1

Inter and Intra-tumoralheterogeneity in pediatric

sarcoma

Jack Shern MDLasker Research Scholar

Pediatric Oncology BranchCenter for Cancer Research

National Cancer Institute

"It is good to have hair-splitters & lumpers.”-Charles Darwin

Rhabdomyosarcoma – “the most common soft tissue sarcoma of childhood”

Shern et al. Cancer Discovery. 4: 216-231, 2014.

Rhabdomyosarcoma: Inter-tumor heterogeneity

Can we use genetic information to further refine risk stratification?

Hawkins et al. Curr Opin Pediatr. 2014 Feb; 26(1): 50–56.

Williamson D et al. JCO 2010;28:2151-2158

ARST14B1 - Project Overview

COG histology review and clinical annotation

NCI extraction and quantification of

nucleotides and sample genotyping

RNA RNAseq – Shern Nanostring assay

Mike Arnold Nationwide ChildrensDNA

Targeted Capture and Illumina sequencing

NCI Oncogenomicspipeline to call point mutations, indels, amplifications and

deletions

Correlation between the observed genetic

alterations (RAS pathway alterations vs no RAS pathway alteration);

expression signatures and clinical outcome

2 unstained slides from

Clinically annotated RMS cases

COG ARST14B1 Summary

AKT1 CDK4 GAB1 MYCN PTEN

ALK CDKN2A HRAS MYOD1 PTPN11

ARID1A CTNNB1 IGF1R NF1 ROBO1

ATM DICER1 IGF2 NRAS SMARCA4

BCOR ERBB2 KRAS PDGFRA SOS1

BRAF FBXW7 MDM2 PIK3CA SOS2

CCND1 FGFR1 MET PKN1 TP53

CCND2 FGFR4 MTOR PTCH1

403 COG cases

Inadequate nucleotides (n=56)

347 Sequenced COG cases

66 Fusion Positive

281 Sequenced COG cases

316 Sequenced UK cases

597 Sequenced Fusion Negative

cases+

Tier I Calls - (Somatic/germline status unknown)• Hot Spot or stop or deleterious indel• High copy number amplification• Deep deletion of any included gene

Clinical Characteristics of the COG cohort

Characteristic n=347 (%)Sex

MaleFemale

234 (67)113 (33)

Age at presentation (years)MedianRange

6.40.02-37.8

Tumor HistologyAlveolarEmbryonalEmbryonal with diffuse anaplasiaEmbryonal with focal anaplasiaMixed Alveolar and EmbryonalMixed Embryonal and Spindle cellSpindle cellBotryoidRhabdomyosarcoma NOSCytology Specimen

66 (19)219 (63)

15 (4)6 (2)

2 (<1)9 (3)

16 (5)10 (3)3 (1)

1 (<1)

Anatomic GroupBladder Prostate Group 3Bladder Prostate Group 4Extremity Group 3Extremity Group 4Female GUHead and NeckOrbitalParameningeal Group 3Parameningeal Group 4ParatesticularPilot Study No DataRetroperitineum/Peritineum/Trunk Group 3Retroperitineum/Peritineum/Trunk Group 4

18 (5)6 (2)

16 (5)35 (10)

8 (2)29 (8)25 (7)

46 (13)14 (4)

64 (19)10 (3)

44 (13)30 (9)

Risk GroupLowIntermediateHigh

93 (27)131 (38)123 (35)

Variant CallsMedianRange

10-5

Mutation Summary

Identified at least one Tier 1 driver in 221/281 (80%) fusion negative cases

Percentage of cases summarized by

anatomy• TP53 pathway mutations are

common in fusion negative extremity lesions

• Female genitourinary cases account for all of the DICER1lesions

• HRAS and KRAS do not occur in orbital tumors

• MYOD1 mutations are restricted to the head

% of cases

1 tumor ≠1 genetic lesion

Hypothesis: Increased number of mutations leads to a worse outcome

Hypothesis: Fusion Negative Rhabdomyosarcoma is polyclonal?

What genes go together?

Infants less than 1 year old and the distribution of RAS isoform mutations by age

Median age: 14.4 years (3.3-21)All cases are head/neck or

parameningeal

MYOD1 mutations are associated with a dismal

outcome

0.0 2.5 5.0 7.5 10.0 12.5 15.0

EFS (Years)

0.0

0.2

0.4

0.6

0.8

1.0

Surv

ival

Pro

babi

lity

YesNoMYOD1

Event-free survival by MYOD1

0.0 2.5 5.0 7.5 10.0 12.5 15.0

EFS (Years)

0.0

0.2

0.4

0.6

0.8

1.0

Surv

ival

Pro

babi

lity

YesNoMYOD1

Log-Rank p< 0.0001

Event-free survival by MYOD1Event-free survival by MYOD1

Surv

ival

Pro

babi

lity

EFS (years)

0.0 2.5 5.0 7.5 10.0 12.5 15.0

EFS (Years)

0.0

0.2

0.4

0.6

0.8

1.0

Surv

ival

Pro

babi

lity

YesNoTP53

Event-free survival by TP53

0.0 2.5 5.0 7.5 10.0 12.5 15.0

EFS (Years)

