pan-HER)

PIK3CA mutation 3q263 3 3 AZD2014

(TORC12)

RET translocation 10q112 2 1

AZD6474

(VEGFR EGFR RET)

BRAF mutation 7p34 2 1-3 AZD6244

(MEKi)

AKT1 mutation 14q3232 1 Very rare AZD5363

(Akt)

MET mutation 7q311 1 2 Volitinib

HER2 mutation 17q112-q12 17q21 1 2 AZD 8931

(pan-HER)

Get COMPLETE PIPELINES

Biopsy metastatic site NGS

CGH 51 alterations

Chemotherapy 6-8 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

Chemotherapy continuation

No PD metastatic Breast cancer patient 1st or

2nd line

RANDOMIZED TRIAL SAFIR 02 BREAST

Molecular alteration Excluding HER2

SAFIR02

BreastAndre

520600

Since 2010 Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics)

SAFIR01

(Andre)

427427

MOSCATO

(Soria)

11501200

SAFIR02

LungSoria

480600

MOST

preSAFIR

(Andre)

108108

SHIVA

(Letourneau

Pilot study 1st Gener Trials 2nd Gener

No NGS NGS WES Randomized trials Sponsor

Gustave Roussy monocentric

Gustave RoussyUnicancer multicenter

L BeacuterardLyon

Curie

Unified Database Pick-up

the winner targets

2nd generation Algorithm for Personnalized

medicine

WINTHER

150200

Profiler

MATCH-R

(WES PDX cfDNA)

200600

FUNDING TOTAL ~50 Million Fondation GR (10) MCM Building (15) IHU (6) INCA (6) ARC (4) Philanthropia (2) WINconsort (4) EU-FP7 (3)

MSN LungMel

BesseRobert

600600

Gustave RoussyWIN multicenter

MOSKIDO

(Goerger)

80100

GETUG

Fizazi

3580 ACSE

Vassal

600

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

12001200 PATIENTS IN 35 Years

bull ATTRITION RATE

bull 100 pts with molecular portrait

bull 50 pts have target identified

bull 25 are actionable

bull 25 overall response rate

bull So in the end 6100 patients benefithellip

bull INNATE RESISTANCE + ORGAN CONTEXT

PROBLEMS with

Molecular Portraits

NEEDS

bull Higher Target Identification Rate

bull Higher of Actionable Targets

bull Higher number diversification of new drugs

bull Rational intrainterpathway combinations

bull Functional Genetics (Crosstalk) ndash Rene Bernards

bull Nodes of convergence ndash Vagner Robert

bull Simplification of models ndash Master Regulators (Andrea Califano)

31

Problems with Targeted Drugs

bull Lack of Durability of responses

ndash Improvement with comborsquos limited

ndash Comborsquos of non-active drugs can be

active

bull Lack of Transversality of impact across

tumor types

ndash Organ of origin

ndash Context 32

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

INHIBIT THE INHIBITOR

vs ACTIVATE THE ACTIVATOR

BREAKING TOLERANCE Immune-Checkpoint-Blockers

REVOLUTION IN IMMUNOTHERAPY

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 15: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

CROSS TALK and RESISTANCE

Prahallad et alhelliphellipBernards R Nature 2012

No drug

EGFR inhibitor

BRAF inhibitor

BRAF+EGFR

inhibitor

Human colon cancer growth in mice

16

bull Targeted Agents will be effective accross different tumor types with that target

ndash importance of organ of origin

ndashAnswer is NO

bull For combination therapies one must demonstrate the antitumor effects for each individual agent

ndashAnswer is NO

PUTS AN END TO TWO PARADIGMS

Molecular profiling

Identification of the molecular alteration

Targeted therapy according to the molecular profile

Tumor Specimen

Precision Medicine identify-hit the target

GUSTAVE ROUSSY PCM PROGRAM

Rational Genomics lsquorsquoMolecular Portraitsrsquorsquo for targeted therapy allocation Rapid through Clinical Trials Logistics ndash Logistics ndash Logistics Focus time pressure culture infrastructure Examples of Current Clinical Trials

SAFIR01 Molecular Screening in

Breast Cancer

Which candidate target FGFR1

FGFR2

FGF4 amp

NOTCH amp

Genetic

instability

PAK1 ampli

PIK3CA AKT PTEN

alteration

VEGFA

amplification

Target

discovery

Primary endpoint

of patients included

in phase III trial according to the

profile

Biopsy of metastatic sites

Frozen sample

CGHhot spot mutations

(PIK3CAAKT)

eligible for phase I

N= 400

Trial A

Trial F

Trial E

Trial D

Trial C

Trial B

Trials XYhellip

Funded by INCA Sanger 30 genes

Andreacute et al ESMO 2012 Lancet Oncol 2014

High level of expectation

Andreacute ESMO 2012

Inclusions

Recruitment completed between May 2011 and August 2012

Molecular alterations

Targetable alterations

Rare alterations

0

5

10

15

20

25

o

f p

atie

nts

wit

h a

vaila

ble

gen

om

ic r

esu

lt mutations

copy number alterations

Andreacute et al Lancet Oncology 2014

29 molecular alteration

IN THE END ONLY 12

gets targeted therapy

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

1200 PATIENTS IN 4 Years

MATCH-R trial

In vitro Cell lines Mouse avatar

Patients with + biomarker tumor exposed to a targeted therapy and an initial response

Resistant Sensitive

Tumor biopy or effusion

Biopsy metastatic site NGS

Array CGH

Chemotherapy 4 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

EGFR TKI if SCC

No PD metastatic NSCLC fisrt line chemotherapy

RANDOMIZED TRIAL SAFIR 02 LUNG

All histologies

Pemetrexed if Non-SCC Molecular alteration Excluding EGFR mut and ALK trl

Genetic abnormality Gene location Squamous Cell

Carcinoma Adenocarcinoma

Therapeutic

intrevention

PIK3CA amplification[ 3q263 33 6 AZD2014

(TORC12)

FGFR1 amplification[ 8p12 22 1 AZD4547

(FGFR)

PTEN mutation 10q233 10 2 AZD8186 (PI3Kbeta)

MET amplification 7q311 3-21 3-21 Volitinib

PTEN loss 10q233 8-20 8-20 AZD8186 (PI3Kbeta)

KRAS mutation[ 12p121 6 21

AZD6244

(MEKi)

Or combo with

AZD2014

LKB1 mutation 19p133 5 23 AZD2014

(TORC12)

HER 2 amplification 17q112-q12 17q21 3-5 5-9 AZD 8931

(pan-HER)

PIK3CA mutation 3q263 3 3 AZD2014

(TORC12)

RET translocation 10q112 2 1

AZD6474

(VEGFR EGFR RET)

BRAF mutation 7p34 2 1-3 AZD6244

(MEKi)

AKT1 mutation 14q3232 1 Very rare AZD5363

(Akt)

MET mutation 7q311 1 2 Volitinib

HER2 mutation 17q112-q12 17q21 1 2 AZD 8931

(pan-HER)

Get COMPLETE PIPELINES

Biopsy metastatic site NGS

CGH 51 alterations

Chemotherapy 6-8 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

Chemotherapy continuation

No PD metastatic Breast cancer patient 1st or

2nd line

RANDOMIZED TRIAL SAFIR 02 BREAST

Molecular alteration Excluding HER2

SAFIR02

BreastAndre

520600

Since 2010 Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics)

SAFIR01

(Andre)

427427

MOSCATO

(Soria)

11501200

SAFIR02

LungSoria

480600

MOST

preSAFIR

(Andre)

108108

SHIVA

(Letourneau

Pilot study 1st Gener Trials 2nd Gener

No NGS NGS WES Randomized trials Sponsor

Gustave Roussy monocentric

Gustave RoussyUnicancer multicenter

L BeacuterardLyon

Curie

Unified Database Pick-up

the winner targets

2nd generation Algorithm for Personnalized

medicine

WINTHER

150200

Profiler

MATCH-R

(WES PDX cfDNA)

200600

FUNDING TOTAL ~50 Million Fondation GR (10) MCM Building (15) IHU (6) INCA (6) ARC (4) Philanthropia (2) WINconsort (4) EU-FP7 (3)

MSN LungMel

BesseRobert

600600

Gustave RoussyWIN multicenter

MOSKIDO

(Goerger)

80100

GETUG

Fizazi

3580 ACSE

Vassal

600

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

12001200 PATIENTS IN 35 Years

bull ATTRITION RATE

bull 100 pts with molecular portrait

bull 50 pts have target identified

bull 25 are actionable

bull 25 overall response rate

bull So in the end 6100 patients benefithellip

bull INNATE RESISTANCE + ORGAN CONTEXT

PROBLEMS with

Molecular Portraits

NEEDS

bull Higher Target Identification Rate

bull Higher of Actionable Targets

bull Higher number diversification of new drugs

bull Rational intrainterpathway combinations

bull Functional Genetics (Crosstalk) ndash Rene Bernards

bull Nodes of convergence ndash Vagner Robert

bull Simplification of models ndash Master Regulators (Andrea Califano)

31

Problems with Targeted Drugs

bull Lack of Durability of responses

ndash Improvement with comborsquos limited

ndash Comborsquos of non-active drugs can be

active

bull Lack of Transversality of impact across

tumor types

ndash Organ of origin

ndash Context 32

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

INHIBIT THE INHIBITOR

vs ACTIVATE THE ACTIVATOR

BREAKING TOLERANCE Immune-Checkpoint-Blockers

REVOLUTION IN IMMUNOTHERAPY

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 16: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

16

bull Targeted Agents will be effective accross different tumor types with that target

ndash importance of organ of origin

ndashAnswer is NO

bull For combination therapies one must demonstrate the antitumor effects for each individual agent

ndashAnswer is NO

PUTS AN END TO TWO PARADIGMS

Molecular profiling

Identification of the molecular alteration

Targeted therapy according to the molecular profile

Tumor Specimen

Precision Medicine identify-hit the target

GUSTAVE ROUSSY PCM PROGRAM

Rational Genomics lsquorsquoMolecular Portraitsrsquorsquo for targeted therapy allocation Rapid through Clinical Trials Logistics ndash Logistics ndash Logistics Focus time pressure culture infrastructure Examples of Current Clinical Trials

SAFIR01 Molecular Screening in

Breast Cancer

Which candidate target FGFR1

FGFR2

FGF4 amp

NOTCH amp

Genetic

instability

PAK1 ampli

PIK3CA AKT PTEN

alteration

VEGFA

amplification

Target

discovery

Primary endpoint

of patients included

in phase III trial according to the

profile

Biopsy of metastatic sites

Frozen sample

CGHhot spot mutations

(PIK3CAAKT)

eligible for phase I

N= 400

Trial A

Trial F

Trial E

Trial D

Trial C

Trial B

Trials XYhellip

Funded by INCA Sanger 30 genes

Andreacute et al ESMO 2012 Lancet Oncol 2014

High level of expectation

Andreacute ESMO 2012

Inclusions

Recruitment completed between May 2011 and August 2012

Molecular alterations

Targetable alterations

Rare alterations

0

5

10

15

20

25

o

f p

atie

nts

wit

h a

vaila

ble

gen

om

ic r

esu

lt mutations

copy number alterations

Andreacute et al Lancet Oncology 2014

29 molecular alteration

IN THE END ONLY 12

gets targeted therapy

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

1200 PATIENTS IN 4 Years

MATCH-R trial

In vitro Cell lines Mouse avatar

Patients with + biomarker tumor exposed to a targeted therapy and an initial response

Resistant Sensitive

Tumor biopy or effusion

Biopsy metastatic site NGS

Array CGH

Chemotherapy 4 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

EGFR TKI if SCC

No PD metastatic NSCLC fisrt line chemotherapy

RANDOMIZED TRIAL SAFIR 02 LUNG

All histologies

Pemetrexed if Non-SCC Molecular alteration Excluding EGFR mut and ALK trl

Genetic abnormality Gene location Squamous Cell

Carcinoma Adenocarcinoma

Therapeutic

intrevention

PIK3CA amplification[ 3q263 33 6 AZD2014

(TORC12)

FGFR1 amplification[ 8p12 22 1 AZD4547

(FGFR)

PTEN mutation 10q233 10 2 AZD8186 (PI3Kbeta)

MET amplification 7q311 3-21 3-21 Volitinib

PTEN loss 10q233 8-20 8-20 AZD8186 (PI3Kbeta)

KRAS mutation[ 12p121 6 21

AZD6244

(MEKi)

Or combo with

AZD2014

LKB1 mutation 19p133 5 23 AZD2014

(TORC12)

HER 2 amplification 17q112-q12 17q21 3-5 5-9 AZD 8931

(pan-HER)

PIK3CA mutation 3q263 3 3 AZD2014

(TORC12)

RET translocation 10q112 2 1

AZD6474

(VEGFR EGFR RET)

BRAF mutation 7p34 2 1-3 AZD6244

(MEKi)

AKT1 mutation 14q3232 1 Very rare AZD5363

(Akt)

MET mutation 7q311 1 2 Volitinib

HER2 mutation 17q112-q12 17q21 1 2 AZD 8931

(pan-HER)

Get COMPLETE PIPELINES

Biopsy metastatic site NGS

CGH 51 alterations

Chemotherapy 6-8 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

Chemotherapy continuation

No PD metastatic Breast cancer patient 1st or

2nd line

RANDOMIZED TRIAL SAFIR 02 BREAST

Molecular alteration Excluding HER2

SAFIR02

BreastAndre

520600

Since 2010 Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics)

SAFIR01

(Andre)

427427

MOSCATO

(Soria)

11501200

SAFIR02

LungSoria

480600

MOST

preSAFIR

(Andre)

108108

SHIVA

(Letourneau

Pilot study 1st Gener Trials 2nd Gener

No NGS NGS WES Randomized trials Sponsor

Gustave Roussy monocentric

Gustave RoussyUnicancer multicenter

L BeacuterardLyon

Curie

Unified Database Pick-up

the winner targets

2nd generation Algorithm for Personnalized

medicine

WINTHER

150200

Profiler

MATCH-R

(WES PDX cfDNA)

