Data from BI-Lilly Diabetes Alliance. Date of preparation: May 2015 | UK/GLA/00036 Data from BI-Lilly Diabetes Alliance Date of preparation: May 2015 |

Data from BI-Lilly Diabetes Alliance. Date of preparation: May 2015 | UK/GLA/00036

Data from BI-Lilly Diabetes AllianceDate of preparation: May 2015 | UK/GLA/00036

ADA 2014 Data Disclosures onLY2963016 Insulin Glargine (LY IGlar)

Adverse events should be reported. Reporting forms and further information can be found at: www.mhra.gov.uk/yellowcard. Adverse events and product complaints should also be

reported to Lilly: please call Lilly UK on 01256 315 000.

Prescribing information for Abasaglar®▼ (insulin glargine injection [rDNA origin] 100 units/mL)can be found at the end of this presentation

This presentation includes reference toBoehringer Ingelheim and Lilly Diabetes Alliance products


• LY IGlar background• Phase 1 PK and PD studies

– Comparative PK and PD of LY2963016 Insulin Glargine andEU- and US-approved Versions of Lantus® Insulin Glargine inHealthy Subjects (889-P)

– Duration of Action of 2 Insulin Glargine Products, LY2963016 and Lantus®,in Subjects with Type 1 Diabetes Mellitus (891-P)

– Comparative PK and PD of 2 Insulin Glargine Products, LY2963016 and Lantus®, in Healthy Subjects at 2 Dose Levels (890-P)

• Phase 3 efficacy and safety studies– Similar Efficacy and Safety with LY2963016 Insulin Glargine Compared with

Lantus® Insulin Glargine in Patients with T1DM: The ELEMENT 1 Study (69-OR)

– Similar Efficacy and Safety with LY2963016 Insulin Glargine Compared with Lantus® Insulin Glargine in Patients with T2DM: The ELEMENT 2 Study (64-OR)

– Evaluation of Immunogenicity of LY2963016 Insulin Glargine Compared with Lantus® Insulin Glargine in Patients with T1DM or T2DM (70-OR)

Contents

PD=pharmacodynamic; PK=pharmacokinetic; T1DM=type 1 diabetes mellitus; T2DM=type 2 diabetes mellitusLantus is a registered trademark of Sanofi-Aventis


• Identical amino acid sequence to Lantus® insulin glargine (IGlar)

• Highly similar to IGlar based on principles of biosimilarity including bioequivalence1,2

• Comprehensive development programme to demonstratesimilarity

LY IGlar Development Programme“The Totality of Evidence”

PD=pharmacodynamic; PK=pharmacokinetic;T1DM=type 1 diabetes mellitus; T2DM=type 2 diabetes mellitusLantus is a registered trademark of Sanofi-Aventis

1. FDA DRAFT Guidance for Industry: Clinical Pharmacology Data to Support a Demonstration of Biosimilarity to a Reference Product; 2. EMEA, Guideline on

non-clinical and clinical development of similar biological medicinal productscontaining recombinant human insulin and insulin analogues, 2015;

3. Blevins et al. ADA 2014: 69-OR; 4. Rosenstock et al. ADA 2014: 64-OR; 5. Deeg et al. ADA 2014: 70-OR; 6. Linnebjerg et al. ADA 2014: 889-P;

7. Zhang et al. ADA 2014: 890-P; 8. Heise et al. ADA 2014: 891-P.

Phase 3studies3–5

ELEMENT 1T1DM

ELEMENT 2T2DM

Phase 1 studies6–8

Preclinical studies

Biochemical and physicochemical characterisation

PK / PD of LY IGlar versus IGlar



The LY IGlar Phase 1 Programme


• All randomized, double-blind, 2-treatment, single-dose (0.5 U/kg), 4-period, crossover, replicate euglycemic clamp studies in healthy subjects1

• Objectives were to assess the similarity of insulin exposure (PK) and insulin action (PD) between treatments*1

Core Comparative PK / PD Studies for the Support of the Phase 3 Programme

Core trials fulfill the regulatory need to demonstrate similarity between LY IGlar and the IGlar products in different regions (US and EU)

*Industry-standard bioequivalence limits (0.8–1.25 of 90% CI). Same analysis of PD using 95% CIs in the EUCI=confidence interval; PD=pharmacodynamic; PK=pharmacokinetic

1. Linnebjerg et al. ADA 2014: 889-P

LY IGlarvs.

EU-approved IGlar(Study ABEA1)

US-approved IGlar

vs. EU-approved IGlar

(Study ABEN1)

LY IGlarvs.

US-approved IGlar(Study ABEO1)


Additional PK / PD Studies

LY IGlarvs.

IGlar

(Study ABEE1)

T1DM patientsDuration of action

RandomizedDouble-blind

Single-dose (0.3 U/kg)Crossover

Euglycemic clamp

LY IGlarvs.

IGlar

(Study ABEM2)

Healthy subjects2 different doses:0.3 and 0.6 U/kg

RandomizedDouble-blindCrossover

Euglycemic clamp

Additional data regarding duration of action and bioavailability across different doses

PD=pharmacodynamic; PK=pharmacokinetic; T1DM=type 1 diabetes mellitus 1. Heise et al. ADA 2014: 891-P;2. Zhang et al. ADA 2014: 890-P


• The Phase 1 studies ABEA, ABEO, and ABEN were designed to test LY IGlar and IGlar against predefined acceptance limits that are standard in bioequivalence studies as defined by regulatory bodies1–3

• The 90% CI of the ratio of the geometric means of keyPK and PD parameters between LY IGlar and IGlar being completely contained within the interval 0.8–1.25*1–3

Criteria for Determining Similarity between LY IGlar and IGlar

*US submissionCI=confidence interval;PD=pharmacodynamic; PK=pharmacokinetic

1. Linnebjerg et al. ADA 2014: 889-P2. FDA DRAFT Guidance for Industry: Clinical Pharmacology Data to Support a Demonstration of Biosimilarity to a Reference Product, Section I, Lines 539-542

3. EMEA, Guideline on non-clinical and clinical development ofsimilar biological medicinal products containing recombinant

human insulin and insulin analogues, 2015



Comparative Pharmacokinetics and Pharmacodynamics of LY2963016 Insulin Glargine and EU- and US-approved Versions of Lantus® Insulin Glargine in Healthy Subjects

Helle Linnebjerg1, Eric Chen Quin Lam2, Mary E. Seger1*, David Coutant1,Laiyi Chua2, Chew Lan Chong2, Maria M. Ferreira3, Danny Soon2, Xin Zhang1

1Eli Lilly and Company, Indianapolis, Indiana, USA;2Lilly-NUS Centre for Clinical Pharmacology, Singapore;3PAREXEL International Bloemfontein Early Phase Unit, Bloemfontein, South Africa*Author has since retired from institution listed

Linnebjerg et al. ADA 2014: 889-P Lantus is a registered trademark of Sanofi-Aventis


• Even with identical amino acid sequences, protein-based therapeutics manufactured by distinct processes must be shown to be similar1,2

– These studies3 aimed to demonstrate similarity in the pharmacokinetics and pharmacodynamics between LY2963016 insulin glargine (LY IGlar) and EU- and US‑approved versions of Lantus® insulin glargine (EU IGlar and US IGlar)

• These were 3 Phase 1, single-site, randomized, double-blind, 2-treatment, 4-period, crossover, replicate euglycemic clamp studies

• Fasted healthy subjects received subcutaneous 0.5 U/kg doses of2 different insulin glargine products on 2 occasions each, following a randomized sequence, with a ≥7-day washout between doses

• During each period, a manual 24-hour euglycemic clamp procedure was performed

– Glucose was infused intravenously at a variable rate to maintain a target blood glucose level of 5 mg/dL (0.3 mmol/L) below each subject’s mean pre-dose fasting blood glucose

Study Rationale, Aim, and Design

1. EMA CHMP. Guideline On Similar Biological Medicinal Products 2014; 2. FDA. Guidance for Industry: Scientific Considerations in Demonstrating

Biosimilarity to a Reference Product 2012; 3. Linnebjerg et al. ADA 2014: 889-P

Lantus is a registered trademark of Sanofi-Aventis


Study Design

PD=pharmacodynamic; PK=pharmacokinetic; T=test treatment; R=reference treatmentLinnebjerg et al. ADA 2014: 889-P

ScreeningUp to 6 weeks before Period 1

dosing Period: 1 2 3 4

Follow-up5–14 days after Period 4 / early

termination

Day: −1 1 2

DischargeAdmission

Clamp*

Dose

Washout ≥7 days

= dosing and 24-hour clamp

T R T R

R T R TTreatment sequence 2:

Treatment sequence 1:

*Including collection of serial blood samples for PK and PD assessments


• Serial blood samples were taken up to 24 hours post-dose to determine concentrations of LY IGlar and/or IGlar, and C-peptide, in serum

• PK parameter estimates for LY IGlar and/or IGlar were calculated by standard non-compartmental methods. The PD parameters were derived from the euglycemic clamp procedure, where the glucose infusion rate over time was used as a measure of insulin effect

• Log-transformed PK and PD parameter estimates were analyzed using a linear mixed effects model with period, sequence, and treatment as fixed effects and subject as a random effect. A non-parametric approach was used to evaluate time of maximum observed drug concentration

• Sample sizes were planned to provide at least 90% power to demonstrate that the 90% confidence interval of the ratios of key PK or PD parameters between treatments would be contained within 0.8–1.25

Study Assessments

PD=pharmacodynamic; PK=pharmacokineticLinnebjerg et al. ADA 2014: 889-P


• For the primary analysis, serum concentrations of LY IGlar and IGlar were corrected for endogenous insulin using C-peptide data1

• Correction was warranted because:

– the studies were conducted in healthy subjects

– the assay used to detect serum LY IGlar and IGlar demonstrates cross-reactivity with endogenous insulin

Correction for Endogenous Insulin

1. Owens DR. Human insulin: clinical pharmacological studiesin normal man. New York: Springer Publishing; 1986



Results


Summary of Studies

Demographics LY IGlar vs.