0.0

0.2

0.4

0.6

0.8

1.0

Surv

ival

Pro

babi

lity

YesNoTP53

Log-Rank p= 0.0022

Event-free survival by TP53

0.0 2.5 5.0 7.5 10.0 12.5

EFS (Years)

0.0

0.2

0.4

0.6

0.8

1.0

Surv

ival

Pro

babi

lity

YesNoTP53

Event-free survival by TP53 (RMS Risk Intermediate only)

0.0 2.5 5.0 7.5 10.0 12.5

EFS (Years)

0.0

0.2

0.4

0.6

0.8

1.0

Surv

ival

Pro

babi

lity

YesNoTP53

Log-Rank p= 0.0231

Event-free survival by TP53 (RMS Risk Intermediate only)

Rhabdomyosarcoma cohort

TCGA cohort

TP53 mutations

Event-free survival by TP53 Event-free survival by TP53 (Intermediate Risk group only)

Surv

ival

Pro

babi

lity

Surv

ival

Pro

babi

lityEFS (years) EFS (years)

MYCN and CDK4 amplifications are poor prognostic modifiers in PAX fusion positive tumors

0 2 4 6 8 10 12

EFS (Years)

0.0

0.2

0.4

0.6

0.8

1.0

Surv

ival

Pro

babi

lity

YesNo

Event-free survival by MYCN Mutation at Enrollment, Alveolar Patients Only

0 2 4 6 8 10 12

EFS (Years)

0.0

0.2

0.4

0.6

0.8

1.0

Surv

ival

Pro

babi

lity

YesNo

p= 0.0003

Event-free survival by MYCN Mutation at Enrollment, Alveolar Patients Only

0 2 4 6 8 10 12

EFS (Years)

0.0

0.2

0.4

0.6

0.8

1.0

Surv

ival P

roba

bility

YesNo

Event-free survival by CDK4 Mutation at Enrollment, Alveolar Patients Only

0 2 4 6 8 10 12

EFS (Years)

0.0

0.2

0.4

0.6

0.8

1.0

Surv

ival P

roba

bility

YesNo

p= 0.0321

Event-free survival by CDK4 Mutation at Enrollment, Alveolar Patients OnlyEvent-free survival by MYCN amplification

Surv

ival

Pro

babi

lity

EFS (years)

Event-free survival by CDK4 amplification

Surv

ival

Pro

babi

lity

EFS (years)

Chen et al. Cancer Cell 24, 710-724, 2013

Pediatric Sarcoma: Intra-tumor Heterogeneity

Single cell sequencing

Carly Sayers Xiyuan Zhang

Skull based metastatic tumor

CD4+, CD37+, CD53+, CD74+Producing Complement and Lysozyme

G2/M – TOP2A, FOXM1, Kinesins, Centrosomal genes

Collagens, Fibrillin, IGFBP’s

BIRC3, KLF10

Atypical Neurofibroma with concern for Malignant Peripheral Nerve Sheath Tumor

Gutmann, D. H. et al. (2017)

Neoplastic Schwann cells: NRXN1, SOX10, S100B

0

1

2

3

4

5

6

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

IL34

Nor

mal

ized

Expr

essio

n

Single Cell Cluster

01234567

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

CSF1RN

orm

alize

d Ex

pres

sion

Single Cell Cluster

00.20.40.60.8

11.21.4

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

CSF1

Nor

mal

ized

Expr

essio

n

Single Cell Cluster

Gutmann, D. H. Nature Reviews. (2017) Felix, J. et al. Structure (2013)

Single cell sequencing identifies IL34 as a novel therapeutic target in NF1 tumors

HDAC inhibitors are potent repressors of PAX3-FOXO1 transcriptional activity.

Abraham, J. et al. Genes Dev (2014)

Nor

mal

ized

Expr

essio

n

Normalized Expression

Even in a “homogenous” cell culture, single cell RNAseqshows cell to cell variability

Bulk sequencing averages across the population

thereby losing information about rare cell populations

Definition of these cell populations has major

implications for our understanding of tumor

evolution and therapeutic resistance

All cells increase target gene

transcription

Cell to cell heterogeneity with

recruitment of transcription

AcknowledgmentsChildren’s Oncology Group

Corinne Linardic

Erin Rudinski

Mike Arnold

Steve Skapek

Dave Hall

Don Barkauskus

Doug Hawkins

Institute of Cancer Research UK

Janet Shipley

Anna Kelsey

Rebecca Brown

Julia Chisholm

Joanna Selfie

NCI DCEG

Kristie Jones

Belynda Hicks

NCI CCR

Javed Khan

Berkley Gryder

Young Song

Ashley Walton

Rajesh Patidar

Jun Wei

Mike Kelly

NCI, CCR, Pediatric Oncology Branch

Carly Sayers

Xiyuan Zhang

Marielle Yohe

Christine Heske

Carol Thiele

Brigitte Widemann

The Lasker Clinical Research Scholars Program at the NIHGoal – To grow the diminishing pool of talented clinical/translational researchers.

Total program duration 8 years:Years 1-5: NIH Intramural Research Program full support (salary and research support) Years 6-8: salary and/or research support of up to $500,000/year at outside institution, OR continuation in Intramural Research Program

Candidates – Early stage clinical researchers, within 10 years of completing core residency, with the ability to conduct independent research.

http://www.nih.gov/science/laskerscholar/