200600

FUNDING TOTAL ~50 Million Fondation GR (10) MCM Building (15) IHU (6) INCA (6) ARC (4) Philanthropia (2) WINconsort (4) EU-FP7 (3)

MSN LungMel

BesseRobert

600600

Gustave RoussyWIN multicenter

MOSKIDO

(Goerger)

80100

GETUG

Fizazi

3580 ACSE

Vassal

600

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

12001200 PATIENTS IN 35 Years

bull ATTRITION RATE

bull 100 pts with molecular portrait

bull 50 pts have target identified

bull 25 are actionable

bull 25 overall response rate

bull So in the end 6100 patients benefithellip

bull INNATE RESISTANCE + ORGAN CONTEXT

PROBLEMS with

Molecular Portraits

NEEDS

bull Higher Target Identification Rate

bull Higher of Actionable Targets

bull Higher number diversification of new drugs

bull Rational intrainterpathway combinations

bull Functional Genetics (Crosstalk) ndash Rene Bernards

bull Nodes of convergence ndash Vagner Robert

bull Simplification of models ndash Master Regulators (Andrea Califano)

31

Problems with Targeted Drugs

bull Lack of Durability of responses

ndash Improvement with comborsquos limited

ndash Comborsquos of non-active drugs can be

active

bull Lack of Transversality of impact across

tumor types

ndash Organ of origin

ndash Context 32

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

INHIBIT THE INHIBITOR

vs ACTIVATE THE ACTIVATOR

BREAKING TOLERANCE Immune-Checkpoint-Blockers

REVOLUTION IN IMMUNOTHERAPY

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 17: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Molecular profiling

Identification of the molecular alteration

Targeted therapy according to the molecular profile

Tumor Specimen

Precision Medicine identify-hit the target

GUSTAVE ROUSSY PCM PROGRAM

Rational Genomics lsquorsquoMolecular Portraitsrsquorsquo for targeted therapy allocation Rapid through Clinical Trials Logistics ndash Logistics ndash Logistics Focus time pressure culture infrastructure Examples of Current Clinical Trials

SAFIR01 Molecular Screening in

Breast Cancer

Which candidate target FGFR1

FGFR2

FGF4 amp

NOTCH amp

Genetic

instability

PAK1 ampli

PIK3CA AKT PTEN

alteration

VEGFA

amplification

Target

discovery

Primary endpoint

of patients included

in phase III trial according to the

profile

Biopsy of metastatic sites

Frozen sample

CGHhot spot mutations

(PIK3CAAKT)

eligible for phase I

N= 400

Trial A

Trial F

Trial E

Trial D

Trial C

Trial B

Trials XYhellip

Funded by INCA Sanger 30 genes

Andreacute et al ESMO 2012 Lancet Oncol 2014

High level of expectation

Andreacute ESMO 2012

Inclusions

Recruitment completed between May 2011 and August 2012

Molecular alterations

Targetable alterations

Rare alterations

0

5

10

15

20

25

o

f p

atie

nts

wit

h a

vaila

ble

gen

om

ic r

esu

lt mutations

copy number alterations

Andreacute et al Lancet Oncology 2014

29 molecular alteration

IN THE END ONLY 12

gets targeted therapy

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

1200 PATIENTS IN 4 Years

MATCH-R trial

In vitro Cell lines Mouse avatar

Patients with + biomarker tumor exposed to a targeted therapy and an initial response

Resistant Sensitive

Tumor biopy or effusion

Biopsy metastatic site NGS

Array CGH

Chemotherapy 4 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

EGFR TKI if SCC

No PD metastatic NSCLC fisrt line chemotherapy

RANDOMIZED TRIAL SAFIR 02 LUNG

All histologies

Pemetrexed if Non-SCC Molecular alteration Excluding EGFR mut and ALK trl

Genetic abnormality Gene location Squamous Cell

Carcinoma Adenocarcinoma

Therapeutic

intrevention

PIK3CA amplification[ 3q263 33 6 AZD2014

(TORC12)

FGFR1 amplification[ 8p12 22 1 AZD4547

(FGFR)

PTEN mutation 10q233 10 2 AZD8186 (PI3Kbeta)

MET amplification 7q311 3-21 3-21 Volitinib

PTEN loss 10q233 8-20 8-20 AZD8186 (PI3Kbeta)

KRAS mutation[ 12p121 6 21

AZD6244

(MEKi)

Or combo with

AZD2014

LKB1 mutation 19p133 5 23 AZD2014

(TORC12)

HER 2 amplification 17q112-q12 17q21 3-5 5-9 AZD 8931

(pan-HER)

PIK3CA mutation 3q263 3 3 AZD2014

(TORC12)

RET translocation 10q112 2 1

AZD6474

(VEGFR EGFR RET)

BRAF mutation 7p34 2 1-3 AZD6244

(MEKi)

AKT1 mutation 14q3232 1 Very rare AZD5363

(Akt)

MET mutation 7q311 1 2 Volitinib

HER2 mutation 17q112-q12 17q21 1 2 AZD 8931

(pan-HER)

Get COMPLETE PIPELINES

Biopsy metastatic site NGS

CGH 51 alterations

Chemotherapy 6-8 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

Chemotherapy continuation

No PD metastatic Breast cancer patient 1st or

2nd line

RANDOMIZED TRIAL SAFIR 02 BREAST

Molecular alteration Excluding HER2

SAFIR02

BreastAndre

520600

Since 2010 Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics)

SAFIR01

(Andre)

427427

MOSCATO

(Soria)

11501200

SAFIR02

LungSoria

480600

MOST

preSAFIR

(Andre)

108108

SHIVA

(Letourneau

Pilot study 1st Gener Trials 2nd Gener

No NGS NGS WES Randomized trials Sponsor

Gustave Roussy monocentric

Gustave RoussyUnicancer multicenter

L BeacuterardLyon

Curie

Unified Database Pick-up

the winner targets

2nd generation Algorithm for Personnalized

medicine

WINTHER

150200

Profiler

MATCH-R

(WES PDX cfDNA)

200600

FUNDING TOTAL ~50 Million Fondation GR (10) MCM Building (15) IHU (6) INCA (6) ARC (4) Philanthropia (2) WINconsort (4) EU-FP7 (3)

MSN LungMel

BesseRobert

600600

Gustave RoussyWIN multicenter

MOSKIDO

(Goerger)

80100

GETUG

Fizazi

3580 ACSE

Vassal

600

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

12001200 PATIENTS IN 35 Years

bull ATTRITION RATE

bull 100 pts with molecular portrait

bull 50 pts have target identified

bull 25 are actionable

bull 25 overall response rate

bull So in the end 6100 patients benefithellip

bull INNATE RESISTANCE + ORGAN CONTEXT

PROBLEMS with

Molecular Portraits

NEEDS

bull Higher Target Identification Rate

bull Higher of Actionable Targets

bull Higher number diversification of new drugs

bull Rational intrainterpathway combinations

bull Functional Genetics (Crosstalk) ndash Rene Bernards

bull Nodes of convergence ndash Vagner Robert

bull Simplification of models ndash Master Regulators (Andrea Califano)

31

Problems with Targeted Drugs

bull Lack of Durability of responses

ndash Improvement with comborsquos limited

ndash Comborsquos of non-active drugs can be

active

bull Lack of Transversality of impact across

tumor types

ndash Organ of origin

ndash Context 32

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

INHIBIT THE INHIBITOR

vs ACTIVATE THE ACTIVATOR

BREAKING TOLERANCE Immune-Checkpoint-Blockers

REVOLUTION IN IMMUNOTHERAPY

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 18: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

GUSTAVE ROUSSY PCM PROGRAM

Rational Genomics lsquorsquoMolecular Portraitsrsquorsquo for targeted therapy allocation Rapid through Clinical Trials Logistics ndash Logistics ndash Logistics Focus time pressure culture infrastructure Examples of Current Clinical Trials

SAFIR01 Molecular Screening in

Breast Cancer

Which candidate target FGFR1

FGFR2

FGF4 amp

NOTCH amp

Genetic

instability

PAK1 ampli

PIK3CA AKT PTEN

alteration

VEGFA

amplification

Target

discovery

Primary endpoint

of patients included

in phase III trial according to the

profile

Biopsy of metastatic sites

Frozen sample

CGHhot spot mutations

(PIK3CAAKT)

eligible for phase I

N= 400

Trial A

Trial F

Trial E

Trial D

Trial C

Trial B

Trials XYhellip

Funded by INCA Sanger 30 genes

Andreacute et al ESMO 2012 Lancet Oncol 2014

High level of expectation

Andreacute ESMO 2012

Inclusions

Recruitment completed between May 2011 and August 2012

Molecular alterations

Targetable alterations

Rare alterations

0

5

10

15

20

25

o

f p

atie

nts

wit

h a

vaila

ble

gen

om

ic r

esu

lt mutations

copy number alterations

Andreacute et al Lancet Oncology 2014

29 molecular alteration

IN THE END ONLY 12

gets targeted therapy

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

1200 PATIENTS IN 4 Years

MATCH-R trial

In vitro Cell lines Mouse avatar

Patients with + biomarker tumor exposed to a targeted therapy and an initial response

Resistant Sensitive

Tumor biopy or effusion

Biopsy metastatic site NGS

Array CGH

Chemotherapy 4 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

EGFR TKI if SCC

No PD metastatic NSCLC fisrt line chemotherapy

RANDOMIZED TRIAL SAFIR 02 LUNG

All histologies

Pemetrexed if Non-SCC Molecular alteration Excluding EGFR mut and ALK trl

Genetic abnormality Gene location Squamous Cell

Carcinoma Adenocarcinoma

Therapeutic

intrevention

PIK3CA amplification[ 3q263 33 6 AZD2014

(TORC12)

FGFR1 amplification[ 8p12 22 1 AZD4547

(FGFR)

PTEN mutation 10q233 10 2 AZD8186 (PI3Kbeta)

MET amplification 7q311 3-21 3-21 Volitinib

PTEN loss 10q233 8-20 8-20 AZD8186 (PI3Kbeta)

KRAS mutation[ 12p121 6 21

AZD6244

(MEKi)

Or combo with

AZD2014

LKB1 mutation 19p133 5 23 AZD2014

(TORC12)

HER 2 amplification 17q112-q12 17q21 3-5 5-9 AZD 8931

(pan-HER)

PIK3CA mutation 3q263 3 3 AZD2014

(TORC12)

RET translocation 10q112 2 1

AZD6474

(VEGFR EGFR RET)

BRAF mutation 7p34 2 1-3 AZD6244

(MEKi)

AKT1 mutation 14q3232 1 Very rare AZD5363

(Akt)

MET mutation 7q311 1 2 Volitinib

HER2 mutation 17q112-q12 17q21 1 2 AZD 8931

(pan-HER)

Get COMPLETE PIPELINES

Biopsy metastatic site NGS

CGH 51 alterations

Chemotherapy 6-8 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

Chemotherapy continuation

No PD metastatic Breast cancer patient 1st or

2nd line

RANDOMIZED TRIAL SAFIR 02 BREAST

Molecular alteration Excluding HER2

SAFIR02

BreastAndre

520600

Since 2010 Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics)

SAFIR01

(Andre)

427427

MOSCATO

(Soria)

11501200

SAFIR02

LungSoria

480600

MOST

preSAFIR

(Andre)

108108

SHIVA

(Letourneau

Pilot study 1st Gener Trials 2nd Gener

No NGS NGS WES Randomized trials Sponsor

Gustave Roussy monocentric

Gustave RoussyUnicancer multicenter

L BeacuterardLyon

Curie

Unified Database Pick-up

the winner targets

2nd generation Algorithm for Personnalized

medicine

WINTHER

150200

Profiler

MATCH-R

(WES PDX cfDNA)

200600

FUNDING TOTAL ~50 Million Fondation GR (10) MCM Building (15) IHU (6) INCA (6) ARC (4) Philanthropia (2) WINconsort (4) EU-FP7 (3)

MSN LungMel

BesseRobert

600600

Gustave RoussyWIN multicenter

MOSKIDO

(Goerger)

80100

GETUG

Fizazi

3580 ACSE

Vassal

600

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

12001200 PATIENTS IN 35 Years

bull ATTRITION RATE

bull 100 pts with molecular portrait

bull 50 pts have target identified

bull 25 are actionable

bull 25 overall response rate

bull So in the end 6100 patients benefithellip

bull INNATE RESISTANCE + ORGAN CONTEXT

PROBLEMS with

Molecular Portraits

NEEDS

bull Higher Target Identification Rate

bull Higher of Actionable Targets

bull Higher number diversification of new drugs

bull Rational intrainterpathway combinations

bull Functional Genetics (Crosstalk) ndash Rene Bernards

bull Nodes of convergence ndash Vagner Robert

bull Simplification of models ndash Master Regulators (Andrea Califano)

31

Problems with Targeted Drugs

bull Lack of Durability of responses

ndash Improvement with comborsquos limited

ndash Comborsquos of non-active drugs can be

active

bull Lack of Transversality of impact across

tumor types

ndash Organ of origin

ndash Context 32

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

INHIBIT THE INHIBITOR

vs ACTIVATE THE ACTIVATOR

BREAKING TOLERANCE Immune-Checkpoint-Blockers

REVOLUTION IN IMMUNOTHERAPY

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 19: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

SAFIR01 Molecular Screening in

Breast Cancer

Which candidate target FGFR1

FGFR2

FGF4 amp

NOTCH amp

Genetic

instability

PAK1 ampli

PIK3CA AKT PTEN

alteration

VEGFA

amplification

Target

discovery

Primary endpoint

of patients included

in phase III trial according to the

profile

Biopsy of metastatic sites

Frozen sample

CGHhot spot mutations

(PIK3CAAKT)

eligible for phase I

N= 400

Trial A

Trial F

Trial E

Trial D

Trial C

Trial B

Trials XYhellip

Funded by INCA Sanger 30 genes

Andreacute et al ESMO 2012 Lancet Oncol 2014

High level of expectation

Andreacute ESMO 2012

Inclusions

Recruitment completed between May 2011 and August 2012

Molecular alterations

Targetable alterations

Rare alterations

0

5

10

15

20

25

o

f p

atie

nts

wit

h a

vaila

ble

gen

om

ic r

esu

lt mutations

copy number alterations

Andreacute et al Lancet Oncology 2014

29 molecular alteration

IN THE END ONLY 12

gets targeted therapy

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

1200 PATIENTS IN 4 Years

MATCH-R trial

In vitro Cell lines Mouse avatar

Patients with + biomarker tumor exposed to a targeted therapy and an initial response