EU IGlar(N = 80)

LY IGlar vs. US IGlar(N = 91)

EU IGlar vs. US IGlar(N = 40)

Sex, male / female (%) 70.0 / 30.0 93.4 / 6.6 82.5 / 17.5

Age (years)a 32.0 � 10.9 32.7 � 9.0 31.9 � 9.9

Weight (kg)a 74.8 � 12.5 70.2 � 10.3 67.8 � 9.4

BMI (kg/m2)a 24.9 � 3.2 24.1 � 2.8 23.8 � 2.4

The PK and PD were compared among the 3 insulin glargine products in 3 separate studies

LY IGlar

US IGlar

EU IGlar

= comparison

aMean ± standard deviationBMI=body mass index; N=number of subjects;PD=pharmacodynamics; PK=pharmacokinetics

Linnebjerg et al. ADA 2014: 889-P


PK Profiles of LY IGlar vs. EU IGlar*

Mean (± SD) C-peptide-corrected Serum Insulin Concentration

*Healthy subjectsPK=pharmacokinetic; SD=standard deviation



PK Profiles of LY IGlar vs. US IGlar*





• C-peptide-corrected serum insulin concentration profiles were similar for all glargine products

PK Profiles of EU IGlar vs. US IGlar*





• 90% CIs for the ratios of LS geometric means for AUC(0–24) and Cmax werecompletely contained within the prespecified bioequivalence limits (0.8–1.25)

Comparison of PK Parameters of LY IGlar and IGlar

aModel: log (parameter)=period+sequence+treatment+error, subject (random), period sequence treatment (categorical)AUC(0–24)=area under the concentration-time curve from time 0 to 24 hours; CI=confidence interval;Cmax=maximum observed drug concentration; CV%=coefficient of variation; LS=least squares;n=number of observations; N=number of subjects; PK=pharmacokinetic


Parameter (units)

Treatment N (n)Geometric

Mean (CV%)Ratio of LS Geometric

Means (90% CI)a

AUC(0–24) (pmol∙hr/L)

LY IGlar 79 (156) 1810 (40)0.91 (0.87, 0.96)

EU IGlar 80 (157) 1980 (36)

LY IGlar 87 (165) 1720 (42)0.90 (0.86, 0.94)

US IGlar 89 (167) 1900 (35)

EU IGlar 40 (75) 2000 (35)0.98 (0.91, 1.05)

US IGlar 40 (76) 2060 (39)

Cmax (pmol/L)

LY IGlar 80 (158) 112 (39)0.95 (0.90, 1.00)

EU IGlar 80 (158) 119 (34)

LY IGlar 88 (167) 103 (41)0.92 (0.87, 0.96)

US IGlar 89 (169) 111 (34)

EU IGlar 40 (76) 120 (33)0.99 (0.92, 1.06)

US IGlar 40 (77) 122 (37)


• No statistically significant difference in tmax between treatments (95% CIs for median differences contain 0; p >0.05)

Comparison of tmax of LY IGlar and IGlara

Treatment Median tmax NMedian Difference (95% CI) [p

value]

LY IGlar 12.00 80 0.00 (−0.75, 0.75)[0.82]EU IGlar 13.50 80

LY IGlar 12.00 88 0.50 (−0.76, 1.25)[0.48]US IGlar 12.00 89

EU IGlar 12.00 40 −0.75 (−1.50, 0.50)[0.28]US IGlar 12.00 40

atmax analyzed using a Wilcoxon signed-rank test and median differences using Hodges-Lehmann methodCI=confidence interval; N=number of subjects; tmax=time of maximum observed drug concentration



PD Profiles of LY IGlar vs. EU IGlar*

*Healthy subjectsPD=pharmacodynamic; SD=standard deviation


• Mean glucose infusion rate profiles were similar between LY IGlar and EU IGlar

0 4 8 12 16 20 240

1

2

3

4

5

6M

ean

Glu

cose

In

fus

ion

Ra

te (

mg

/kg

/min

)LY IGlarEU IGlar

0 4 8 12 16 20 24

2

3

4

5

6

Mea

n B

loo

dG

luco

se (

mm

ol/L

)

Time (hour)

Mean (± SD) Glucose Infusion Rate (Top) and Corresponding Blood Glucose Levels (Bottom)


PD Profiles of LY IGlar vs. US IGlar*




• Mean glucose infusion rate profiles were similar between LY IGlar and US IGlar


PD Profiles of EU IGlar vs. US IGlar*




• Mean glucose infusion rate profiles were similar between EU IGlar and US IGlar

0 4 8 12 16 20 240123456

Mea

n G

luco

se I

nfu

sio

nR

ate

(mg

/kg

/min

)

0 4 8 12 16 20 242

3

4

5

6

Time (hour)

Mea

n B

loo

dG

luco

se (

mm

ol/L

)EU IGlarUS IGlar


• 90% CIs for the ratios of LS geometric means for Gtot and Rmax were completely contained within the prespecified bioequivalence limits (0.8–1.25)

Comparison of PD Parameters of LY IGlar and IGlar

Parameter (units)

Treatment N (n)Geometric

Mean (CV%)

Ratio of LS Geometric Means

(90% CI)a

Gtot (mg/kg)

LY IGlar 80 (158) 2580 (45)0.95 (0.91, 1.00)

EU IGlar 80 (158) 2710 (40)

LY IGlar 88 (171) 1670 (60)0.91 (0.85, 0.98)

US IGlar 88 (170) 1820 (74)

EU IGlar 40 (76) 1870 (84)1.00 (0.89, 1.13)

US IGlar 40 (77) 1880 (77)

Rmax

(mg/kg/min)

LY IGlar 80 (158) 2.85 (46)0.99 (0.94, 1.04)

EU IGlar 80 (158) 2.88 (41)

LY IGlar 88 (171) 2.12 (54)0.93 (0.88, 0.98)

US IGlar 88 (170) 2.27 (58)

EU IGlar 40 (76) 2.35 (67)0.97 (0.88, 1.07)

US IGlar 40 (77) 2.44 (63)

aModel: log (parameter)=period+sequence+treatment+error, subject (random), period sequence treatment (categorical)CI=confidence interval; CV%=coefficient of variation; Gtot=total glucose infusion over the clamp duration;LS=least squares; n=number of observations; N=number of subjects; PD=pharmacodynamic;Rmax=maximum glucose infusion rate



• The most common treatment-emergent AEs reported across all studies were procedural complications and headache

• No differences in the types or incidence ofdrug-related AEs were noted between treatments

• No safety concerns were noted in the clinical laboratory, vital sign, or ECG data

Safety Results

AE=adverse event; ECG=electrocardiogramLinnebjerg et al. ADA 2014: 889-P


• The studies demonstrated that the PK(AUC[0–24] and Cmax) and PD (Rmax and Gtot) properties of LY IGlar, EU IGlar, and US IGlar were similar following subcutaneous dosing– The 90% confidence intervals for the ratios of least squares

geometric means were completely contained within the prespecified bioequivalence limits (0.8–1.25)

• No safety concerns were noted in the adverse-event, clinical laboratory, vital sign, or electrocardiogram data

Conclusions

AUC(0–24)=area under the concentration-time curve from time 0 to 24 hours;Cmax=maximum observed drug concentration; Gtot=total glucose infusion over the clamp duration; PD=pharmacodynamic; PK=pharmacokinetic; Rmax=maximum glucose infusion rate




Duration of Action of 2 Insulin Glargine Products, LY2963016 and Lantus®, in Subjects with Type 1 Diabetes MellitusTim Heise1, Xin Zhang2, Eric Chen Quin Lam3,Mary E. Seger2*, David Coutant2,Laiyi Chua3, Helle Linnebjerg2

1Profil, Neuss, Germany; 2Eli Lilly and Company, Indianapolis, Indiana, USA;3Lilly-NUS Centre for Clinical Pharmacology, Singapore*Author has since retired from institution listed

Heise et al. ADA 2014: 891-PLantus is a registered trademark of Sanofi-Aventis


• This was a Phase 1, single-site, randomized, double-blind, single-dose,2-period, 2-sequence, crossover euglycemic clamp study

• Fasted male subjects with T1DM received single subcutaneous dosesof 0.3 U/kg LY2963016 insulin glargine (LY IGlar) and Lantus® insulin glargine (IGlar), with a ≥7-day washout between doses