Resistant Sensitive

Tumor biopy or effusion

Biopsy metastatic site NGS

Array CGH

Chemotherapy 4 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

EGFR TKI if SCC

No PD metastatic NSCLC fisrt line chemotherapy

RANDOMIZED TRIAL SAFIR 02 LUNG

All histologies

Pemetrexed if Non-SCC Molecular alteration Excluding EGFR mut and ALK trl

Genetic abnormality Gene location Squamous Cell

Carcinoma Adenocarcinoma

Therapeutic

intrevention

PIK3CA amplification[ 3q263 33 6 AZD2014

(TORC12)

FGFR1 amplification[ 8p12 22 1 AZD4547

(FGFR)

PTEN mutation 10q233 10 2 AZD8186 (PI3Kbeta)

MET amplification 7q311 3-21 3-21 Volitinib

PTEN loss 10q233 8-20 8-20 AZD8186 (PI3Kbeta)

KRAS mutation[ 12p121 6 21

AZD6244

(MEKi)

Or combo with

AZD2014

LKB1 mutation 19p133 5 23 AZD2014

(TORC12)

HER 2 amplification 17q112-q12 17q21 3-5 5-9 AZD 8931

(pan-HER)

PIK3CA mutation 3q263 3 3 AZD2014

(TORC12)

RET translocation 10q112 2 1

AZD6474

(VEGFR EGFR RET)

BRAF mutation 7p34 2 1-3 AZD6244

(MEKi)

AKT1 mutation 14q3232 1 Very rare AZD5363

(Akt)

MET mutation 7q311 1 2 Volitinib

HER2 mutation 17q112-q12 17q21 1 2 AZD 8931

(pan-HER)

Get COMPLETE PIPELINES

Biopsy metastatic site NGS

CGH 51 alterations

Chemotherapy 6-8 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

Chemotherapy continuation

No PD metastatic Breast cancer patient 1st or

2nd line

RANDOMIZED TRIAL SAFIR 02 BREAST

Molecular alteration Excluding HER2

SAFIR02

BreastAndre

520600

Since 2010 Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics)

SAFIR01

(Andre)

427427

MOSCATO

(Soria)

11501200

SAFIR02

LungSoria

480600

MOST

preSAFIR

(Andre)

108108

SHIVA

(Letourneau

Pilot study 1st Gener Trials 2nd Gener

No NGS NGS WES Randomized trials Sponsor

Gustave Roussy monocentric

Gustave RoussyUnicancer multicenter

L BeacuterardLyon

Curie

Unified Database Pick-up

the winner targets

2nd generation Algorithm for Personnalized

medicine

WINTHER

150200

Profiler

MATCH-R

(WES PDX cfDNA)

200600

FUNDING TOTAL ~50 Million Fondation GR (10) MCM Building (15) IHU (6) INCA (6) ARC (4) Philanthropia (2) WINconsort (4) EU-FP7 (3)

MSN LungMel

BesseRobert

600600

Gustave RoussyWIN multicenter

MOSKIDO

(Goerger)

80100

GETUG

Fizazi

3580 ACSE

Vassal

600

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

12001200 PATIENTS IN 35 Years

bull ATTRITION RATE

bull 100 pts with molecular portrait

bull 50 pts have target identified

bull 25 are actionable

bull 25 overall response rate

bull So in the end 6100 patients benefithellip

bull INNATE RESISTANCE + ORGAN CONTEXT

PROBLEMS with

Molecular Portraits

NEEDS

bull Higher Target Identification Rate

bull Higher of Actionable Targets

bull Higher number diversification of new drugs

bull Rational intrainterpathway combinations

bull Functional Genetics (Crosstalk) ndash Rene Bernards

bull Nodes of convergence ndash Vagner Robert

bull Simplification of models ndash Master Regulators (Andrea Califano)

31

Problems with Targeted Drugs

bull Lack of Durability of responses

ndash Improvement with comborsquos limited

ndash Comborsquos of non-active drugs can be

active

bull Lack of Transversality of impact across

tumor types

ndash Organ of origin

ndash Context 32

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

INHIBIT THE INHIBITOR

vs ACTIVATE THE ACTIVATOR

BREAKING TOLERANCE Immune-Checkpoint-Blockers

REVOLUTION IN IMMUNOTHERAPY

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 20: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

High level of expectation

Andreacute ESMO 2012

Inclusions

Recruitment completed between May 2011 and August 2012

Molecular alterations

Targetable alterations

Rare alterations

0

5

10

15

20

25

o

f p

atie

nts

wit

h a

vaila

ble

gen

om

ic r

esu

lt mutations

copy number alterations

Andreacute et al Lancet Oncology 2014

29 molecular alteration

IN THE END ONLY 12

gets targeted therapy

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

1200 PATIENTS IN 4 Years

MATCH-R trial

In vitro Cell lines Mouse avatar

Patients with + biomarker tumor exposed to a targeted therapy and an initial response

Resistant Sensitive

Tumor biopy or effusion

Biopsy metastatic site NGS

Array CGH

Chemotherapy 4 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

EGFR TKI if SCC

No PD metastatic NSCLC fisrt line chemotherapy

RANDOMIZED TRIAL SAFIR 02 LUNG

All histologies

Pemetrexed if Non-SCC Molecular alteration Excluding EGFR mut and ALK trl

Genetic abnormality Gene location Squamous Cell

Carcinoma Adenocarcinoma

Therapeutic

intrevention

PIK3CA amplification[ 3q263 33 6 AZD2014

(TORC12)

FGFR1 amplification[ 8p12 22 1 AZD4547

(FGFR)

PTEN mutation 10q233 10 2 AZD8186 (PI3Kbeta)

MET amplification 7q311 3-21 3-21 Volitinib

PTEN loss 10q233 8-20 8-20 AZD8186 (PI3Kbeta)

KRAS mutation[ 12p121 6 21

AZD6244

(MEKi)

Or combo with

AZD2014

LKB1 mutation 19p133 5 23 AZD2014

(TORC12)

HER 2 amplification 17q112-q12 17q21 3-5 5-9 AZD 8931

(pan-HER)

PIK3CA mutation 3q263 3 3 AZD2014

(TORC12)

RET translocation 10q112 2 1

AZD6474

(VEGFR EGFR RET)

BRAF mutation 7p34 2 1-3 AZD6244

(MEKi)

AKT1 mutation 14q3232 1 Very rare AZD5363

(Akt)

MET mutation 7q311 1 2 Volitinib

HER2 mutation 17q112-q12 17q21 1 2 AZD 8931

(pan-HER)

Get COMPLETE PIPELINES

Biopsy metastatic site NGS

CGH 51 alterations

Chemotherapy 6-8 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

Chemotherapy continuation

No PD metastatic Breast cancer patient 1st or

2nd line

RANDOMIZED TRIAL SAFIR 02 BREAST

Molecular alteration Excluding HER2

SAFIR02

BreastAndre

520600

Since 2010 Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics)

SAFIR01

(Andre)

427427

MOSCATO

(Soria)

11501200

SAFIR02

LungSoria

480600

MOST

preSAFIR

(Andre)

108108

SHIVA

(Letourneau

Pilot study 1st Gener Trials 2nd Gener

No NGS NGS WES Randomized trials Sponsor

Gustave Roussy monocentric

Gustave RoussyUnicancer multicenter

L BeacuterardLyon

Curie

Unified Database Pick-up

the winner targets

2nd generation Algorithm for Personnalized

medicine

WINTHER

150200

Profiler

MATCH-R

(WES PDX cfDNA)

200600

FUNDING TOTAL ~50 Million Fondation GR (10) MCM Building (15) IHU (6) INCA (6) ARC (4) Philanthropia (2) WINconsort (4) EU-FP7 (3)

MSN LungMel

BesseRobert

600600

Gustave RoussyWIN multicenter

MOSKIDO

(Goerger)

80100

GETUG

Fizazi

3580 ACSE

Vassal

600

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

12001200 PATIENTS IN 35 Years

bull ATTRITION RATE

bull 100 pts with molecular portrait

bull 50 pts have target identified

bull 25 are actionable

bull 25 overall response rate

bull So in the end 6100 patients benefithellip

bull INNATE RESISTANCE + ORGAN CONTEXT

PROBLEMS with

Molecular Portraits

NEEDS

bull Higher Target Identification Rate

bull Higher of Actionable Targets

bull Higher number diversification of new drugs

bull Rational intrainterpathway combinations

bull Functional Genetics (Crosstalk) ndash Rene Bernards

bull Nodes of convergence ndash Vagner Robert

bull Simplification of models ndash Master Regulators (Andrea Califano)

31

Problems with Targeted Drugs

bull Lack of Durability of responses

ndash Improvement with comborsquos limited

ndash Comborsquos of non-active drugs can be

active

bull Lack of Transversality of impact across

tumor types

ndash Organ of origin

ndash Context 32

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

INHIBIT THE INHIBITOR

vs ACTIVATE THE ACTIVATOR

BREAKING TOLERANCE Immune-Checkpoint-Blockers

REVOLUTION IN IMMUNOTHERAPY

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 21: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Molecular alterations

Targetable alterations

Rare alterations

0

5

10

15

20

25

o

f p

atie

nts

wit

h a

vaila

ble

gen

om

ic r

esu

lt mutations

copy number alterations

Andreacute et al Lancet Oncology 2014

29 molecular alteration

IN THE END ONLY 12

gets targeted therapy

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

1200 PATIENTS IN 4 Years

MATCH-R trial

In vitro Cell lines Mouse avatar

Patients with + biomarker tumor exposed to a targeted therapy and an initial response

Resistant Sensitive

Tumor biopy or effusion

Biopsy metastatic site NGS

Array CGH

Chemotherapy 4 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

EGFR TKI if SCC

No PD metastatic NSCLC fisrt line chemotherapy

RANDOMIZED TRIAL SAFIR 02 LUNG

All histologies

Pemetrexed if Non-SCC Molecular alteration Excluding EGFR mut and ALK trl

Genetic abnormality Gene location Squamous Cell

Carcinoma Adenocarcinoma

Therapeutic

intrevention

PIK3CA amplification[ 3q263 33 6 AZD2014

(TORC12)

FGFR1 amplification[ 8p12 22 1 AZD4547

(FGFR)

PTEN mutation 10q233 10 2 AZD8186 (PI3Kbeta)

MET amplification 7q311 3-21 3-21 Volitinib

PTEN loss 10q233 8-20 8-20 AZD8186 (PI3Kbeta)

KRAS mutation[ 12p121 6 21

AZD6244

(MEKi)

Or combo with

AZD2014

LKB1 mutation 19p133 5 23 AZD2014

(TORC12)

HER 2 amplification 17q112-q12 17q21 3-5 5-9 AZD 8931

(pan-HER)

PIK3CA mutation 3q263 3 3 AZD2014

(TORC12)

RET translocation 10q112 2 1

AZD6474

(VEGFR EGFR RET)

BRAF mutation 7p34 2 1-3 AZD6244

(MEKi)

AKT1 mutation 14q3232 1 Very rare AZD5363

(Akt)

MET mutation 7q311 1 2 Volitinib

HER2 mutation 17q112-q12 17q21 1 2 AZD 8931

(pan-HER)

Get COMPLETE PIPELINES

Biopsy metastatic site NGS

CGH 51 alterations

Chemotherapy 6-8 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

Chemotherapy continuation

No PD metastatic Breast cancer patient 1st or

2nd line

RANDOMIZED TRIAL SAFIR 02 BREAST

Molecular alteration Excluding HER2

SAFIR02

BreastAndre

520600

Since 2010 Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics)

SAFIR01

(Andre)

427427

MOSCATO

(Soria)

11501200

SAFIR02

LungSoria

480600

MOST

preSAFIR

(Andre)

108108

SHIVA

(Letourneau

Pilot study 1st Gener Trials 2nd Gener

No NGS NGS WES Randomized trials Sponsor

Gustave Roussy monocentric

Gustave RoussyUnicancer multicenter

L BeacuterardLyon

Curie

Unified Database Pick-up

the winner targets

2nd generation Algorithm for Personnalized

medicine

WINTHER

150200

Profiler

MATCH-R

(WES PDX cfDNA)

200600

FUNDING TOTAL ~50 Million Fondation GR (10) MCM Building (15) IHU (6) INCA (6) ARC (4) Philanthropia (2) WINconsort (4) EU-FP7 (3)

MSN LungMel

BesseRobert

600600

Gustave RoussyWIN multicenter

MOSKIDO

(Goerger)

80100

GETUG

Fizazi

3580 ACSE

Vassal

600

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

12001200 PATIENTS IN 35 Years

bull ATTRITION RATE

bull 100 pts with molecular portrait

bull 50 pts have target identified

bull 25 are actionable

bull 25 overall response rate

bull So in the end 6100 patients benefithellip

bull INNATE RESISTANCE + ORGAN CONTEXT

PROBLEMS with

Molecular Portraits

NEEDS

bull Higher Target Identification Rate

bull Higher of Actionable Targets

bull Higher number diversification of new drugs

bull Rational intrainterpathway combinations

bull Functional Genetics (Crosstalk) ndash Rene Bernards

bull Nodes of convergence ndash Vagner Robert

bull Simplification of models ndash Master Regulators (Andrea Califano)

31

Problems with Targeted Drugs

bull Lack of Durability of responses

ndash Improvement with comborsquos limited

ndash Comborsquos of non-active drugs can be

active

bull Lack of Transversality of impact across

tumor types

ndash Organ of origin

ndash Context 32

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

INHIBIT THE INHIBITOR

vs ACTIVATE THE ACTIVATOR

BREAKING TOLERANCE Immune-Checkpoint-Blockers

REVOLUTION IN IMMUNOTHERAPY

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 22: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