• This study compared the duration of action (time post-dosing at which a subject’s blood glucose was >150 mg/dL [8.3 mmol/L] without any glucose infusion) of 0.3 U/kg LY IGlar and IGlar in subjects with T1DM during a euglycemic clamp procedure

• During each period, a euglycemic clamp procedure, lasting up to 42 hours post-dose, was performed. Blood samples were taken to determine:

– duration of action and pharmacodynamic parameters for LY IGlar and IGlar

– serum concentrations of LY IGlar, IGlar, and insulin lispro for pharmacokinetic evaluation; analyzed using a validated radioimmunoassay method

Study Design

T1DM=type 1 diabetes mellitusLantus is a registered trademark of Sanofi-Aventis

Heise et al. ADA 2014: 891-P


Washoutpre-

study insulin

Run-in

period

Euglycem

ic clam

p

Resume pre-

study insulin

Euglycemic Clamp Procedure

IV=intravenous; NPH=Neutral Protamine Hagedorn; SC=subcutaneous 1. Heise et al. ADA 2014: 891-P;2. Lepore M et al. Diabetes 2000;49:2142–2148

LY IGlar / IGlar SC (0.3 U/kg) (0 hours)

• Begin variable IV infusion of insulin lispro or glucose at −1 to −6 hours (automated clamp using a Biostator)

• Target blood glucose level: 100 mg/dL (5.6 mmol/L); maintained for ≥1 hours before dosing

• From −1 hour, insulin lispro infusion rate (if any) minimized while ensuring no glucose infused

−48 hours: start switching subjects on insulin detemir or IGlar to NPH insulin (tailored washout regimen)

By −22 hours: last dose of intermediate-acting insulin administered

−12 to −9 hours: total dose of short- or rapid-acting insulin to be <8 units

By −9 hours: last dose of short- or rapid-acting insulin

−4 hours: stop basal rate of any continuous SC insulin infusion; washout complete forall subjects

• Variable IV glucose infusionto maintain target glucoselevel for ≤42 hours

• Insulin lispro infusion terminated when LY IGlar / IGlar effect is seen (defined as a drop in blood glucose of approximately 5 mg/dL[0.3 mmol/L])

• Clamp ended at 42 hourspost-dose, unless blood glucose reaches 250 mg/dL(13.8 mmol/L) before this time

• End of action = time post-dosing at which a subject’s blood glucose level is >150 mg/dL[8.3 mmol/L] without any glucose infusion2


• Estimates of duration of action were compared for LY IGlar and IGlar using a linear mixed effects model with treatment, period, and sequence as fixed effects and subject as a random effect1

• As 14 / 40 clamps were terminated at 42 hours, before end of action (EoA) was reached, a time-to-event (survival) analysis that allows for censored observations was applied

– Each EoA observation was considered an “event”, and “survival” was defined as EoA not yet being reached

– Duration of action was “censored” (i.e. not recorded) when a subject did not reach EoA before clamp termination

– Estimated time-to-event (survival) curves for LY IGlar and IGlar were plotted and compared using the log-rank test of equality

• Cox proportional hazard model2 was used to estimate the hazard ratio

Estimating Duration of Action

1. Heise et al. ADA 2014: 891-P; 2. Cox DR. J R Stat Soc Series B 1972;20:187–220


Subject Demographics

*All of the 20 subjects who entered the study completed the study as plannedBMI=body mass index; HbA1c=glycosylated haemoglobin; N=number of subjects; SD=standard deviation


ParameterMean � SD (N = 20)*

Age (years) 41.5 � 9.1

Sex, male (%) 100

Race, white (%) 100

Weight (kg) 84.1 � 9.8

BMI (kg/m2) 25.6 � 2.4

HbA1c (%) 7.99 � 0.62

Duration of diabetes (years) 18.9 � 9.8



Results


Mean (± SD) Glucose Infusion Rate and Corresponding Blood Glucose Levels

LY IGlar (0.3 U/kg)IGlar (0.3 U/kg)

0 8 16 24 32 40 480.0

0.5

1.0

1.5

2.0

Me

an

Glu

co

se

In

fus

ion

Ra

te (

mg

/kg

/min

)

0 8 16 24 32 40 480

6

912

15

Time (hour)

Blo

od

Glu

co

se

(mm

ol/

L)

Time (hour)

Smoothing factor ranged

from 0.075 to 0.2

3

Heise et al. ADA 2014: 891-PSD=standard deviation


• Survival curves for LY IGlar and IGlar were similar (log-rank test of equality, p = 0.859)

• Cox proportional hazard estimate (LY IGlar / IGlar) for duration of actionwas 1.063; p = 0.8777

• Results demonstrate similar duration of action for LY IGlar and IGlar

Time-to-event (Survival) Plot of Duration of Action for LY IGlar and IGlar

Test of equality of survival curves (based on log-rank test): Chi-square statistic = 0.031, p = 0.859Heise et al. ADA 2014: 891-P


Comparison of Duration of Action of LY IGlar and IGlar Based on Survival Analysis (All Subjects)

aDuration of action was “censored” (i.e. not recorded) when a subject did not reach EoA before clamp termination; bMaximum and mean are based on subjects who reached EoA before 42 hours (i.e. not censored); cThe Xth percentile of the survival time (duration of action) is the time beyond which (100−X)% of the subjects are expected to “survive” (i.e. EoA has not been reached); dDue to censoringCI=confidence interval; EoA=end of action; NA=not applicable; SE=standard error


Summary Statistics IGlar (0.3 U/kg) LY IGlar (0.3 U/kg)

Number (%) of events 13 (65.0) 13 (65.0)

Number (%) censoreda 7 (35.0) 7 (35.0)

Duration of action (hour)

Range 2.0–41.5b 2.8–40.5b

25th percentilec (95% CI) 19.50 (12.23, 39.50) 19.75 (7.00, 37.00)

Median (95% CI) 40.00 (20.00, NAd) 37.13 (20.00, NAd)

Meanb (SE) 25.54 (3.91) 23.78 (3.75)


• No statistically significant difference in Gtot and Rmax between LY IGlar and IGlar, with the 90% CIs for the ratios of LS geometric means containing 1

Comparison of Gtot and Rmax of LY IGlar and IGlar

Statistical model: log (parameter)=period+sequence+treatment+error, subject (random), period sequence treatment (categorical)CI=confidence interval; CV%=coefficient of variation; Gtot=total glucose infusion over the clamp duration; LS=least squares; N=number of subjects; Rmax=maximum glucose infusion rate


Parameter (units)

Treatment (0.3 U/kg)

NGeometric

Mean (CV%)


(90% CI)

Gtot

(mg/kg)

LY IGlar 20 4.60 (1090)0.77 (0.46, 1.30)

IGlar 19 6.52 (1160)

Rmax

(mg/kg/min)

LY IGlar 20 0.530 (254)0.91 (0.52, 1.61)

IGlar 19 0.611 (310)


• Serum insulin concentration profiles were similar for LY IGlar and IGlar

Mean (± SD) Serum Insulin Concentrationwith LY IGlar and IGlar

11 subjects did not have analyzable insulin concentration data for both study periods, mostly due to concentrations being below the lower limit of quantification. Data available from 17 subjects were used to generate the mean profile. The pharmacokinetic assay detects endogenous insulin as well as IGlar / LY IGlar; however, due to the limited quantifiable concentrations available, the concentrations were not corrected for endogenous insulinSD=standard deviation


Time (hour)

Ser

um

In

sulin

(p

mo

l/L)

20

40

60

80

100

120

0 6 12 18 24 30 36 42

LY IGlar (0.3 U/kg)IGlar (0.3 U/kg)


Adverse Events and Tolerability with LY IGlar and IGlar

Frequency of Treatment-emergent Adverse Events (All Causalities)

aMedDRA version 14.1; bAEs with a change in severity are counted only onceAE=adverse event; ECG=electrocardiogram; MedDRA=Medical Dictionary for Regulatory Activities;N=number of subjects


MedDRA Preferred Terma

Number of AEsb

(Number of Subjects with AEs)

0.3 U/kg IGlar(N = 20)

0.3 U/kg LY IGlar(N = 20)

Headache 2 (2) 1 (1)

Back pain 1 (1) 1 (1)

Abdominal pain upper 1 (1)

Arthralgia 1 (1)

Dizziness 1 (1)

Nausea 1 (1)

Total 3 (3) 6 (4)


• LY IGlar (0.3 U/kg) was well tolerated, with a similar AE profile observed to that following IGlar dosing (0.3 U/kg)

• No safety concerns noted in the AE, clinical laboratory, vital sign, or ECG data

• No drug-related AEs reported

Safety Results

AE=adverse event; ECG=electrocardiogramHeise et al. ADA 2014: 891-P


• Duration of action was similar for both treatments

• No statistically significant difference in Rmax and Gtot was observed between LY IGlar and IGlar during a 42-hour glucose clamp

• Pharmacokinetic profiles appeared to be similar betweenLY IGlar and IGlar

• Single subcutaneous doses of 0.3 U/kg LY IGlar and IGlar werewell tolerated in subjects with T1DM

Conclusions

In subjects with T1DM, following a single 0.3 U/kgdose of LY IGlar and IGlar:

Gtot=total glucose infusion over the clamp duration;Rmax=maximum glucose infusion rate; T1DM=type 1 diabetes mellitus




Comparative Pharmacokinetics and Pharmacodynamics of 2 Insulin Glargine Products, LY2963016 and Lantus®, in Healthy Subjects at 2 Dose Levels

Xin Zhang1, Eric Chen Quin Lam2, Mary E. Seger1*, David Coutant1, Laiyi Chua2, Lai Hock Tan2, Danny Soon2, Helle Linnebjerg1

1Eli Lilly and Company, Indianapolis, Indiana, USA; 2Lilly-NUS Centre for Clinical Pharmacology, Singapore*Author has since retired from institution listed

Zhang et al. ADA 2014: 890-PLantus is a registered trademark of Sanofi-Aventis


• This was a Phase 1, single-site, randomized, subject- and investigator-blinded, 4-treatment, 4-period, crossover study to evaluate the pharmacokinetics and pharmacodynamics ofLY2963016 insulin glargine (LY IGlar) and Lantus® insulin glargine (IGlar) at 2 dose levels (0.3 and 0.6 U/kg)1

• The study was carried out to supplement the results of a larger, core study at the 0.5 U/kg dose level2

• Fasted healthy subjects were randomly assigned to a dosing sequence, with ≥6-day washout between doses1

• During each period, a manual 24-hour euglycemic clamp procedure was performed– Glucose was infused intravenously at a variable rate to maintain a target

blood glucose level of 5 mg/dL (0.3 mmol/L) below the mean pre-dose fasting blood glucose for each subject

Study Design and Rationale

1. Zhang et al. ADA 2014: 890-P; 2. Linnebjerg et al. ADA 2014: 889-P

Lantus is a registered trademark of Sanofi-Aventis


• PK parameter estimates for LY IGlar and IGlar were calculated by standard non-compartmental methods. The PD parameters were derived from the euglycemic clamp procedure, where the glucose infusion rate over time was used as a measure of insulin effect

• Log-transformed PK and PD parameter estimates were analyzed using a linear mixed effects model with period, sequence, and treatment as fixed effects and subject as a random effect. A non-parametric approach was used to evaluate time of maximum observed drug concentration

Statistical Analyses

PD=pharmacodynamic; PK=pharmacokineticZhang et al. ADA 2014: 890-P


• For the primary analysis, serum concentrations of LY IGlar and IGlar were corrected for endogenous insulin usingC-peptide data using the following equation1:

[LY IGlar or IGlar] = [immunoreactive LY IGlar or IGlar] – F*[C-peptide]

where F is the average of the ratios of immunoreactiveLY IGlar or IGlar to C-peptide at baseline (−30 and 0 minutes)

• Correction was warranted because:

– the study was conducted in healthy subjects

– the assay used to detect serum LY IGlar and IGlar demonstrates cross-reactivity with endogenous insulin

Correction for Endogenous Insulin

1. Owens DR. Human insulin: clinical pharmacological studies in normal man. New York: Springer

Publishing; 1986


• Serial blood samples were taken pre-dose and up to 24 hourspost-dose to determine serum LY IGlar, IGlar, and C-peptide concentrations

Dosing and Assessments

Dosing Schedule (U/kg)

Zhang et al. ADA 2014: 890-P

Sequence Period 1 Period 2 Period 3 Period 4

1 LY IGlar 0.3 LY IGlar 0.6 IGlar 0.3 IGlar 0.6

2 LY IGlar 0.6 IGlar 0.6 LY IGlar 0.3 IGlar 0.3

3 IGlar 0.3 LY IGlar 0.3 IGlar 0.6 LY IGlar 0.6

4 IGlar 0.6 IGlar 0.3 LY IGlar 0.6 LY IGlar 0.3



Results


Subject Demographics

ParameterMean � SD (N = 24a)

Age (years) 32.1 � 7.7

Sex, male / female (%) 83.3 / 16.7

Race, Asian (%) 100

Weight (kg) 66.4 � 9.2

BMI (kg/m2) 22.7 � 2.8

a1 subject was withdrawn after dosing in Period 2 (subject decision); 23 subjects completed the studyBMI=body mass index; N=number of subjects studied; SD=standard deviation



• C-peptide-corrected serum insulin concentration profiles were similar for LY IGlarand IGlar at each dose level

PK of LY IGlar and IGlar with different doses*

*Healthy subjectsPK=pharmacokinetics; SD=standard deviation



LY IGlar 0.3 U/kgLY IGlar 0.6 U/kgIGlar 0.3 U/kgIGlar 0.6 U/kg

Time (hour)

0

100

200

300

400

0 6 12 18 24

Mea

n C

-pep

tid

e-co

rrec

ted

Insu

lin C

on

cen

trat

ion

(p

mo

l/L)


Mean (± SD) C-peptide Concentration

• C-peptide concentration profiles are similar at each dose between LY IGlar and IGlar, suggesting a similar degree of suppression of endogenous insulin

Mean C-peptide Concentration with Different Doses of LY IGlar and IGlar*

LY IGlar 0.3 U/kg

Time (hour)

Mea

n C

-pep

tid

eC

on

cen

trat

ion

(p

mo

l/L)

0

200

400

600

0 6 12 18 24

LY IGlar 0.6 U/kgIGlar 0.3 U/kgIGlar 0.6 U/kg

*Healthy subjectsSD=standard deviation



• No statistically significant differences in AUCs and Cmax between LY IGlar and IGlar, with the 90% CIs for the ratios of LS geometric means containing 1

• AUCs and Cmax were also consistent across dose levels

Comparison of PK Parameters with Different Doses of LY IGlar and IGlar*

Parameter (units)

Dose (U/kg)

Treatment NGeometric

Mean (CV%)


(90% CI)

AUC(0–24) (pmol∙hr/L)

0.3LY IGlar 23 1730 (20)

1.03 (0.91, 1.16)IGlar 23 1690 (30)

0.6LY IGlar 24 3160 (27)

1.07 (0.95, 1.21)IGlar 24 2940 (45)

AUC(0–∞)

(pmol∙hr/L)

0.3LY IGlar 23 2330 (39)

0.97 (0.83, 1.12)IGlar 22 2390 (33)

0.6LY IGlar 24 4470 (15)

1.04 (0.90, 1.20)IGlar 24 4310 (51)

Cmax

(pmol/L)

0.3LY IGlar 23 108 (20)

1.03 (0.92, 1.15)IGlar 23 105 (33)

0.6LY IGlar 24 180 (28)

1.03 (0.92, 1.16)IGlar 24 174 (38)

*Healthy subjectsAUC(0–24)=area under the concentration-time curve from time 0 to 24 hours; AUC(0–∞)=AUC from 0 to infinity; CI=confidence interval; Cmax=maximum observed drug concentration; CV%=coefficient of variation;LS=least squares; N=number of subjects; PK=pharmacokinetic



Comparison of tmax with Different Doses ofLY IGlar and IGlar*

• No statistically significant difference in tmax between LY IGlar and IGlar, with the 90% CIs for the median differences containing 0

Dose (U/kg)

Median tmaxMedian Difference

(Approximate 90% CI)

LY IGlar – IGlarIGlar (N) LY IGlar (N)

0.3 9.00 (23) 9.00 (23) 0.00 (−3.00, 3.00)

0.6 10.50 (24) 12.00 (24) 2.00 (0.00, 3.00)

*Healthy subjectsCI=confidence interval; N=number of subjects; tmax=time of maximum observed drug concentration



PD of LY IGlar and IGlar with 0.3 U/kg dosing*

LY IGlar 0.3 U/kgIGlar 0.3 U/kg

0 4 8 12 16 20 24

2

3

4

5

6

Time (hour)

Mea

n B

loo

dG

luco

se (

mm

ol/L

)

0 4 8 12 16 20 24

0

2

4

6M

ean

Glu

cose

In

fusi

on

Rat

e (m

g/k

g/m

in)

Time (hour)

Mean (± SD) Glucose Infusion Rate (Top) andCorresponding Glucose Levels (Bottom)

*Healthy subjectsPD=pharmacodynamics; SD=standard deviation



• Mean glucose infusion rate profiles were similar between treatments at each dose level. The clamps were well conducted, as evidenced by the relatively flat mean blood glucose profiles at each dose level

PD of LY IGlar and IGlar with 0.6 U/kg dosing*Mean (± SD) Glucose Infusion Rate (Top) andCorresponding Glucose Levels (Bottom)

*Healthy subjectsPD=pharmacodynamics; SD=standard deviation


0 4 8 12 16 20 242

3

4

5

6

Time (hour)

0 4 8 12 16 20 240

2

4

6

Time (hour)Mea

n G

luco

se I

nfu

sio

nR

ate

(mg

/kg

/min

)M

ean

Blo

od

Glu

cose

(m

mo

l/L)

LY IGlar 0.6 U/kgIGlar 0.6 U/kg


• The PD properties (Gtot and Rmax) were not statistically significantly different between LY IGlar and IGlar, with the 90% CIs for the ratios of geometric means containing 1, and were consistent across dose levels