1200 PATIENTS IN 4 Years

MATCH-R trial

In vitro Cell lines Mouse avatar

Patients with + biomarker tumor exposed to a targeted therapy and an initial response

Resistant Sensitive

Tumor biopy or effusion

Biopsy metastatic site NGS

Array CGH

Chemotherapy 4 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

EGFR TKI if SCC

No PD metastatic NSCLC fisrt line chemotherapy

RANDOMIZED TRIAL SAFIR 02 LUNG

All histologies

Pemetrexed if Non-SCC Molecular alteration Excluding EGFR mut and ALK trl

Genetic abnormality Gene location Squamous Cell

Carcinoma Adenocarcinoma

Therapeutic

intrevention

PIK3CA amplification[ 3q263 33 6 AZD2014

(TORC12)

FGFR1 amplification[ 8p12 22 1 AZD4547

(FGFR)

PTEN mutation 10q233 10 2 AZD8186 (PI3Kbeta)

MET amplification 7q311 3-21 3-21 Volitinib

PTEN loss 10q233 8-20 8-20 AZD8186 (PI3Kbeta)

KRAS mutation[ 12p121 6 21

AZD6244

(MEKi)

Or combo with

AZD2014

LKB1 mutation 19p133 5 23 AZD2014

(TORC12)

HER 2 amplification 17q112-q12 17q21 3-5 5-9 AZD 8931

(pan-HER)

PIK3CA mutation 3q263 3 3 AZD2014

(TORC12)

RET translocation 10q112 2 1

AZD6474

(VEGFR EGFR RET)

BRAF mutation 7p34 2 1-3 AZD6244

(MEKi)

AKT1 mutation 14q3232 1 Very rare AZD5363

(Akt)

MET mutation 7q311 1 2 Volitinib

HER2 mutation 17q112-q12 17q21 1 2 AZD 8931

(pan-HER)

Get COMPLETE PIPELINES

Biopsy metastatic site NGS

CGH 51 alterations

Chemotherapy 6-8 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

Chemotherapy continuation

No PD metastatic Breast cancer patient 1st or

2nd line

RANDOMIZED TRIAL SAFIR 02 BREAST

Molecular alteration Excluding HER2

SAFIR02

BreastAndre

520600

Since 2010 Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics)

SAFIR01

(Andre)

427427

MOSCATO

(Soria)

11501200

SAFIR02

LungSoria

480600

MOST

preSAFIR

(Andre)

108108

SHIVA

(Letourneau

Pilot study 1st Gener Trials 2nd Gener

No NGS NGS WES Randomized trials Sponsor

Gustave Roussy monocentric

Gustave RoussyUnicancer multicenter

L BeacuterardLyon

Curie

Unified Database Pick-up

the winner targets

2nd generation Algorithm for Personnalized

medicine

WINTHER

150200

Profiler

MATCH-R

(WES PDX cfDNA)

200600

FUNDING TOTAL ~50 Million Fondation GR (10) MCM Building (15) IHU (6) INCA (6) ARC (4) Philanthropia (2) WINconsort (4) EU-FP7 (3)

MSN LungMel

BesseRobert

600600

Gustave RoussyWIN multicenter

MOSKIDO

(Goerger)

80100

GETUG

Fizazi

3580 ACSE

Vassal

600

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

12001200 PATIENTS IN 35 Years

bull ATTRITION RATE

bull 100 pts with molecular portrait

bull 50 pts have target identified

bull 25 are actionable

bull 25 overall response rate

bull So in the end 6100 patients benefithellip

bull INNATE RESISTANCE + ORGAN CONTEXT

PROBLEMS with

Molecular Portraits

NEEDS

bull Higher Target Identification Rate

bull Higher of Actionable Targets

bull Higher number diversification of new drugs

bull Rational intrainterpathway combinations

bull Functional Genetics (Crosstalk) ndash Rene Bernards

bull Nodes of convergence ndash Vagner Robert

bull Simplification of models ndash Master Regulators (Andrea Califano)

31

Problems with Targeted Drugs

bull Lack of Durability of responses

ndash Improvement with comborsquos limited

ndash Comborsquos of non-active drugs can be

active

bull Lack of Transversality of impact across

tumor types

ndash Organ of origin

ndash Context 32

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

INHIBIT THE INHIBITOR

vs ACTIVATE THE ACTIVATOR

BREAKING TOLERANCE Immune-Checkpoint-Blockers

REVOLUTION IN IMMUNOTHERAPY

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 23: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

MATCH-R trial

In vitro Cell lines Mouse avatar

Patients with + biomarker tumor exposed to a targeted therapy and an initial response

Resistant Sensitive

Tumor biopy or effusion

Biopsy metastatic site NGS

Array CGH

Chemotherapy 4 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

EGFR TKI if SCC

No PD metastatic NSCLC fisrt line chemotherapy

RANDOMIZED TRIAL SAFIR 02 LUNG

All histologies

Pemetrexed if Non-SCC Molecular alteration Excluding EGFR mut and ALK trl

Genetic abnormality Gene location Squamous Cell

Carcinoma Adenocarcinoma

Therapeutic

intrevention

PIK3CA amplification[ 3q263 33 6 AZD2014

(TORC12)

FGFR1 amplification[ 8p12 22 1 AZD4547

(FGFR)

PTEN mutation 10q233 10 2 AZD8186 (PI3Kbeta)

MET amplification 7q311 3-21 3-21 Volitinib

PTEN loss 10q233 8-20 8-20 AZD8186 (PI3Kbeta)

KRAS mutation[ 12p121 6 21

AZD6244

(MEKi)

Or combo with

AZD2014

LKB1 mutation 19p133 5 23 AZD2014

(TORC12)

HER 2 amplification 17q112-q12 17q21 3-5 5-9 AZD 8931

(pan-HER)

PIK3CA mutation 3q263 3 3 AZD2014

(TORC12)

RET translocation 10q112 2 1

AZD6474

(VEGFR EGFR RET)

BRAF mutation 7p34 2 1-3 AZD6244

(MEKi)

AKT1 mutation 14q3232 1 Very rare AZD5363

(Akt)

MET mutation 7q311 1 2 Volitinib

HER2 mutation 17q112-q12 17q21 1 2 AZD 8931

(pan-HER)

Get COMPLETE PIPELINES

Biopsy metastatic site NGS

CGH 51 alterations

Chemotherapy 6-8 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

Chemotherapy continuation

No PD metastatic Breast cancer patient 1st or

2nd line

RANDOMIZED TRIAL SAFIR 02 BREAST

Molecular alteration Excluding HER2

SAFIR02

BreastAndre

520600

Since 2010 Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics)

SAFIR01

(Andre)

427427

MOSCATO

(Soria)

11501200

SAFIR02

LungSoria

480600

MOST

preSAFIR

(Andre)

108108

SHIVA

(Letourneau

Pilot study 1st Gener Trials 2nd Gener

No NGS NGS WES Randomized trials Sponsor

Gustave Roussy monocentric

Gustave RoussyUnicancer multicenter

L BeacuterardLyon

Curie

Unified Database Pick-up

the winner targets

2nd generation Algorithm for Personnalized

medicine

WINTHER

150200

Profiler

MATCH-R

(WES PDX cfDNA)

200600

FUNDING TOTAL ~50 Million Fondation GR (10) MCM Building (15) IHU (6) INCA (6) ARC (4) Philanthropia (2) WINconsort (4) EU-FP7 (3)

MSN LungMel

BesseRobert

600600

Gustave RoussyWIN multicenter

MOSKIDO

(Goerger)

80100

GETUG

Fizazi

3580 ACSE

Vassal

600

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

12001200 PATIENTS IN 35 Years

bull ATTRITION RATE

bull 100 pts with molecular portrait

bull 50 pts have target identified

bull 25 are actionable

bull 25 overall response rate

bull So in the end 6100 patients benefithellip

bull INNATE RESISTANCE + ORGAN CONTEXT

PROBLEMS with

Molecular Portraits

NEEDS

bull Higher Target Identification Rate

bull Higher of Actionable Targets

bull Higher number diversification of new drugs

bull Rational intrainterpathway combinations

bull Functional Genetics (Crosstalk) ndash Rene Bernards

bull Nodes of convergence ndash Vagner Robert

bull Simplification of models ndash Master Regulators (Andrea Califano)

31

Problems with Targeted Drugs

bull Lack of Durability of responses

ndash Improvement with comborsquos limited

ndash Comborsquos of non-active drugs can be

active

bull Lack of Transversality of impact across

tumor types

ndash Organ of origin

ndash Context 32

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

INHIBIT THE INHIBITOR

vs ACTIVATE THE ACTIVATOR

BREAKING TOLERANCE Immune-Checkpoint-Blockers

REVOLUTION IN IMMUNOTHERAPY

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 24: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Biopsy metastatic site NGS

Array CGH

Chemotherapy 4 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

EGFR TKI if SCC

No PD metastatic NSCLC fisrt line chemotherapy

RANDOMIZED TRIAL SAFIR 02 LUNG

All histologies

Pemetrexed if Non-SCC Molecular alteration Excluding EGFR mut and ALK trl

Genetic abnormality Gene location Squamous Cell

Carcinoma Adenocarcinoma

Therapeutic

intrevention

PIK3CA amplification[ 3q263 33 6 AZD2014

(TORC12)

FGFR1 amplification[ 8p12 22 1 AZD4547

(FGFR)

PTEN mutation 10q233 10 2 AZD8186 (PI3Kbeta)

MET amplification 7q311 3-21 3-21 Volitinib

PTEN loss 10q233 8-20 8-20 AZD8186 (PI3Kbeta)

KRAS mutation[ 12p121 6 21

AZD6244

(MEKi)

Or combo with

AZD2014

LKB1 mutation 19p133 5 23 AZD2014

(TORC12)

HER 2 amplification 17q112-q12 17q21 3-5 5-9 AZD 8931

(pan-HER)

PIK3CA mutation 3q263 3 3 AZD2014

(TORC12)

RET translocation 10q112 2 1

AZD6474

(VEGFR EGFR RET)

BRAF mutation 7p34 2 1-3 AZD6244

(MEKi)

AKT1 mutation 14q3232 1 Very rare AZD5363

(Akt)

MET mutation 7q311 1 2 Volitinib

HER2 mutation 17q112-q12 17q21 1 2 AZD 8931

(pan-HER)

Get COMPLETE PIPELINES

Biopsy metastatic site NGS

CGH 51 alterations

Chemotherapy 6-8 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

Chemotherapy continuation

No PD metastatic Breast cancer patient 1st or

2nd line

RANDOMIZED TRIAL SAFIR 02 BREAST

Molecular alteration Excluding HER2

SAFIR02

BreastAndre

520600

Since 2010 Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics)

SAFIR01

(Andre)

427427

MOSCATO

(Soria)

11501200

SAFIR02

LungSoria

480600

MOST

preSAFIR

(Andre)

108108

SHIVA

(Letourneau

Pilot study 1st Gener Trials 2nd Gener

No NGS NGS WES Randomized trials Sponsor

Gustave Roussy monocentric

Gustave RoussyUnicancer multicenter

L BeacuterardLyon

Curie

Unified Database Pick-up

the winner targets

2nd generation Algorithm for Personnalized

medicine

WINTHER

150200

Profiler

MATCH-R

(WES PDX cfDNA)

200600

FUNDING TOTAL ~50 Million Fondation GR (10) MCM Building (15) IHU (6) INCA (6) ARC (4) Philanthropia (2) WINconsort (4) EU-FP7 (3)

MSN LungMel

BesseRobert

600600

Gustave RoussyWIN multicenter

MOSKIDO

(Goerger)

80100

GETUG

Fizazi

3580 ACSE

Vassal

600

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

12001200 PATIENTS IN 35 Years

bull ATTRITION RATE

bull 100 pts with molecular portrait

bull 50 pts have target identified

bull 25 are actionable

bull 25 overall response rate

bull So in the end 6100 patients benefithellip

bull INNATE RESISTANCE + ORGAN CONTEXT

PROBLEMS with

Molecular Portraits

NEEDS

bull Higher Target Identification Rate

bull Higher of Actionable Targets

bull Higher number diversification of new drugs

bull Rational intrainterpathway combinations

bull Functional Genetics (Crosstalk) ndash Rene Bernards

bull Nodes of convergence ndash Vagner Robert

bull Simplification of models ndash Master Regulators (Andrea Califano)

31

Problems with Targeted Drugs

bull Lack of Durability of responses

ndash Improvement with comborsquos limited

ndash Comborsquos of non-active drugs can be

active

bull Lack of Transversality of impact across

tumor types

ndash Organ of origin

ndash Context 32

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

INHIBIT THE INHIBITOR

vs ACTIVATE THE ACTIVATOR

BREAKING TOLERANCE Immune-Checkpoint-Blockers

REVOLUTION IN IMMUNOTHERAPY

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 25: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Genetic abnormality Gene location Squamous Cell

Carcinoma Adenocarcinoma

Therapeutic

intrevention

PIK3CA amplification[ 3q263 33 6 AZD2014

(TORC12)

FGFR1 amplification[ 8p12 22 1 AZD4547

(FGFR)

PTEN mutation 10q233 10 2 AZD8186 (PI3Kbeta)

MET amplification 7q311 3-21 3-21 Volitinib

PTEN loss 10q233 8-20 8-20 AZD8186 (PI3Kbeta)

KRAS mutation[ 12p121 6 21

AZD6244

(MEKi)

Or combo with

AZD2014

LKB1 mutation 19p133 5 23 AZD2014

(TORC12)

HER 2 amplification 17q112-q12 17q21 3-5 5-9 AZD 8931

(pan-HER)

PIK3CA mutation 3q263 3 3 AZD2014

(TORC12)

RET translocation 10q112 2 1

AZD6474

(VEGFR EGFR RET)

BRAF mutation 7p34 2 1-3 AZD6244

(MEKi)

AKT1 mutation 14q3232 1 Very rare AZD5363

(Akt)

MET mutation 7q311 1 2 Volitinib

HER2 mutation 17q112-q12 17q21 1 2 AZD 8931

(pan-HER)