Comparison of PD Parameters with Different Doses of LY IGlar and IGlar*

*Healthy subjectsCI=confidence interval; CV%=coefficient of variation; Gtot=total glucose infusion over the clamp duration;LS=least squares; N=number of subjects; PD=pharmacodynamic; Rmax=maximum glucose infusion rate


Parameter (units)

Dose (U/kg)

Treatment NGeometric

Mean (CV%)


(90% CI)

Gtot

(mg/kg)

0.3LY IGlar 23 1060 (178)

0.98 (0.78, 1.24)IGlar 23 1050 (130)

0.6LY IGlar 24 2260 (80)

0.87 (0.70, 1.09)IGlar 24 2590 (62)

Rmax

(mg/kg/min)

0.3LY IGlar 23 1.81 (100)

1.04 (0.87, 1.25)IGlar 23 1.70 (92)

0.6LY IGlar 24 3.05 (59)

0.94 (0.79, 1.12)IGlar 24 3.25 (54)


Drug-related, Treatment-emergent Adverse Events with Different Doses of LY IGlar and IGlar*

MedDRA Preferred Terma

Number of AEsb (Number of Subjects with AEs)

0.3 U/kg IGlar

(N = 23)

0.3 U/kg LY IGlar

(N = 23)

0.6 U/kg IGlar

(N = 24)

0.6 U/kgLY IGlar(N = 24)

Vomiting 2 (1) 3 (2)

Dizziness 1 (1) 1 (1)

Headache 1 (1) 1 (1)

Hyperhidrosis 1 (1) 1 (1)

Abdominal discomfort 1 (1)

Head discomfort 1 (1)

Injection site erythema 1 (1)

Injection site pruritus 1 (1)

Nausea 1 (1)

Total 4 (1) 0 4 (3) 8 (4)

*Healthy subjectsaMedDRA version 15.0; bAEs with a change in severity are counted only onceAE=adverse event; MedDRA=Medical Dictionary for Regulatory Activities; N=number of subjects


Frequency of Drug-related, Treatment-emergent Adverse Events


• No safety concerns were noted in the AE, clinical laboratory, vital sign, or ECG data

• No episodes of hypoglycaemia were recorded• The types of drug-related AEs reported were similar

between treatments

Safety Summary with Different Doses ofLY IGlar and IGlar*

*Healthy subjectsAE=adverse event; ECG=electrocardiogram



• There were no statistically significant differences in PK (AUCs and Cmax) or PD (Rmax and Gtot) properties of LY IGlar and IGlar in healthy subjects following single subcutaneous doses of 0.3 and 0.6 U/kg

• Single subcutaneous doses of 0.3 and 0.6 U/kgLY IGlar and IGlar were well tolerated in healthy subjects

Conclusions

AUC=area under the curve; Cmax=maximum observed drug concentration; Gtot=total glucose infusionover the clamp duration; PD=pharmacodynamic; PK=pharmacokinetic; Rmax=maximum glucose infusion rate




The LY IGlar Phase 3 Programme



Similar Efficacy and Safety with LY2963016 Insulin Glargine Compared with Lantus® Insulin Glargine in Patients with T1DM: The ELEMENT 1 Study

Thomas Blevins1, Dominik Dahl2, Julio Rosenstock3, Liza L. Ilag4, William J. Huster4, Robyn K. Pollom4, Melvin J. Prince4

for the ELEMENT 1 Study Group1Texas Diabetes & Endocrinology, Texas, USA;2Gemeinschaftspraxis für Innere Medizin und Diabetologie, Hamburg, Germany; 3Dallas Diabetes and Endocrine Center at Medical City, Texas, USA; 4Eli Lilly and Company, Indiana, USA

Blevins et al. ADA 2014: 69-ORLantus is a registered trademark of Sanofi-Aventis


Phase 3 Open-label Study in T1DM(ELEMENT 1): Design

Blevins et al. ADA 2014: 69-OR

a1 patient (LY IGlar) discontinued before receiving study drug BMI=body mass index; HbA1c=glycosylated haemoglobin; IGlar=Lantus® insulin glargine; LY IGlar=LY2963016 insulin glargine;NPH=Neutral Protamine Hagedorn; QD=once daily; SC=subcutaneous; T1DM; type 1 diabetes mellitus; TID=thrice daily

• ≥18 years• T1DM ≥1 years• HbA1c ≤11%• BMI ≤35 kg/m2

• Basal-bolus with QD NPH, IGlar or detemir

Study Criteria

Week

Screening

−2 (±1) 0 6 12 24 52 56

Randomization N = 536a

Treatment Period

Extension Period Follow-upTitration Period

Primary Endpoint

at 24 weeks

IGlar QD SC + TID Insulin Lispro

LY IGlar QD SC + TID Insulin Lispro

N = 267 (IGlar)N = 268 (LY IGlar)


ELEMENT 1: Objectives

• Primary– To demonstrate non-inferiority of LY IGlar to IGlar on change in HbA1c from

baseline to 24 weeks in T1DM (the non-inferiority margin was 0.4% and, if met, the upper limit of the 95% CI was compared with the 0.3% non-inferiority margin)

• Key secondary– Non-inferiority of IGlar to LY IGlar

– Change in HbA1c at 6, 12, 36, and 52 weeks

– 7-point SMBG profiles

– Proportion of patients with HbA1c <7% or ≤6.5%

– Change in body weight

– Insulin dose

– Hypoglycaemia

– Adverse events

– Incidence of anti-insulin antibodies


CI=confidence interval; HbA1c=glycosylated haemoglobin;SMBG=self-monitored blood glucose; T1DM; type 1 diabetes mellitus


ELEMENT 1: Baseline Patient Characteristics

DemographicsIGlar

N = 267a

LY IGlarN = 268a

p value

Age (years) 41 ± 13 41 ± 14 0.72

Sex, male (%) 58 58 >0.99

Race, Caucasian / Asian / other (%) 75 / 19 / 6 74 / 18 / 8 0.29

Weight (kg) 75 ± 15 76 ± 17 0.46

BMI (kg/m2) 25 ± 4 26 ± 4 0.50

HbA1c (%) 7.8 ± 1.0 7.8 ± 1.1 0.72

FBG by SMBG (mmol/L) 8.2 ± 3.0 8.4 ± 3.0 0.49

Basal insulin, IGlar / other (%) 88 / 12 81 / 19 0.06

Diabetes duration (years) 17 ± 11 16 ± 11 0.73


Data are mean ± standard deviation unless otherwise indicatedaFull analysis set, N numbers reflect maximum sample sizeBMI=body mass index; FBG=fasting blood glucose;HbA1c=glycosylated haemoglobin; SMBG=self-monitored blood glucose


ELEMENT 1: HbA1c Change from Baseline and over Time with LY IGlar and IGlar


Data are least squares mean ± standard error*p = 0.03; no significant differences between treatment at any other time pointCI=confidence interval; HbA1c=glycosylated haemoglobin; LOCF=last observation carried forward

24 Weeks LOCF

∆ = 0.10895% CI (-0.002, 0.219)

p=0.055

∆ = 0.02095% CI (-0.099, 0.140)

p=0.737

52 Weeks LOCF

Ch

ang

e i

n H

bA

1c (

%)

IGlar (N=268)LY IGlar (N=267)

0 6 12 24 36 52

*


ELEMENT 1: Other Glucose Measures


BG values are least squares mean ± standard erroraNumbers reflect maximal sample size; bFrom 7-point self-monitored blood glucoseBG=blood glucose; FBG=fasting blood glucose;HbA1c=glycosylated haemoglobin; LOCF=last observation carried forward

IGlarN = 267a

LY IGlarN = 268a

p value

FBG(mmol/L)b

Baseline 8.2 ± 0.2 8.4 ± 0.2 0.49

24 weeks LOCF 7.8 ± 0.2 8.0 ± 0.2 0.40

52 weeks LOCF 8.3 ± 0.2 8.0 ± 0.2 0.23

Daily mean BG(mmol/L)b

Baseline 8.7 ± 0.1 8.7 ± 0.1 0.76

24 weeks LOCF 8.3 ± 0.1 8.3 ± 0.1 0.95

52 weeks LOCF 8.5 ± 0.1 8.3 ± 0.1 0.13

Patients with HbA1c

<7% (%)

Baseline 20 29 0.02

24 weeks LOCF 32 35 0.65

52 weeks LOCF 25 30 0.21


ELEMENT 1: Daily Insulin Dose and Body Weight

IGlarN = 267a

LY IGlarN = 268a p value

Total daily insulin dose (U/kg) Baseline 0.71 ± 0.02 0.72 ± 0.02 0.63

∆ at 24 weeks LOCF 0.00 ± 0.02 0.01 ± 0.02 0.70

∆ at 52 weeks LOCF 0.03 ± 0.02 0.03 ± 0.02 0.79

Body weight (kg) Baseline 74.8 ± 1.0 75.8 ± 1.0 0.46

∆ at 24 weeks LOCF 0.1 ± 0.2 0.4 ± 0.2 0.32

∆ at 52 weeks LOCF 0.4 ± 0.3 0.7 ± 0.3 0.25

Blevins et al. ADA 2014: 69-ORData are least squares mean ± standard error. LOCF=last observation carried forward

60

70

80

90

Da

ily In

su

lin D

os

e (U

/kg

)

0

0.2

0.4

0.6

0.8

Baseline 24 Weeks 52 Weeks

IGlar prandialIGlar basal

LY IGlar prandialLY IGlar basal

Bo

dy

We

igh

t (k

g)

LY IGlar (N = 268)IGlar (N = 267)

Baseline 24 Weeks 52 Weeks


ELEMENT 1: Total, Nocturnal, and Severe hypoglycaemia

Blevins et al. ADA 2014: 69-ORAll p values >0.05

97

88

4

96

86

40

20

40

60

80

100

Total Nocturnal Severe

Pa

tie

nts

(%)

80

17

77

160

20

40

60

80

100

120

140

160


Eve

nts

/Pa

tie

nts

/Yea

r, M

ea

n (S

D)

Incidence Rate

IGlar (N = 268)LY IGlar (N = 267)

<1 <1


ELEMENT 1: Summary of Adverse Events

AEsa IGlar (N = 267)

LY IGlar(N = 268)

Deaths 1 (0.4) 0

Serious AEs 24 (9) 20 (8)

Discontinuations due to an AE 6 (2) 2 (0.7)

Injection site AEs 3 (1) 7 (3)

TEAEs 166 (62) 167 (62)

Possibly related to study drug 14 (5) 17 (6)

Possibly related to study procedure 2 (0.7) 2 (0.7)

Possibly related to study disease state (diabetes) 16 (6) 21 (8)

Special topic assessment of AEs (allergic) 11 (4) 20 (8)


Data are n (%)All p values >0.05aPatients may be counted in >1 categoryAE=adverse event; TEAE=treatment-emergent adverse event


AEsa

IGlar (N = 267)

LY IGlar(N = 268)

Special topic assessment of allergic events 11 (4.1) 20 (7.5)

Pruritus, rash, dermatitis, otherb 4 (1.5) 7 (2.6)

Arthralgia, arthritis 5 (1.9) 4 (1.5)

Injection site (reaction, induration, nodule, swelling) 2 (0.7) 6(2.2)

Hypersensitivity 1 (0.4) 1 (0.4)

Allergic respiratory symptom, asthma 0 2 (0.7)

Injection site reaction (patient questionnaires) 3 (1.1) 7 (2.6)

Pain 2 (0.7) 6 (2.2)

Pruritus 1 (0.4) 2 (0.7)

Rash 1 (0.4) 2 (0.7)

ELEMENT 1: Summary of Allergic Events and Injection Site Reactions


Data are n (%) for patients with ≥1 treatment-emergent AEAll p values >0.05aPatients may be counted in >1 category;bPhotosensitivity reaction, urticariaAE=adverse event


IGlar LY IGlar p value

Patients with TEAR, n (%)

Week 52 (LOCF) 12 (5) 18 (7) 0.27

Overall 24 weeks 17 (6) 25 (9) 0.20

Overall 52 weeks 25 (9) 29 (11) 0.57

0

10

20

30

40

Pa

tie

nts

wit

h T

EA

R (%

)

0 6 12 24 52

Week

TEAR by weekLY IGlar (N = 268)IGlar (N = 267)

TEAR criteria

If antibody was not detected at baseline % antibody binding ≥1.26%

If antibody was detected at baselineAbsolute increase in % antibody binding of 1%

AND 30% relative increase from baseline

ELEMENT 1: Incidence of Treatment-emergent Antibody Response (TEAR)

Blevins et al. ADA 2014: 69-ORLOCF=last observation carried forward


• LY IGlar compared with IGlar at 24 weeks and 52 weeks demonstrated similar:– glucose-lowering effect (HbA1c, FBG, mean BG)

– insulin doses (basal, prandial, total)

– changes in body weight

– hypoglycaemia incidence and rate

– adverse-event profile

– allergic and injection site reactions

– incidence of treatment-emergent antibody response

Summary of Results from ELEMENT 1

Blevins et al. ADA 2014: 69-ORBG=blood glucose; FBG=fasting blood glucose; HbA1c=glycosylated haemoglobin


• LY IGlar compared with IGlar, used in combination with insulin lispro, provided equivalent efficacy and a similar safety profile, with no clinically meaningful differences in patients with type 1 diabetes mellitus

ELEMENT 1: Conclusions




Similar Efficacy and Safety with LY2963016 Insulin Glargine Compared with Lantus® Insulin Glargine in Patients with T2DM: The ELEMENT 2 Study

Julio Rosenstock1, Priscilla Hollander2,Anuj Bhargava3, Liza Ilag4, Robyn K. Pollom4,William J. Huster4, Melvin Prince4 for the ELEMENT 2 Study Group1Dallas Diabetes and Endocrine Center at Medical City, Texas, USA; 2Baylor Endocrine Center, Texas, USA; 3Iowa Diabetes andEndocrinology Research Center, Iowa, USA; 4Eli Lilly and Company, Indiana, USA

Rosenstock et al. ADA 2014: 64-OR Lantus is a registered trademark of Sanofi-Aventis


ELEMENT 2: Study Design

Rosenstock et al. ADA 2014: 64-OR

a3 patients (LY IGlar) discontinued before receiving study drugBMI=body mass index; FBG=fasting blood glucose; HbA1c=glycosylated haemoglobin;IGlar=Lantus® insulin glargine; LY IGlar=LY2963016 insulin glargine;OAD=oral antidiabetic drug; QD=once daily; SC=subcutaneous; T2DM; type 2 diabetes mellitus

• ≥18 years• T2DM• ≥2 OADs ± IGlar• HbA1c ≥7% and ≤11%

if insulin-naïve• HbA1c ≤11% if

previously on IGlar• BMI ≤45 kg/m2

Study Criteria Patient-driven titration(1U/d until FBG ≤5.6 mmol/L)

Screening

−2 0 12 24 28Week

N = 380 (IGlar)N = 376 (LY IGlar)

Treatment Period

Maintenance Period Follow-upTitration Period

Primary endpoint

at 24 weeks

Randomization (Total N = 759a)

LY IGlar QD SC + OADs

IGlar QD SC + OADs

Phase 3double-blind


• Primary – To demonstrate the non-inferiority of LY IGlar to IGlar as measured by change in HbA1c from

baseline to 24 weeks, when used in combination with OADs (the non-inferiority margin was 0.4% and, if met, the upper limit of the 95% confidence interval was compared with the 0.3% non-inferiority margin)

• Key secondary– Non-inferiority of IGlar to LY IGlar

– Change in HbA1c over time

– 7-point SMBG

– Fasting blood glucose changes over time (by SMBG)

– Proportion of patients with HbA1c <7% or ≤6.5%

– Change in body weight

– Insulin dose

– Hypoglycaemia

– Adverse events

– Incidence of anti-insulin antibodies

ELEMENT 2: Objectives


HbA1c=glycosylated haemoglobin; OAD=oral antidiabetic drug;SMBG=self-monitored blood glucose


ELEMENT 2: Patient Disposition

Patients screenedN = 1026

Patients randomizedN = 759

Screen failuresN = 267

Patients discontinued before receiving study drug

N = 3 (LY IGlar)

LY IGlarN = 376

IGlarN = 380

Completed 24-weektreatment periodN = 334 (88.8%)

Completed 24-weektreatment periodN = 328 (86.3%)

DiscontinuedAdverse event = 5 (1.3%)

Death = 1 (0.3%)Lack of efficacy = 1 (0.3%)Lost to follow-up = 7 (1.9%)

Physician decision = 9 (2.4%)Protocol violation = 8 (2.1%)Patient decision = 11 (2.9%)

DiscontinuedAdverse event = 10 (2.6%)

Death = 1 (0.3%)Lack of efficacy = 2 (0.5%)Lost to follow-up = 9 (2.4%)

Physician decision = 9 (2.4%)Protocol violation = 5 (1.3%)Patient decision = 16 (4.2%)



ELEMENT 2: Baseline Demographics and Patient Characteristics

Total Insulin-naïve Prior IGlar

DemographicsIGlar

N = 380a

LY IGlarN = 376a

IGlarN = 236a

LY IGlarN = 221a

IGlarN = 144a

LY IGlarN = 155a

Age (years) 59 ± 10 59 ± 10 58 ± 10 58 ± 11 60 ± 10 60 ± 9

<65 years, n (%) 278 (73) 264 (70) 182 (77) 158 (72) 96 (67) 106 (68)

Sex, male, n (%) 199 (52) 179 (48) 131 (56) 106 (48) 68 (47) 73 (47)

Race, Caucasian / Asian / black / other (%)