Get COMPLETE PIPELINES

Biopsy metastatic site NGS

CGH 51 alterations

Chemotherapy 6-8 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

Chemotherapy continuation

No PD metastatic Breast cancer patient 1st or

2nd line

RANDOMIZED TRIAL SAFIR 02 BREAST

Molecular alteration Excluding HER2

SAFIR02

BreastAndre

520600

Since 2010 Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics)

SAFIR01

(Andre)

427427

MOSCATO

(Soria)

11501200

SAFIR02

LungSoria

480600

MOST

preSAFIR

(Andre)

108108

SHIVA

(Letourneau

Pilot study 1st Gener Trials 2nd Gener

No NGS NGS WES Randomized trials Sponsor

Gustave Roussy monocentric

Gustave RoussyUnicancer multicenter

L BeacuterardLyon

Curie

Unified Database Pick-up

the winner targets

2nd generation Algorithm for Personnalized

medicine

WINTHER

150200

Profiler

MATCH-R

(WES PDX cfDNA)

200600

FUNDING TOTAL ~50 Million Fondation GR (10) MCM Building (15) IHU (6) INCA (6) ARC (4) Philanthropia (2) WINconsort (4) EU-FP7 (3)

MSN LungMel

BesseRobert

600600

Gustave RoussyWIN multicenter

MOSKIDO

(Goerger)

80100

GETUG

Fizazi

3580 ACSE

Vassal

600

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

12001200 PATIENTS IN 35 Years

bull ATTRITION RATE

bull 100 pts with molecular portrait

bull 50 pts have target identified

bull 25 are actionable

bull 25 overall response rate

bull So in the end 6100 patients benefithellip

bull INNATE RESISTANCE + ORGAN CONTEXT

PROBLEMS with

Molecular Portraits

NEEDS

bull Higher Target Identification Rate

bull Higher of Actionable Targets

bull Higher number diversification of new drugs

bull Rational intrainterpathway combinations

bull Functional Genetics (Crosstalk) ndash Rene Bernards

bull Nodes of convergence ndash Vagner Robert

bull Simplification of models ndash Master Regulators (Andrea Califano)

31

Problems with Targeted Drugs

bull Lack of Durability of responses

ndash Improvement with comborsquos limited

ndash Comborsquos of non-active drugs can be

active

bull Lack of Transversality of impact across

tumor types

ndash Organ of origin

ndash Context 32

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

INHIBIT THE INHIBITOR

vs ACTIVATE THE ACTIVATOR

BREAKING TOLERANCE Immune-Checkpoint-Blockers

REVOLUTION IN IMMUNOTHERAPY

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 26: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Biopsy metastatic site NGS

CGH 51 alterations

Chemotherapy 6-8 cycles

No alteration Followed up but not included

R

Targeted therapy According to

Molecular alteration

Chemotherapy continuation

No PD metastatic Breast cancer patient 1st or

2nd line

RANDOMIZED TRIAL SAFIR 02 BREAST

Molecular alteration Excluding HER2

SAFIR02

BreastAndre

520600

Since 2010 Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics)

SAFIR01

(Andre)

427427

MOSCATO

(Soria)

11501200

SAFIR02

LungSoria

480600

MOST

preSAFIR

(Andre)

108108

SHIVA

(Letourneau

Pilot study 1st Gener Trials 2nd Gener

No NGS NGS WES Randomized trials Sponsor

Gustave Roussy monocentric

Gustave RoussyUnicancer multicenter

L BeacuterardLyon

Curie

Unified Database Pick-up

the winner targets

2nd generation Algorithm for Personnalized

medicine

WINTHER

150200

Profiler

MATCH-R

(WES PDX cfDNA)

200600

FUNDING TOTAL ~50 Million Fondation GR (10) MCM Building (15) IHU (6) INCA (6) ARC (4) Philanthropia (2) WINconsort (4) EU-FP7 (3)

MSN LungMel

BesseRobert

600600

Gustave RoussyWIN multicenter

MOSKIDO

(Goerger)

80100

GETUG

Fizazi

3580 ACSE

Vassal

600

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

12001200 PATIENTS IN 35 Years

bull ATTRITION RATE

bull 100 pts with molecular portrait

bull 50 pts have target identified

bull 25 are actionable

bull 25 overall response rate

bull So in the end 6100 patients benefithellip

bull INNATE RESISTANCE + ORGAN CONTEXT

PROBLEMS with

Molecular Portraits

NEEDS

bull Higher Target Identification Rate

bull Higher of Actionable Targets

bull Higher number diversification of new drugs

bull Rational intrainterpathway combinations

bull Functional Genetics (Crosstalk) ndash Rene Bernards

bull Nodes of convergence ndash Vagner Robert

bull Simplification of models ndash Master Regulators (Andrea Califano)

31

Problems with Targeted Drugs

bull Lack of Durability of responses

ndash Improvement with comborsquos limited

ndash Comborsquos of non-active drugs can be

active

bull Lack of Transversality of impact across

tumor types

ndash Organ of origin

ndash Context 32

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

INHIBIT THE INHIBITOR

vs ACTIVATE THE ACTIVATOR

BREAKING TOLERANCE Immune-Checkpoint-Blockers

REVOLUTION IN IMMUNOTHERAPY

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 27: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

SAFIR02

BreastAndre

520600

Since 2010 Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics)

SAFIR01

(Andre)

427427

MOSCATO

(Soria)

11501200

SAFIR02

LungSoria

480600

MOST

preSAFIR

(Andre)

108108

SHIVA

(Letourneau

Pilot study 1st Gener Trials 2nd Gener

No NGS NGS WES Randomized trials Sponsor

Gustave Roussy monocentric

Gustave RoussyUnicancer multicenter

L BeacuterardLyon

Curie

Unified Database Pick-up

the winner targets

2nd generation Algorithm for Personnalized

medicine

WINTHER

150200

Profiler

MATCH-R

(WES PDX cfDNA)

200600

FUNDING TOTAL ~50 Million Fondation GR (10) MCM Building (15) IHU (6) INCA (6) ARC (4) Philanthropia (2) WINconsort (4) EU-FP7 (3)

MSN LungMel

BesseRobert

600600

Gustave RoussyWIN multicenter

MOSKIDO

(Goerger)

80100

GETUG

Fizazi

3580 ACSE

Vassal

600

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

12001200 PATIENTS IN 35 Years

bull ATTRITION RATE

bull 100 pts with molecular portrait

bull 50 pts have target identified

bull 25 are actionable

bull 25 overall response rate

bull So in the end 6100 patients benefithellip

bull INNATE RESISTANCE + ORGAN CONTEXT

PROBLEMS with

Molecular Portraits

NEEDS

bull Higher Target Identification Rate

bull Higher of Actionable Targets

bull Higher number diversification of new drugs

bull Rational intrainterpathway combinations

bull Functional Genetics (Crosstalk) ndash Rene Bernards

bull Nodes of convergence ndash Vagner Robert

bull Simplification of models ndash Master Regulators (Andrea Califano)

31

Problems with Targeted Drugs

bull Lack of Durability of responses

ndash Improvement with comborsquos limited

ndash Comborsquos of non-active drugs can be

active

bull Lack of Transversality of impact across

tumor types

ndash Organ of origin

ndash Context 32

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

INHIBIT THE INHIBITOR

vs ACTIVATE THE ACTIVATOR

BREAKING TOLERANCE Immune-Checkpoint-Blockers

REVOLUTION IN IMMUNOTHERAPY

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 28: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

MOSCATO MOlecular Screening

for CAncer Treatment Optimization

Histological

analysis Bio

psy

Molecular analysis

CGH NGS --- WES

Gene-panel sequencing

14 calendar days

CGH+RNAseq+NGS - WES

phase I candidates

TARGET IDENTIFIED IN 45-50

Targeted therapy in 20-25

12001200 PATIENTS IN 35 Years

bull ATTRITION RATE

bull 100 pts with molecular portrait

bull 50 pts have target identified

bull 25 are actionable

bull 25 overall response rate

bull So in the end 6100 patients benefithellip

bull INNATE RESISTANCE + ORGAN CONTEXT

PROBLEMS with

Molecular Portraits

NEEDS

bull Higher Target Identification Rate

bull Higher of Actionable Targets

bull Higher number diversification of new drugs

bull Rational intrainterpathway combinations

bull Functional Genetics (Crosstalk) ndash Rene Bernards

bull Nodes of convergence ndash Vagner Robert

bull Simplification of models ndash Master Regulators (Andrea Califano)

31

Problems with Targeted Drugs

bull Lack of Durability of responses

ndash Improvement with comborsquos limited

ndash Comborsquos of non-active drugs can be

active

bull Lack of Transversality of impact across

tumor types

ndash Organ of origin

ndash Context 32

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

INHIBIT THE INHIBITOR

vs ACTIVATE THE ACTIVATOR

BREAKING TOLERANCE Immune-Checkpoint-Blockers

REVOLUTION IN IMMUNOTHERAPY

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 29: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

bull ATTRITION RATE

bull 100 pts with molecular portrait

bull 50 pts have target identified

bull 25 are actionable

bull 25 overall response rate

bull So in the end 6100 patients benefithellip

bull INNATE RESISTANCE + ORGAN CONTEXT

PROBLEMS with

Molecular Portraits

NEEDS

bull Higher Target Identification Rate

bull Higher of Actionable Targets

bull Higher number diversification of new drugs

bull Rational intrainterpathway combinations

bull Functional Genetics (Crosstalk) ndash Rene Bernards

bull Nodes of convergence ndash Vagner Robert

bull Simplification of models ndash Master Regulators (Andrea Califano)

31

Problems with Targeted Drugs

bull Lack of Durability of responses

ndash Improvement with comborsquos limited

ndash Comborsquos of non-active drugs can be

active

bull Lack of Transversality of impact across

tumor types

ndash Organ of origin

ndash Context 32

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

INHIBIT THE INHIBITOR

vs ACTIVATE THE ACTIVATOR

BREAKING TOLERANCE Immune-Checkpoint-Blockers

REVOLUTION IN IMMUNOTHERAPY

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 30: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

NEEDS

bull Higher Target Identification Rate

bull Higher of Actionable Targets

bull Higher number diversification of new drugs

bull Rational intrainterpathway combinations

bull Functional Genetics (Crosstalk) ndash Rene Bernards

bull Nodes of convergence ndash Vagner Robert

bull Simplification of models ndash Master Regulators (Andrea Califano)

31

Problems with Targeted Drugs

bull Lack of Durability of responses

ndash Improvement with comborsquos limited

ndash Comborsquos of non-active drugs can be

active

bull Lack of Transversality of impact across

tumor types

ndash Organ of origin

ndash Context 32

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

INHIBIT THE INHIBITOR

vs ACTIVATE THE ACTIVATOR

BREAKING TOLERANCE Immune-Checkpoint-Blockers

REVOLUTION IN IMMUNOTHERAPY

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 31: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Problems with Targeted Drugs

bull Lack of Durability of responses

ndash Improvement with comborsquos limited

ndash Comborsquos of non-active drugs can be

active

bull Lack of Transversality of impact across

tumor types

ndash Organ of origin

ndash Context 32

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

INHIBIT THE INHIBITOR

vs ACTIVATE THE ACTIVATOR

BREAKING TOLERANCE Immune-Checkpoint-Blockers

REVOLUTION IN IMMUNOTHERAPY

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 32: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

THE MELANOMA PARADIGM

MUTATION DRIVEN DRUG DEVELOPMENT

INNOVATIVE IMMUNOMODULATION

INHIBIT THE INHIBITOR

vs ACTIVATE THE ACTIVATOR

BREAKING TOLERANCE Immune-Checkpoint-Blockers

REVOLUTION IN IMMUNOTHERAPY

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 33: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

INHIBIT THE INHIBITOR

vs ACTIVATE THE ACTIVATOR

BREAKING TOLERANCE Immune-Checkpoint-Blockers

REVOLUTION IN IMMUNOTHERAPY

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 34: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Anti CTLA-4 Monoclonal Antibodies

Perpetuate T Cell Activation

Reawaken silenced Immune Responses

IL-2

B7 MHC

TCR

CD28

~ Antigen

APC

T-cell

CTLA-4

B7 MHC

TCR

CD28

~ Antigen

B7 MHC

TCR

CD28 CTLA-4

Antigen

Anti-CTLA-4 mAb

IL-2

~

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 35: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Balancing T cell activation

playing with T cell receptors

bull PD-1 and CTLA4 play

distinct roles in

regulating T cell

immunity

bull CTLA4 modulates early

phases of T cell priming

(naiumlve and memory T

cells)

bull PD-1 is expressed on

antigen-experienced T

cells (TILs and Tregs)

bull PD-1PDL-1 interaction

downregulates overt

inflammation in lesions

bull PDL-1 expressed on

Tumor Cells and

Plasmoid DCs 38

PD-1

CTLA-4

Inhibitory

receptors

Activating

receptors

TIM-3

LAG-3

Blocking

antibodies

Agonistic

antibodies T-cell stimulation

CD28

OX40

CD137

Specific response to tumour regardless of its

characteristics including mutation status

Adapted from Mellman et al Nature 2011480(7378)480ndash489 Pardoll DM Nat Rev Cancer 201212252ndash264

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 36: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Cytokines

IFN

IL2

IL7

IL21

GM-CSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy strategies

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 37: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

ANTI-CTLA4

Ipilimumab Data

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 38: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Potential for long-term survival with IT

anti-CTLA-4 (ipilimumab)

bull Ipilimumab was the first therapy to improve overall survival in unresectable or metastatic melanoma in a randomised phase 3 trial1

41

Median OS months 95 CI HR P value

Ipilimumab + gp100 100 85ndash115 068 lt0001

Ipilimumab 101 80ndash138 066 0003

gp100 64 55ndash87

Pro

po

rtio

n o

f p

ati

en

ts a

live

(

)

Years

0

20

40

60

80

100

0 1 2 3 4

bull In clinical trials most adverse events associated with ipilimumab were immune related and managed using ipilimumab-specific treatment guidelines2

bull Most frequently reported adverse events associated with ipilimumab monotherapy (all grades) in a clinical study were diarrhoea (27) rash (26) and pruritus (26)2

1 Adapted from Hodi FS et al N Engl J Med 2010363711ndash723 2 Data on File Bristol-Myers-Squibb Company Princeton NJ

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 39: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

42

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 40: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Ipilimumab + DTIC in Melanoma in 1st line IMPACT MEDIAN SURVIVAL only 21 MONTHS

Robert C et al

N Engl J Med 2011

DTIC may have rather limited the ipilimumab

effect than enhance it

DITC is does NOT LEAD TO IMMUNOGENIC

CELL DEATH and thus may be a poor

combination therapy candidate

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 41: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Lancet Oncol 2015 May16(5)522-30

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 42: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

EORTC 18071CA184-029

Study Design

INDUCTION

Ipi 10 mgkg

Q3W X4 High-risk stage III

completely resected

melanoma INDUCTION

Placebo (Pbo)

Q3W X4

R

MAINTENANCE

Ipi 10 mgkg

Q12W up to 3 years

MAINTENANCE

Placebo (Pbo)

Q12W up to 3 years

45

Treatment up to a maximum 3 years or until disease progression intolerable toxicity or

withdrawal

N=475

N=476

Week 1 Week 12 Week 24

Stratification factors ndash Stage (IIIA vs IIIB vs IIIC 1-3 positive lymph nodes vs IIIC ge4 positive lymph nodes)

ndash Regions (North America European countries and Australia)

bull Primary Endpoint RFS

ndash Secondary endpoints DMFS and OS

N=951

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 43: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Primary Endpoint Recurrence-free

Survival (IRC)

Ipi Pbo

Eventspatients 234475 294476

HR (95 CI) 075 (064ndash090)

Log-rank P value 00013

2-Year RFS rate () 515 438

3-Year RFS rate () 465 348

Ipi 10 mgkg

Pbo

Patients at Risk

O N

Ipi

Pbo

Pa

tie

nts

Ali

ve

Wit

ho

ut

Re

cu

rre

nc

e (

)

100

90

80

70

60

50

40

30

20

10

0 0 12 24 36 48 60

Months

475

476

276

260

205

193

67

62

5

4

0

0

Median 171 mo

Median 261 mo

Stratified by stage

Data are not yet mature

46

234

294

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 44: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

POST HOC ANALYSES

ULCERATED PRIMARY

VS

NON-ULCERATED PRIMARY

47

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 45: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

EORTC 18071 Importance of

ULCERATION for RFS

48

Microscopic and

macroscopic Stage III

Microscopic Stage III

(sentinel nodes positive)

Macroscopic Stage III

(palpable nodes)

Primary tumor

Ulcerated

(N=400)

Non-ulcer

(N=501)

Ulcerated

(N=187)

Non-ulcer

(N=201)

Ulcerated

(N=213)

Non-ulcer

(N=300)

HR (CI) 063

(045-088)dagger

082

(059-114)dagger

053

(031-090)Dagger

072

(040-131)Dagger

068

(044-105)Dagger

085

(057-128)Dagger

HR hazard ratio for Ipi versus Pbo CI confidence interval at 95 (all patients)

or CI at 99 (subgroups Post hoc analyses)

(1) Cox model stratified by stage at randomization (primary analysis)

daggerCox model treatment effect adjusted by type of lymph node (LN) involvement (micro vs macroscopic) no of LN+ (1 2-3 ge4) ulceration (No Yes) Breslow thickness (le2 gt2-4 or Unknown gt4 mm)

DaggerCox model as above but without type of LN involvement

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 46: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Time to Onset of Grade 2-5 irAEs

49

Median time to onset (ipilimumab)

43 wks (range 03ndash1441)

Skin Gastrointestinal

Hepatic Endocrine

10 mgkg Ipilimumab (n=471)

Placebo (n=474) 035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

035

030

025

020

015

010

005

000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Pro

po

rtio

n W

ith

Even

t

Number of Doses

Median time to onset (ipilimumab)

63 wks (range 03ndash1450)

Median time to onset (ipilimumab)

87 wks (range 19ndash480) Median time to onset (ipilimumab)

108 wks (range 03ndash901)

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 47: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

ANTI-PD1PD1-L

DRUGS OF THE YEAR

20132014201520162017hellip

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 48: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

CTLA-4 and PD-1PDL1

T cell Tumor cell

MHC TCR

PD-L1 PD-1

- - - T cell

Dendritic

cell

MHC TCR

CD28

B7 CTLA-4 - - -

Activation

(cytokines lysis proliferation

migration to tumor)

B7 + + +

+ + +

CTLA-4 Blockade (ipilimumab tremelimumab) PD-1 Blockade (nivolumab pembrolizumab)

anti-CTLA-4

anti-PD-1

Mostly PERIPHERAL

Tumor Microenvironment

+ + +

PD-L2 PD-1

anti-PD-1

- - -

Mostly CENTRAL in LNN

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 49: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Anti-PD1

Nivolumab

Pembrolizumab

Data

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 50: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Efficacy and Safety of the Anti-PD-1

Monoclonal Antibody PEMBROLIZUMAB

(MK-3475) in 411 Patients With Melanoma

Antoni Ribas1 F Stephen Hodi2 Richard Kefford34 Omid Hamid5

Adil Daud6 Jedd D Wolchok7 Wen-Jen Hwu8 Tara C Gangadhar9

Amita Patnaik10 Anthony M Joshua11 Peter Hersey4

Jeffrey Weber12 Roxana Dronca13 Hassane Zarour14

Kevin Gergich15 Xiaoyun (Nicole) Li15 Robert Iannone15

S Peter Kang15 Scot Ebbinghaus15 Caroline Robert16

1University of California Los Angeles CA 2Dana-Farber Cancer Institute Boston MA 3Crown Princess Mary Cancer Centre

Westmead Hospital and Melanoma Institute Australia Sydney Australia 4University of Sydney Sydney Australia 5The Angeles Clinic and Research Institute Los Angeles CA 6University of California

San Francisco CA 7Memorial Sloan-Kettering Cancer Center New York NY 8MD Anderson Cancer Center Houston TX 9Abramson Cancer Center at the University of Pennsylvania Philadelphia PA 10South Texas Accelerated Research Therapeutics

San Antonio TX 11Princess Margaret Hospital Toronto Ontario 12H Lee Moffitt Cancer Center Tampa FL 13Mayo Clinic Rochester MN 14University of Pittsburgh Pittsburgh PA 15Merck amp

Co Inc Whitehouse Station NJ 16Institut Gustave-Roussy Villejuif France

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 51: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Pembrolizumab in advanced Melanoma

Presented by Antoni Ribas

aIn patients with measurable disease at baseline by RECIST v11 by central review and ge1 postbaseline assessment (n = 317)

Percentage changes gt100 were truncated at 100

Analysis cut-off date October 18 2013

Individual Patients Treated With Pembrolizumab -100

-80

-60

-40

-20

0

20

40

60

80

100

Ch

an

ge

Fro

m B

as

eli

ne

in

Su

m o

f

Lo

ng

es

t D

iam

ete

r o

f Ta

rge

t L

es

ion

IPI-T

IPI-N

72

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 52: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Presented by

Time to and Durability of Response Swimmer Plot Pembrolizumab in advanced melanoma

Presented by Antoni Ribas

aOngoing response defined as alive progression free and without new anticancer therapy

Analysis cut-off date October 18 2013

bull 88 of responses ongoinga

bull Median response duration not reached (range 6+ to 76+ weeks)

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 53: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Pembrolizumab ORR

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

a1-sided P values calculated by logistic regression adjusting for doseschedule PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

40

49

13

0

10

20

30

40

50

60

Overall Response Rate

Rat

e

Unselected PD-L1+ PD-L1ndash

P = 00007a

10 mgkg Q2W n = 35

10 mgkg Q3W n = 47

2 mgkg Q3W n = 31

Proportion PD-L1+

86 77 55

Overall response rate to Pembro

Unselected 51 32 39

PD-L1+ 57 39 59

PD-L1ndash 20 9 14

bull Differences in PD-L1 positivity may

partly explain ORR differences between

dosing cohorts

ORR by PD-L1 Positivity

Across DosesSchedules n=113

ORR by PD-L1 Positivity

By DoseSchedule

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 54: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

PFS and OS

Based on Tumor PD-L1 Expression

Presented by Richard Kefford

PD-L1 positivity defined as staining in ge1 of tumor cells Analysis cut-off date 18 October 2013 1 Daud A et al Presented at 2014 Annual AACR Meeting April 5-9 2014 San Diego CA

80 60 40 20 0

0

20

40

60

80

100

PFS

Time weeks

PD-L1+

PD-L1ndash

P = 00051

80 60 40 20 0

0

20

40

60

80

100

Time weeks

OS

PD-L1+

PD-L1ndash

P = 03165

Overall Survival Progression-Free Survival

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 55: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

58

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 56: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

59

Anti-PD1 vs DTIC in BRAF wild type

Advanced Melanoma

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 57: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

BRAFinh in BRAFmutant compared to

anti-PD1 in Wildtype Advanced Melanoma

60

OS 97-136 mts Gain 39 mts

HR 070

PFS 16- 69 mts Gain53mts

HR 038

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 58: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Nivolumab activity equal

in BRAF wild type V600 Mutant

61

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 59: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

62

Nivolumab activity equal

in BRAF wild type V600 Mutant

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 60: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Durable Long-term Survival in Previously Treated

Patients With Advanced Melanoma Who Received

Nivolumab Monotherapy in a Phase I Trial

F Stephen Hodi1 Harriet M Kluger2 Mario Sznol2 Richard D Carvajal3 Donald P

Lawrence4 Michael B Atkins5 John D Powderly6 William H Sharfman7 Igor Puzanov8

David C Smith9 Philip D Leming10 Evan J Lipson7 Janis M Taube7 Robert A

Anders7 Christine E Horak11 Joel Jiang11 David F McDermott12 Jeffrey A Sosman8

Julie R Brahmer7 Drew M Pardoll7 Suzanne L Topalian7

1Dana Farber Cancer Institute Boston MA USA 2Yale University School of Medicine and Smilow Cancer Center Yale-New

Haven Hospital New Haven CT USA 3Columbia University Medical Center New York NY USA 4Massachusetts General

Hospital Cancer Center Boston MA USA 5Georgetown-Lombardi Comprehensive Cancer Center Washington DC USA 6Carolina BioOncology Institute Huntersville NC USA 7The Sidney Kimmel Comprehensive Cancer Center at Johns

Hopkins Baltimore MD USA 8Vanderbilt University Medical Center Nashville TN USA 9University of Michigan Ann

Arbor MI USA 10The Christ Hospital Cancer Center Cincinnati OH USA 11Bristol-Myers Squibb Princeton NJ USA 12Beth Israel Deaconess Medical Center Boston MA USA

AACR 2016 Annual Meeting (Abstract CT001)

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 61: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Overall Survival at 5 Years of Follow-up

Nivolumab in Advanced Melanoma

64

Pro

bab

ilit

y o

f S

urv

iva

l

Months

00

01

02

03

04

05

06

07

08

09

10

0 12 24 36 48 60 72 84 6 18 30 42 54 66 78

Database lock Oct 2015

107 64 86 51 49 41 29 0 3 15 43 36 17 12 1

Number of Patients at Risk

All Patients

All Patients (events 69107) median and 95 CI 173 (125ndash378)

NIVO 3 mgkg (events 1117) median and 95 CI 203 (72ndashNR)

17 11 15 9 8 7 6 1 6 7 6 6 6 0 NIVO 3 mgkg

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 62: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

65

OS Rate (95 CI)

Landmark timepoint All Patients

(N = 107) NIVO 3 mgkg

(n = 17)

12-month 63 (53ndash71) 65 (38ndash82)

24-month 48 (38ndash57) 47 (23ndash68)

36-month 42 (32ndash51) 41 (19ndash63)

48-month 35 (26ndash44) 35 (15ndash57)

60-month 34 (25ndash43) 35 (15ndash57)

Median OS months (95 CI) 173 (125ndash

378) 203 (72-NR)

Database lock Oct 2015

Summary of Overall Survival

Based on Kaplan-Meier estimates

NR not reached

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 63: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

66

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 64: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Pembrolizumab vs ipilimumab OS

67

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 65: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

CHANGING ADJUVANT LANDSCAPE

MELANOMA

bull To be reported

Ipilimumab Trials

ndash EORTC 18071 DMFSOS analysis adjuvant ipilimumab

ndash ECOG 1609 Ipilimumab 3 vs 10 vs HDI

BRAFi (+ MEKi) Trials

ndash Adjuvant vemurafenib ()

ndash Adjuvant dabrafenib + trametinib

bull To be launched Anti-PD1 Trials

ndash EORTC 1235 adjuvant pembrolizumab

ndash ECOGSWOG adjuvant pembrolizumab vs HDI

ndash BMS adjuvant ipilimumab vs nivolumab

68

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 66: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

bull Sample size needed N=950 patients (randomized)

bull Randomization lt 12 weeks after complete lymph node dissection

bull Stratify by Stage by region (Europe Australia NAmerica)

bull AT RELAPSE unblinding ALL PLACEBO PATIENTS CAN GET PEMBRO

bull AT RELAPSE for Pembro patients physicians choice

bull PREDEFINED GROUP OF INTEREST PDL-1 positive patients

bull Coordinator Caroline Robert Study Chair Alexander Eggermont

R

(double blind)

Placebo iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

200 mg anti-PD1 (Pembrolizumab) iv q3 wks for 12 mts or until disease progression unacceptable toxicity or withdrawal

Eligible patient

EORTC 1325

69

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 67: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Anti-PD1

(nivolumab)

(pembrolizumab)

TRANSVERSAL

IMPACT

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 68: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Anti-PD1

(nivolumab)

(pembrolizumab)

LUNG CANCER

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 69: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

73

Nivolumab vs Docetaxel in NSCLC 2nd line

Overall Survival (FDA et al 2015)

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 70: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

74

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 71: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Nivolumab vs Docetaxel 2nd line

Squamous NSCLC

75

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 72: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Nivolumab vs Docetaxel in 2nd line in

Squamous NSCLC

76

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 73: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Nivolumab activity Squamous NSCLC

PD-L1 Expression

77

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 74: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