77 / 9 /8 / 6

80 / 8 /7 / 5

79 / 7 /9 / 4

77 / 9 /9 / 6

72 / 13 / 7 / 8

85 / 7 /4 / 5*

Weight (kg) 90 ± 19 90 ± 20 91 ± 19 89 ± 20 88 ± 20 92 ± 20

BMI (kg/m2) 32 ± 5 32 ± 6 32 ± 5 32 ± 5 32 ± 6 32 ± 6

HbA1c (%) 8.3 ± 1.1 8.3 ± 1.1 8.4 ± 1.0 8.5 ± 1.0 8.1 ± 1.1 8.1 ± 1.2

FBG (mmol/L)b 8.9 ± 2.4 8.8 ± 2.5 9.5 ± 2.3 9.6 ± 2.4 7.8 ± 2.3 7.7 ± 2.2

Basal insulin, IGlar / none (%) 38 / 62 41 / 59 0 / 100 0 / 100 100 / 0 100 / 0


Data are mean ± standard deviation unless otherwise indicatedaFull analysis set, N numbers reflect maximum sample size for all demographics / characteristics with the exception of FBG;bTotal (IGlar, N = 359; LY IGlar, N = 353): insulin-naïve (IGlar, N = 224; LY IGlar, N = 209): prior IGlar (IGlar, N = 135;LY IGlar, N = 144); *p = 0.04BMI=body mass index; FBG=fasting blood glucose; HbA1c=glycosylated haemoglobin


LY IGlar (N = 353)

IGlar (N = 359)

0 2 12 24Week

Fa

sti

ng

Glu

co

se

(m

mo

l/L

)


FBG (mmol/L)LSM ± SE

Baseline 8.9 ± 0.1 8.8 ± 0.1 0.84

Week 24 LOCF 6.1 ± 0.1 5.9 ± 0.1 0.27

Change at Week 24 LOCF −2.6 ± 0.2 −2.7 ± 0.2 0.69

10

9

8

7

6

5

ELEMENT 2: FBG Changes over Time


FBG=fasting blood glucose; LOCF=last observation carried forward;LSM=least squares mean; SE=standard error


ELEMENT 2: 7-point SMBG

Baseline

24 Weeks LOCF

LY IGlar (N = 369)

IGlar (N = 375)

Glu

co

se

(m

mo

l/L

)

*

13

12

11

10

9

8

7

6

5


*p = 0.04LOCF=last observation carried forward; PPG=post-prandial glucose;SMBG=self-monitored blood glucose


ELEMENT 2: HbA1c Changes over Time


HbA1c, %LSM � SE

Baseline 8.31 � 0.06 8.35 � 0.06 0.61

Endpoint LOCF 6.99 � 0.06 7.04 � 0.06 0.40

6.5

7.0

7.5

8.0

8.5

9.0

LY IGlar (N = 369)

IGlar (N = 375)

Hb

A1

c(%

)

0 4 8 12 16 20 24Week


HbA1c=glycosylated haemoglobin; LOCF=last observation carried forward;LSM=least squares mean; SE=standard error


ELEMENT 2: Change at Study End in HbA1c and % of Patients Reaching Target HbA1c

Rosenstock et al. ADA 2014: 64-OR HbA1c=glycosylated haemoglobin; NS=not significant

−1.34 −1.29

-2.0

-1.5

-1.0

-0.5

0.0

∆ = 0.05295% CI (−0.07, 0.18)

p = NS

Ch

an

ge

in H

bA

1c

(%)

Total

−1.54 −1.48

Insulin-naïve

∆ = 0.06195% CI (−0.09, 0.21)

p = NS

−1.01 −1.02

Prior IGlar

∆ = −0.00495% CI (−0.19, 0.19)

p = NS

LY IGlarIGlar

Total

IGlar (N = 375)

LY IGlar(N = 369)

HbA1c <7% 53% 49%

Insulin-naïve

IGlar (N = 232)

LY IGlar(N = 217)

60% 54%

Prior IGlar

IGlar (N = 143)

LY IGlar(N = 152)

41% 41%


ELEMENT 2: Insulin Dose and Body Weight Changes at Study End


Data are least squares mean ± standard erroraN = 374 for body weightLOCF=last observation carried forward

0

0.2

0.4

0.6

0.8

U/k

g

Baseline 24 Weeks ∆ at 24 Weeks(LOCF) (LOCF)

Daily Basal Insulin Dose

0

20

40

60

80

100

120

kg

Body Weight

Baseline 24 Weeks ∆ at 24 Weeks(LOCF) (LOCF)

LY IGlar (N = 370)IGlar (N = 372a)


ELEMENT 2: Insulin Dose Subgroup Analysis


Data are least squares mean ± standard error*p = 0.038LOCF=last observation carried forward

Total Insulin-naïve Prior IGlar

IGlar (N = 372)

LY IGlar(N = 370)

IGlar (N = 229)

LY IGlar(N = 218)

IGlar (N = 143)

LY IGlar(N = 152)

Daily basal insulin dose (U/kg)

Baseline 0.14 � 0.01 0.16 � 0.01 0.00 � 0.00 0.00 � 0.00 0.35 � 0.01 0.39 � 0.01*

Week 24 (LOCF)

0.48 � 0.03 0.50 � 0.03 0.44 � 0.03 0.42 � 0.03 0.53 � 0.03 0.60 � 0.03

Change at Week 24 (LOCF)

0.37 � 0.02 0.36 � 0.02 0.46 � 0.03 0.45 � 0.03 0.19 � 0.03 0.22 � 0.03


0

25

50

75

100

Pa

tie

nts

(%

)

Incidence


78 79

5457

<1 <1

22

8

21

70

20

40

60

80 Event Rates

Eve

nts

/Pa

tie

nts

/1 Y

ea

r, M

ea

n (

SD

)

<1 <1




All p values >0.05SD=standard deviation

ELEMENT 2: Total, Nocturnal, and Severe hypoglycaemia


ELEMENT 2: Summary of Adverse Events

AEsa

IGlar (N = 380)

n (%)

LY IGlar(N = 376)

n (%)

Deaths 1 (0.3) 1 (0.3)

Serious AEs 18 (5) 15 (4)

Discontinuations due to an AE 11 (3) 6 (2)

Injection site AEs 11 (3) 13 (4)

TEAEs 184 (48) 196 (52)

Possibly related to study drug 23 (6) 26 (7)

Possibly related to study procedure 8 (2) 6 (2)

Possibly related to study disease state (diabetes) 18 (5) 19 (5)

Special topic assessment of AEs (allergic) 27 (7) 21 (6)


aPatients may be counted in >1 categoryAE=adverse event; TEAE=treatment-emergent adverse event


ELEMENT 2: Summary of Allergic Events and Injection Site Reactions


aPatients may be counted in >1 category;bMacular rash, papular rash, pruritic rash, vesicular rashAE=adverse event

AEsa

IGlar (N = 380)

n (%)

LY IGlar(N = 376)

n (%)

Special topic assessment of allergic events 27 (7) 21 (6)

Pruritus, rash, dermatitis, otherb 12 (3) 8 (2)

Arthralgia, peri-arthritis 9 (2) 7 (2)

Injection site (reaction, pruritis, induration) 4 (1) 5 (1)

Asthma, nasal edema 5 (1) 3 (1)

Injection site reaction (patient questionnaires) 11 (3) 13 (4)

Pain 5 (1) 10 (3)

Pruritus 4 (1) 4 (1)

Rash 3 (1) 3 (1)


ELEMENT 2: Incidence of Treatment-emergent Antibody Response (TEAR)


Patients with TEAR, n (%)

Week 24 (LOCF) 7 (2) 12 (3) 0.35

Overall 14 (4) 14 (4) >0.99

0

10

20

30

40

Pa

tie

nts

wit

h T

EA

R (

%)

Week

TEAR by week

0 4 12 24

LY IGlar (N = 365)

IGlar (N = 365)

TEAR criteria




Rosenstock et al. ADA 2014: 64-OR LOCF=last observation carried forward


• LY IGlar compared with IGlar demonstrated similar:– glucose-lowering effect (FBG, SMBG, HbA1c)

– insulin doses

– changes in body weight

– hypoglycaemia incidence and rates

– adverse-event profile

– allergic and injection site reactions

– incidence of treatment-emergent antibody response

ELEMENT 2: Summary

Rosenstock et al. ADA 2014: 64-OR FBG=fasting blood glucose; HbA1c=glycosylated haemoglobin; SMBG=self-monitored blood glucose


• LY IGlar compared with IGlar, in combination with oral antidiabetic drugs, provided equivalent efficacy and similar safety profiles, with no clinically meaningful differences in patients with type 2 diabetes mellitus

ELEMENT 2: Conclusion




Evaluation of Immunogenicity of LY2963016 Insulin Glargine Compared with Lantus® Insulin Glargine in Patients with T1DM or T2DM

Mark A. Deeg, Liza L. Ilag, William J. Huster,Robyn K. Pollom, Jason S. Zielonka,Melvin J. Prince, Robert J. Konrad

Eli Lilly and Company, Indiana, USA

Deeg et al. ADA 2014: 70-ORLantus is a registered trademark of Sanofi-Aventis


• To compare the immunogenicity profile of LY2963016 insulin glargine (LY IGlar) and Lantus® insulin glargine (IGlar) in patients with T1DM (ELEMENT 1) and T2DM (ELEMENT 2) using the following measurements:– Proportion of patients with detectable antibodies

– Treatment-emergent antibody response (TEAR)

– Treatment-emergent allergic events

– Relationships between clinical outcomes (HbA1c,basal insulin dose, and total hypoglycaemia) and TEAR status