78

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 75: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Nivolumab vs Docetaxel in 2nd line

NONSquamous NSCLC

79

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 76: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Nivolumab vs Docetaxel in 2nd line in

NONSquamous NSCLC

80

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 77: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

PEMBROLIZUMAB IN NSCLC

ROLE OF PDL-1

81

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 78: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Role PD-L1 expression in

Pembrolizumab NSCLC Therapy

82

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 79: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Nivolumab Versus Investigatorrsquos Choice (IC) for

Recurrent or Metastatic (RM) Head and Neck

Squamous Cell Carcinoma (SCCHN)

CheckMate-141

1The Ohio State University Columbus OH USA 2MD Anderson Cancer Center Houston TX USA 3Centre Leon Berard Lyon France 4Centre Antoine

Lacassagne Nice France 5Stanford Cancer Institute Stanford CA USA 6IRCCS Istituto Nazionale Tumori Milan Italy 7Institute of Cancer Research London UK 8University Hospital Essen Essen Germany 9University of Chicago Chicago IL USA 10Institut Gustave Roussy Villejuif Cedex France 11University of Michigan

Ann Arbor MI USA 12Winship Cancer Institute Emory University Atlanta GA USA 13Hospital Universitario 12 de Octubre Madrid Spain 14Dana-Farber Cancer

Institute Boston MA USA 15Universitaumltsspital Zurich Zurich Switzerland 16Kobe University Hospital Kobe-City Japan 17National Cancer Center Hospital East

Kashiwa Japan 18Bristol-Myers Squibb Princeton NJ USA 19University of Pittsburgh Medical Center and Cancer Institute Pittsburgh PA USA

Maura L Gillison1 George Blumenschein Jr2 Jerome Fayette3 Joel Guigay4 A Dimitrios Colevas5 Lisa Licitra6

Kevin Harrington7 Stefan Kasper8 Everett E Vokes9 Caroline Even10

Francis Worden11 Nabil F Saba12 Lara Carmen Iglesias Docampo13 Robert Haddad14

Tamara Rordorf15 Naomi Kiyota16 Makoto Tahara17 Manish Monga18 Mark Lynch18

William J Geese18 Justin Kopit18 James W Shaw18 Robert L Ferris19

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 80: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

0 3 6 9 12 15 18

Median OS mo (95 CI)

HR (9773

CI)

p-value

Nivolumab (n = 240) 75 (55ndash

91) 070 (051ndash096)

00101 Investigatorrsquos Choice (n =

121) 51 (40ndash

60)

Overall Survival in SCCHN

Nivolumab vs Investig Choice 2nd line

Months

Nivolumab 240 167 109 52 24 7 0

Investigatorrsquos

Choice

121 87 42 17 5 1

No at Risk

0

0

10

20

30

40

50

60

70

80

90

100

Ove

rall

Su

rviv

al (

of

pa

tie

nts

)

1-year OS rate (95 CI)

360 (285ndash434)

166 (86ndash268)

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 81: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Overall Survival in SCCHN

by PD-L1 Expression

PD-L1 Expression lt 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 73) 57 (44ndash127) 089

(054ndash145) Investigatorrsquos Choice (n

= 38) 58 (40ndash98)

PD-L1 Expression ge 1

Median OS mo (95 CI)

HR (95 CI)

Nivolumab (n = 88) 87 (57ndash91) 055

(036ndash083) Investigatorrsquos Choice (n =

61) 46 (38ndash

58)

88 67 44 18 6 0

61 42 20 6 2 0

73 52 33 17 8 3 0

38 29 14 6 2 0 0

Nivolumab

Investigatorrsquos Choice

Nivolumab

Investigatorrsquos

Choice

No at Risk

Ove

rall S

urv

iva

l (

of

pa

tie

nts

)

Nivolumab

Investigatorrsquos Choice

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Months

0 3 6 9 12 15 18

0

10

20

30

40

50

60

70

80

90

100

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 82: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Pembrolizumab in Mesothelioma

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 83: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Pembrolizumab in Mesothelioma

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 84: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

PEMBROLIZUMAB in

GASTRIC CANCER

39 pts median FU 88 months 23 of pts had gt two prior therapies 30

achieved PR 50 of pts some degree of tumor shrinkage median duration of

response 24 weeks (range 8+ to 33+ wks

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 85: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Nivolumab in RCC

90

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 86: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

NO DOSE-RESPONSE EFFECT In RCC

91

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 87: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Nivolumab in 2nd line in RCC

92

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 88: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

93

Nivolumab in 2nd line in RCC

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 89: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

No clear impact PD-L1 Expression

94

Nivolumab in 2nd line in RCC

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 90: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

95

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 91: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Pembrolizumab in Triple Negative

Breast Cancer

96

J Clin Oncol 2016 May 2 pii JCO648931 [Epub ahead of print]

Pembrolizumab in Patients With Advanced Triple-

Negative Breast Cancer Phase Ib KEYNOTE-012 Study Nanda R1 Chow LQ2 Dees EC2 Berger R2 Gupta S2 Geva R2 Pusztai L2 Pathiraja K2 Aktan G2 Cheng JD2 Karantza

V2 Buisseret L2

RESULTS

Among 111 patients with TNBC whose tumor samples were screened for PD-L1

expression 586 had PD-L1-positive tumorsAmong the 27 patients who were

evaluable for antitumor activity the overall response rate was 185 the

median time to response was 179 weeks (range 73 to 324 weeks) and the

median duration of response was not yet reached (range 150 to ge 473 weeks)

CONCLUSION

clinical activity and a potentially acceptable safety profile of pembrolizumab given

every 2 weeks to patients with heavily pretreated advanced TNBC

A single-agent phase II study examining a 200-mg dose given once every 3

weeks (ClinicalTrialsgov identifier NCT02447003) is ongoing

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 92: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Pembrolizumab in Merkel Cell

Carcinoma

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 93: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Pembrolizumab in Merkel Cell Carcinoma

RR 56 and PFS

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 94: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Pembrolizumab in

Hodgkin Lymphoma News | December 08 2014 | ASH 2014 Hematologic Malignancies Leukemia amp

Lymphoma

99

According to Moskowitz (MSKCC) it is believed that

classic Hodgkinrsquos lymphoma may represent a uniquely

vulnerable target for PD-1 blockade

Specifically amplification of 9p241 is frequent in the

disease and results in the overexpression of PD-L1

and PD-L2

ORR 86

CR 21

PR 45

SD 21

DURABILITY Seventeen of the 19 responses were ongoing

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 95: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

100

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 96: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Pembrolizumab in Mismatched

Repair Deficient Tumors

101

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 97: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Pembrolizumab in Mismatched

Repair Deficient Tumors

102

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 98: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Pembrolizumab in Mismatched

Repair Deficient Tumors

103

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 99: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

No more blockbusters

TRANSVERSAL IMPACT IMMUNO

Anti-PD1 is most important drug in history cancer medicine

bull IMMUNOTHERAPY TRANSVERSAL IMPACT

ndash Melanoma approved 2014 will take all first line + adjuvant

ndash Renal approval 2016 all the way to first line

ndash Bladder (ASCOESMO 201415) will take first line

ndash Lung approved 2015 will take first line + adjuvant

ndash Mesothelioma AACR 2016

ndash Head and Neck (ASCO 2015) like lung major results in 2015

ndash OesophStomach (ASCO GI 2015) may take first place

ndash HCC (ASCO 2015) potentially in first place

ndash MSI CRC tumors 60 response rate (ASCO 2015) various types

ndash Various Mismatched Repair Deficient 60 response rate (ASCO 15)

ndash Anal Cancer ESMO 2015

ndash Merkel Cell NEJM 2016

ndash Hodgkin approval in 2016 (ASH 2014)

ndash TNBC JCO 2016 104

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 100: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Mutational Load and Sensitivity

to anti-CTLA4PD1

105

MSI tumors (CRC and others) 60 response rate (ASCO 2015)

CRC MSI RR 62 CRC RR 0 Others MSI RR 60

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 101: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Tumor antigens and response

to Ab CTLA-4

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 102: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

IMMUNO ndash COMBOS

INHIBITOR COMBOS

Anti-CTLA4 + Anti-PD1

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 103: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Initial Report of Overall Survival Rates From a Randomized Phase II

Trial Evaluating the Combination of Nivolumab and Ipilimumab in

Patients With Advanced Melanoma CHECKMATE 069

Michael A Postow1 Jason Chesney2 Anna C Pavlick3 Caroline Robert4 Kenneth Grossmann5

David McDermott6 Gerald Linette7 Nicolas Meyer8 Jeffrey Giguere9 Sanjiv S Agarwala10

Montaser Shaheen11 Marc S Ernstoff12 David R Minor13 April Salama14 Matthew H Taylor15

Patrick A Ott16 Joel Jiang17 Christine Horak17 Paul Gagnier17 Jedd D Wolchok1 F Stephen

Hodi16

1Memorial Sloan Kettering Cancer Center New York NY USA 2University of Louisville Louisville KY USA 3New York University New York NY USA 4Gustave Roussy Villejuif-Paris-Sud France 5Huntsman Cancer Institute Salt Lake City UT USA 6Beth Israel Deaconess Medical Center Boston MA

USA 7Washington University St Louis MO USA 8Institut Universitaire du Cancer Toulouse France 9Greenville Health System Greenville SC USA 10St Lukersquos Cancer Center and Temple University Bethlehem PA USA 11University of New Mexico Albuquerque NM USA 12Cleveland Clinic Cleveland OH

USA 13California Pacific Center for Melanoma Research San Francisco CA USA 14Duke University Durham NC USA 15Oregon Health amp Science University Portland OR USA 16Dana-Farber Cancer Institute Boston MA USA 17Bristol-Myers Squibb Princeton NJ USA

AACR 2016 Annual Meeting (Abstract CT002)

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 104: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Phase II (CheckMate 069) Study Design

109

Eligible patients with unresectable stage III or IV melanoma

bull Treatment-naiumlve bull BRAF WT (N = 109) or

BRAF MT (N = 33) bull Stratified by BRAF

status

R 21

NIVO 1 mgkg

+ IPI

3 mgkg

Placebo +

IPI 3 mgkg

NIVO 3 mgkg

Placebo

Double-blind

Q3W

x4

Q3W

x4

Treat until disease progressiona or unacceptable toxicity

bull IPI-treated patients could receive NIVO monotherapy after disease progression

Q2W

Q2W

aTreatment beyond initial investigator-assessed RECIST v11- defined progression is permitted in patients experiencing clinical benefit and tolerating study

therapy Upon confirmed progression and change of treatment all patients were unblinded

MT = mutation-positive PFS = progression-free survival Q3W = every 3 weeks R = randomization WT = wild-type

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 105: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Response to Treatment

(11 months of follow-up)

110

BRAF WT All Randomized

NIVO+IPI (N = 72)

IPI (N = 37)

NIVO+IPI (N = 95)

IPI (N = 47)

Objective Response Rate ndash (95 CI)a 61 (49‒72) 11 (3‒25) 59 (48‒69) 11 (4‒23)

Best Overall Response ndash b

Complete response 22 0 22 0

Partial response 39 11 37 11

Stable disease 12 35 13 30

Progressive disease 14 41 16 47

Could not be determined

12 14 13 13

aProportion of patients with a confirmed complete or partial response 95 CI is based on Clopper and Pearson method bAs assessed by the investigator with the use of the Response Evaluations Criteria in Solid Tumors version 11

Database lock Jan 2015

Postow et al N Engl J Med 2015 3722006-17

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 106: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Tumor Burden Change From Baseline at

2 Years of Follow-up (All Randomized Patients)

111

Database lock Feb 2016

NIVO + IPI

Median change -70

IPI

Median change +5

Patients

100

75

50

25

0

-25

-50

-75

-100

Best

Red

ucti

on

Fro

m B

aseli

ne in

Targ

et

Lesio

n (

)

= confirmed responder

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 107: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

PFS at 2 Years of Follow-up

(BRAF Wild-type Patients)

112

NIVO + IPI IPI

Death or disease progression nN

3172 2837

Median PFS months (95 CI) NR (86-NR) 44 (28‒53)

HR (95 CI) P value

035 (021‒059) lt00001

NR = not reached

Number of Patients at Risk

72 54 45 0 38 34 34 31 29 18 2 NIVO+ IPI

37 20 9 0 7 4 3 2 2 2 0 IPI

Months

NIVO + IPI

IPI

17 11

55 54

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 108: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

PFS at 2 Years of Follow-up

(All Randomized Patients)

113

47 22 10 0 8 5 4 3 3 3 0 IPI

53

16

51

12

Number of Patients at Risk

Months

95 69 58 0 47 43 43 40 38 24 2 NIVO+ IPI

NIVO + IPI

IPI

NIVO + IPI IPI

Death or disease progression nN

4395 3547

Median PFS months (95 CI) NR (736ndashNR) 30 (27‒51)

HR (95 CI) P value

036 (022‒056) lt00001

NR = not reached

Pro

bab

ilit

y o

f P

rog

ress

ion

-fre

e S

urv

iva

l

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 109: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

OS at 2 Years of Follow-up

(BRAF Wild-type Patients)

114

NIVO + IPI (n = 72)

IPI (n = 37)

Median OS months (95 CI)

NR 248 (103‒NR)

HR (95 CI) 058 (031‒108) Exploratory endpoint

NR = not reached

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Months

79

69

62

53

Number of Patients at Risk

72 63 59 0 58 56 54 52 51 46 5 NIVO+ IPI

37 32 27 0 25 22 21 20 20 17 3 IPI

10

09

08

07

06

05

04

03

02

00

01

3 6 30 9 12 15 18 21 24 27 0

NIVO + IPI

IPI

bull 2237 (60) of patients randomized to IPI crossed over to receive any systemic therapy at progression

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 110: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

10

09

08

07

06

05

04

03

02

00

01

0 3 6 30 9 12 15 18 21 24 27

OS at 2 Years of Follow-up

(All Randomized Patients)

115

bull 3047 (64) of patients randomized to IPI crossed over to receive any systemic therapy at progression

Number of Patients at Risk

95 82 77 0 74 69 67 65 63 57 6 NIVO+ IPI

47 41 36 0 33 29 27 26 25 22 3 IPI

Months

73

64

65

54

NIVO + IPI (N = 95)

IPI (N = 47)

Median OS months (95 CI)

NR NR (119‒NR)

HR (95 CI) 074 (043‒126)

Exploratory endpoint

NR = not reached

NIVO + IPI

IPI

Pro

bab

ilit

y o

f O

ve

rall

Su

rviv

al

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 111: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Most Common Subsequent Therapies

116

All Randomized Patients (n)

NIVO+IPI (N = 95) IPI (N = 47)

Any subsequent therapya 35 (33) 70 (33)

Systemic therapy 28 (27) 64 (30)

Anti-PD-1 agents 18 (17) 62 (29)b

BRAF inhibitor 4 (4) 13 (6)

MEK inhibitor 3 (3) 15 (7)

Investigational agent 4 (4) 4 (2)

Radiotherapy 18 (17) 36 (17)

Surgery 12 (11) 28 (13)

aPatients may have received more than one subsequent therapy bIncluding 26 (55) patients who received NIVO after progression on IPI (per protocol) 3 additional patients received

NIVO or pembrolizumab off study

bull Median time to subsequent therapy Not reached for NIVO+IPI and 61 months (95 CI 42‒74) for IPI

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 112: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

ORR (11 months)

Similar Efficacy Regardless of Tumor PD-L1

Status (All Randomized Patients NIVO + IPI)

117

Database lock Jan 2015 Database lock Feb 2016 Database lock Feb 2016

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

PFS Rate (2 Year)

0

10

20

30

40

50

60

70

80

90

100

PD-L1 ge5 PD-L1 lt5

Perc

ent

OS Rate (2 Year)

58

55 48

48

67

60

Of the 94 all randomized patients treated with the combination 80 (85) had quantifiable PD-L1 expression

30 had PD-L1 tumor expression ge5 and 70 had PD-L1 tumor expression lt5

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 113: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Nivo + Ipi vs Nivolumab vs Ipilimumab in Melanoma CHECKMATE 067

118

Randomized double-blind phase III study

to compare NIVO + IPI or NIVO alone to IPI alone

Unresectable or

Metatastic Melanoma

bull Previously untreated

bull 945 patients

Treat until

progression

or

unacceptable

toxicity

NIVO 3 mgkg Q2W + IPI-matched placebo

NIVO 1 mgkg + IPI 3 mgkg Q3W for 4 doses then

NIVO 3 mgkg Q2W

IPI 3 mgkg Q3W for 4 doses +

NIVO-matched placebo

Randomize

111

Stratify by

bull PD-L1 expression

bull BRAF status

bull AJCC M stage

Verified PD-L1 assay with 5 expression level was used for the stratification

of patients validated PD-L1 assay was used for efficacy analyses

Patients could have been treated beyond progression under protocol-defined circumstances

N=314

N=316

N=315

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 114: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Response to Treatment

NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

ORR (95 CI) 576 (520ndash

632) 437 (381ndash

493) 190 (149ndash

238) Two-sided P value vs IPI lt0001 lt0001 --

Best overall response mdash

Complete response 115 89 22

Partial response 462 348 168

Stable disease 131 108 219

Progressive disease 226 377 489

Unknown 67 79 102

Duration of response (months)

Median (95 CI) NR (131

NR) NR (117

NR) NR (69 NR)

By RECIST v11

NR not reached

119

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 115: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

PFS (Intent-to-Treat) NIVO + IPI (N=314)

NIVO (N=316)

IPI (N=315)

Median PFS months (95 CI)

115 (89ndash167)

69 (43ndash95)

29 (28ndash34)

HR (995 CI) vs IPI

042 (031ndash057)

057 (043ndash076)

--

HR (95 CI) vs NIVO

074 (060ndash

092) -- --

Stratified log-rank Plt000001 vs IPI

Exploratory endpoint

120

No at Risk

314 NIVO + IPI 173 151 65 11 1 219 0

316 NIVO 147 124 50 9 1 177 0

315 IPI 77 54 24 4 0 137 0

0 6 9 12 15 18 3 21

NIVO

NIVO + IPI

IPI

Months

10

09

08

07

06

05

04

03

02

01

00

Pro

po

rtio

n a

liv

e a

nd

pro

gre

ssio

n-f

ree

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 116: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

No at Risk

IPI ndash 202 82 44 31 12 1

NIVO 208 108 88 74 31 5 2

NIVO + IPI 210 142 112 96 42 9 2

0 3 6 9 12 15 17

Months

10

08

06

04

02

00

0 3 6 9 12 15 17

No at Risk

IPI 0 75 40 22 17 9 2

NIVO 80 57 51 43 16 4 0

NIVO + IPI 68 53 44 39 16 1 0

Months

NIVO + IPI

NIVO

IPI

NIVO + IPI

NIVO

IPI

PFS by PD-L1 Expression Level (5) Ipilimumab vs Nivolumab vs Combo

PD-L1 ge5 PD-L1 lt5

Per validated PD-L1 immunohistochemical assay based on PD-L1 staining of tumor cells in a section of at least 100 evaluable tumor cells

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

Pro

po

rtio

n a

live

an

d p

rog

res

sio

n-f

ree

10

08

06

04

02

00

mPFS HR

NIVO + IPI 140 040

NIVO 140 040

IPI 39 --

mPFS HR

NIVO + IPI 112 042

NIVO 53 060

IPI 28 --

121 ( Wolchok ASCO 2015)

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 117: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

122

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 118: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Cytokines

IFN

IL2

IL7

IL21

GmCSF

Vaccination

DC

DNA

RNA

Immunocyte

depletion

Treg

MDSC

Adoptive Tcell

therapy

Activated

TCR engineered

CARs

MoAb-conjugates

Immunotherapy combo strategies

Breaking Tolerance is Prerequisite

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 119: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

Immunotherapy + Other Modalities

Guidance by Immunogenic Cell Death Zitvogel amp Kroemer

124

Uploading of

antigen processing

machinery

CD8 T-cell Adhesion molecules

(CAM-1) and death

receptors (FAS)

Peptide

pools

Vascular normalisation

T-cell initiation

Cytokine release

CD8 T-cells

T-cell function MDSC

Treg cells

Activation of apoptosis

Blockage of cell cycle

TAA

Upregulates MHC-1

Adhesion molecules

death receptors

APM

CD8 T-cells

(homeostatic peripheral

expansion)

MDSC

Treg cells

M2 macrophages

TAA cross-

presentation

CD8 T-cells

Effector immune

infiltrate Release of tumour

antigens (cascade)

Translocation of

calreticulin

Radiation

Chemotherapy Targeted therapies

Dendritic

cell

Upregulation of MHC-1

Adapted from Hodge JW Semin Oncol 201239(3)323ndash339 Drake CG Ann Oncol 201223 Suppl 8viii41-6 Meacutenard C et al Cancer Immunol Immunother 2008571579-87 Hannani D et al Cancer J 201117351-358 Ribas A at al Curr Opin Immunol 201325291-296

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 120: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

PREDICTIONS

bull IMMUNO COMBOS will dominate the scene for years to come

bull Breaking tolerance is first prerequisite and will get Nobel Price

bull Will be backbone for any treatment of metastatic melanoma

patients and many tumors to come

bull Anti-PD-1PD-L1 is key Anti-CTLA4 remains key for ldquotail effectrdquo

bull Neoantigens private antigens sequencing and TcR cloning will

lead further understandingrdquo ldquolet the body decide what is key

bull Will cure ldquoclinically curerdquo gt 50 of advanced melanoma patients

over next 5 years Will stabelize 50 of many tumor types new

ceiling to be broken with new insights

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU

Page 121: Precision Cancer Medicine · Breast/Andre 520/600 Since 2010: Ongoing precision medicine programs 15 GR-initiated trials (high throughput genomics) SAFIR01 (Andre) 427/427 MOSCATO

126

COME VISIT GUSTAVE ROUSSY

Cancer Campus Grand Paris

THANK YOU


ROHIT PRECISION TECHNOLOGIES PVT LTD.3.imimg.com/data3/VY/CI/MY-8085681/cl_mittal-extrusions.pdfROHIT PRECISION TECHNOLOGIES PVT LTD. A VASANT GROUP ENTERPRISE WORKS : GAT NO : 427/2,

ROHIT PRECISION TECHNOLOGIES PVT LTD.3.imimg.com/data3/VY/CI/MY-8085681/cl_mittal-extrusions.pdfROHIT PRECISION TECHNOLOGIES PVT LTD. A VASANT GROUP ENTERPRISE WORKS : GAT NO : 427/2,

Documents
WRT 427: Introduction

WRT 427: Introduction

Documents
Manual Refrey 427

Manual Refrey 427

Documents
Correio Fraterno 427

Correio Fraterno 427

Documents
Mslrp 427-2014sp

Mslrp 427-2014sp

Spiritual
Guyancourt Magazine 427

Guyancourt Magazine 427

Documents
decizionala/HCL IMP SI TAXE...hectare cu suma corespunzatoare prevazuta in urmatorul tabel. LEI/ha 2017 5236 3558 1690 984 569 427 569 569 427 427 284 427 284 142 427 284 284 142 VIZAT

decizionala/HCL IMP SI TAXE...hectare cu suma corespunzatoare prevazuta in urmatorul tabel. LEI/ha 2017 5236 3558 1690 984 569 427 569 569 427 427 284 427 284 142 427 284 284 142 VIZAT

Documents
FreeStyle Precision Neo · FreeStyle Precision Neo bruger enkeltpakkede teststrimler som er sikret mod luftbårne artikler, fugt og andre kontaminanter. Derved sikres at strimlen

FreeStyle Precision Neo · FreeStyle Precision Neo bruger enkeltpakkede teststrimler som er sikret mod luftbårne artikler, fugt og andre kontaminanter. Derved sikres at strimlen

Documents
Edicion 427

Edicion 427

Documents
Anboto 427

Anboto 427

Education
KMBT C554e-20161208134456 · Donna Gambol, Chair r Andre Andre Andre Andre Andre Andre Andre Andre Gambol/ Andre Andre Andre Bèsolutibn- Locked fine to Saw no bees on 913 Remove

KMBT C554e-20161208134456 · Donna Gambol, Chair r Andre Andre Andre Andre Andre Andre Andre Andre Gambol/ Andre Andre Andre Bèsolutibn- Locked fine to Saw no bees on 913 Remove

Documents
skpress #427

skpress #427

Documents
Serviço Social do Comércio - sescgo.com.br · andre luis caiado souza faria andre luiz coelho de matos andre luiz da silva amorim andre luiz mendes ferreira andre pereira lopes

Serviço Social do Comércio - sescgo.com.br · andre luis caiado souza faria andre luiz coelho de matos andre luiz da silva amorim andre luiz mendes ferreira andre pereira lopes

Documents
Semanal 427

Semanal 427

Documents
Jornal 427

Jornal 427

Documents
LEGENDE (Teil 1) Legende FÖRDERGEBIET BARMBEK-NORD S1 ... · P Fußgänger-Passage H H H 427 / 6 427 / 4 427 / 20 427 / 16 427 / 14 427 / 17 427 / 29 427 / 38 421 / 5 427 / 21 427

LEGENDE (Teil 1) Legende FÖRDERGEBIET BARMBEK-NORD S1 ... · P Fußgänger-Passage H H H 427 / 6 427 / 4 427 / 20 427 / 16 427 / 14 427 / 17 427 / 29 427 / 38 421 / 5 427 / 21 427

Documents
INSTRUCTION MANUAL MODEL 427 CURRENT AMPLIFIER 427 Instruction Manual.pdfMODEL 427 GENERAL DESCRIPTION SECTION 1. GENERAL DESCRIPTION l-1. GENERAL; The Model 427 Current Amplifier

INSTRUCTION MANUAL MODEL 427 CURRENT AMPLIFIER 427 Instruction Manual.pdfMODEL 427 GENERAL DESCRIPTION SECTION 1. GENERAL DESCRIPTION l-1. GENERAL; The Model 427 Current Amplifier

Documents
427 Roccia di fedeltà - RNS 427 mi affido a Te

427 Roccia di fedeltà - RNS 427 mi affido a Te

Documents
ROHIT PRECISION TECHNOLOGIES PVT LTD. · ROHIT PRECISION TECHNOLOGIES PVT LTD. A VASANT GROUP ENTERPRISE ... GAT NO : 427/2, Chakan-Talegaon Road, Mahalunge, Chakan, Pune 410501

ROHIT PRECISION TECHNOLOGIES PVT LTD. · ROHIT PRECISION TECHNOLOGIES PVT LTD. A VASANT GROUP ENTERPRISE ... GAT NO : 427/2, Chakan-Talegaon Road, Mahalunge, Chakan, Pune 410501

Documents
427 Report

427 Report

Documents
SPED 427/527

SPED 427/527

Documents
427 Cerâmica 67 (2021) 427-433

427 Cerâmica 67 (2021) 427-433

Documents
427 kiowas

427 kiowas

Documents
grsm.org.ingrsm.org.in/notice/PG BENGALI MERIT LIST_1408463508… ·  · 2014-08-19ruma roy 427 427 427 427 427 427 424 424 423 423 423 423 423 422 422 422 422 421 421 420 419 marks

grsm.org.ingrsm.org.in/notice/PG BENGALI MERIT LIST_1408463508… ·  · 2014-08-19ruma roy 427 427 427 427 427 427 424 424 423 423 423 423 423 422 422 422 422 421 421 420 419 marks

Documents
Sociology 427

Sociology 427

Education
брой76 (427)

брой76 (427)

Documents
Snippetz Issue 427

Snippetz Issue 427

Documents
Internship For 427

Internship For 427

Documents
Opinião Socialista 427

Opinião Socialista 427

Documents
Andre Scară modulară dreaptă - magazindescari.ro · Andre detaliu de prindere balustradă Andre dimensiune treaptă și reglaj Andre gol de scară și gardă de corp etaj Andre

Andre Scară modulară dreaptă - magazindescari.ro · Andre detaliu de prindere balustradă Andre dimensiune treaptă și reglaj Andre gol de scară și gardă de corp etaj Andre

Documents