Objective

Deeg et al. ADA 2014: 70-OR

HbA1c=glycosylated haemoglobin; T1DM=type 1 diabetes mellitus; T2DM=type 2 diabetes mellitusLantus is a registered trademark of Sanofi-Aventis


0

25

50

75

100

Pa

tie

nts

(%)

Week

0 6 12 24 52 52(LOCF)


Patients with detectable antibodies, n (%)

Overall 24 weeks 90 (34) 80 (30) 0.40

Overall 52 weeks 105 (39) 107 (40) 0.86

By weekLY IGlar (N = 265)IGlar (N = 267)

Proportion of Patients with Detectable Antibodies: ELEMENT 1

Deeg et al. ADA 2014: 70-ORLOCF=last observation carried forward


0

5

10

15

20

n = 12

n = 18

Pa

tie

nts

wit

h T

EA

R (%

)

Overall(24 Weeks)

Overall(52 Weeks)

Endpoint(52 Weeks)

n = 17

n = 25 n = 25

n = 29


p = 0.27

p = 0.57p = 0.20

TEAR criteria




Treatment-emergent Antibody Response (TEAR): ELEMENT 1



Relationship between HbA1c and Insulin Antibody Level at Endpoint: ELEMENT 1

Deeg et al. ADA 2014: 70-ORHbA1c=glycosylated haemoglobin


Effect of Overall TEAR Status on Change in Clinical Outcomes: ELEMENT 1


-0.6

-0.4

-0.2

0.0

HbA1c

TEAR No TEAR

∆ %

n = 25 n = 29 n = 242 n = 236

Basal Insulin Dose

-0.02

0.00

0.02

0.04

0.06

0.08

∆ U

/kg

/da

y

TEAR No TEAR

n = 25 n = 29 n = 241 n = 236

Total Hypoglycemia Rate

-6

-4

-2

0

2

∆ E

pis

od

es/

30

da

ys

TEAR No TEAR

n = 25 n = 29 n = 242 n = 236


Data are least squares mean (standard error) change from baseline to LOCF endpointp >0.05 for all treatment-by-TEAR interactionsHbA1c=glycosylated haemoglobin; LOCF=last observation carried forward;TEAR=treatment-emergent antibody response


0

25

50

75

100

Week

p = 0.047


Pa

tie

nts

(%)

0 4 12 24 24(LOCF)


Patients with detectable antibodies, n (%)

Overall 24 weeks 40 (11) 56 (15) 0.10

By week

Proportion of Patients with Detectable Antibodies: ELEMENT 2

Deeg et al. ADA 2014: 70-ORLOCF=last observation carried forward


Treatment-emergent Antibody Response (TEAR): ELEMENT 2


0

5

10

15

20

n = 14n = 14

Pa

tie

nts

wit

h T

EA

R (%

)

n = 12

n = 7

Endpoint(24 Weeks)

Overall(24 Weeks)


p >0.99p = 0.35


Relationship between HbA1c and Insulin Antibody Level at Endpoint: ELEMENT 2

Deeg et al. ADA 2014: 70-ORHbA1c=glycosylated haemoglobin



-2.0

-1.5

-1.0

-0.5

0.0

TEAR No TEAR

∆ %

n = 14 n = 14 n = 351 n = 351

HbA1c

0.0

0.2

0.4

0.6

0.8∆

U/k

g/d

ay

Basal Insulin Dose

TEAR No TEAR

n = 14 n = 14 n = 346 n = 3500

1

2

3

∆ E

pis

od

es/

30

da

ys

Total Hypoglycemia Rate

TEAR No TEAR

n = 14 n = 14 n = 347 n = 350

Effect of Overall TEAR Status on Change in Clinical Outcomes: ELEMENT 2


Data are least squares mean (standard error) change from baseline to LOCF endpointp >0.05 for all treatment-by-TEAR interactionsHbA1c=glycosylated haemoglobin; LOCF=last observation carried forward;TEAR=treatment-emergent antibody response


Summary of Allergic Events and Injection Site Reactions


Data are n (%) for patients with ≥1 treatment-emergent adverse eventAll p values >0.05aPatients may be counted in >1 category;bMacular rash, papular rash, pruritic rash, vesicular rash, photosensitivity reaction, urticaria;cInduration, nodule, swelling

ELEMENT 1 ELEMENT 2

Adverse eventsa IGlar (N = 267)

LY IGlar(N = 268)

IGlar (N = 380)

LY IGlar(N = 376)

Special topic assessment of allergic vents 11 (4) 20 (8) 27 (7) 21 (6)

Pruritus, rash, dermatitis, otherb 4 (2) 7 (3) 12 (3) 8 (2)

Arthralgia, arthritis, peri-arthritis 5 (2) 4 (2) 9 (2) 7 (2)

Injection site (reaction, pruritus, otherc) 2 (1) 6 (2) 4 (1) 5 (1)

Hypersensitivity 1 (0.4) 1 (0.4) -- --

Allergic respiratory symptom, asthma, nasal edema

0 2 (0.7) 5 (1) 3 (1)

Injection site reaction (patient questionnaires) 3 (1) 7 (3) 11 (3) 13 (4)

Pain 2 (0.7) 6 (2) 5 (1) 10 (3)

Pruritus 1 (0.4) 2 (0.7) 4 (1) 4 (1)

Rash 1 (0.4) 2 (0.7) 3 (1) 3 (1)


• In patients with T1DM or T2DM treated withLY IGlar or IGlar, there were no treatment differences in:– proportion of patients with detectable antibodies at baseline and

throughout the treatment period

– the incidence of TEAR

– relationships between clinical outcomes(HbA1c, basal insulin dose, and total hypoglycaemia) and TEAR status

– the incidence of treatment-emergent allergic events

Immunogenicity Summary


HbA1c=glycosylated haemoglobin; T1DM=type 1 diabetes mellitus;T2DM=type 2 diabetes mellitus; TEAR=treatment-emergent antibody response


Immunogenicity Conclusions

• LY IGlar and IGlar have a similar immunogenicity profile, with no effects of anti-insulin glargine antibodies on efficacy and safety outcomes in patients with type 1 diabetes mellitus or type 2 diabetes mellitus




BACK-UP SLIDES


ELEMENT 1: Summary of Allergic EventsBy System Organ Class and Preferred Term

System Organ ClassPreferred Term

IGlar (N = 267)

LY IGlar(N = 268)

Patients with ≥1 treatment-emergent allergic event 11 (4.1) 20 (7.5)

Skin and subcutaneous tissue disorders 4 (1.5) 7 (2.6)

Pruritus 1 (0.4) 3 (1.1)

Rash 2 (0.7) 2 (0.7)

Dermatitis allergic 0 1 (0.4)

Photosensitivity reaction 0 1 (0.4)

Urticaria 1 (0.4) 0

Musculoskeletal and connective tissue disorders 5 (1.9) 4 (1.5)

Arthralgia 5 (1.9) 3 (1.1)

Arthritis 0 1 (0.4)

General disorders and administration site conditions 2 (0.7) 6 (2.2)

Injection site reaction 2 (0.7) 3 (1.1)

Injection site induration 0 1 (0.4)

Injection site nodule 0 1 (0.4)

Local swelling 0 1 (0.4)

Immune system disorders 1 (0.4) 1 (0.4)

Drug hypersensitivity 0 1 (0.4)

Hypersensitivity 1 (0.4) 0

Respiratory, thoracic, and mediastinal disorders 0 2 (0.7)

Allergic respiratory symptom 0 1 (0.4)

Asthma 0 1 (0.4)


Data are n (%) for patients with ≥1 treatment-emergent allergic eventAll p values >0.05


ELEMENT 2: Summary of Allergic EventsBy System Organ Class and Preferred TermSystem Organ Class

Preferred Term

IGlar (N = 380)

n (%)

LY IGlar(N = 376)

n (%)

Patients with ≥1 treatment-emergent allergic event 27 (7.1) 21 (5.6)

Skin and subcutaneous tissue disorders 12 (3.2) 8 (2.1)

Pruritus 4 (1.1) 4 (1.1)

Rash 3 (0.8) 3 (0.8)

Dermatitis 2 (0.5) 1 (0.3)

Angioedema 0 1 (0.3)

Rash macular 1 (0.3) 0

Rash papular 1 (0.3) 0

Rash pruritic 1 (0.3) 0

Rash vesicular 1 (0.3) 0

Musculoskeletal and connective tissue disorders 9 (2.4) 7 (1.9)

Arthralgia 8 (2.1) 7 (1.9)

Peri-arthritis 1 (0.3) 0

General disorders and administration site conditions 4 (1.1) 5 (1.3)

Injection site reaction 3 (0.8) 3 (0.8)

Injection site pruritis 1 (0.3) 1 (0.3)

Injection site induration 0 1 (0.3)

Respiratory, thoracic, and mediastinal disorders 5 (1.3) 3 (0.8)

Asthma 5 (1.3) 2 (0.5)

Nasal edema 0 1 (0.3)


Date of preparation: May 2015 | UK/GLA/00036

Date of preparation: May 2015 | UK/GLA/00036

Documents

Data from BI-Lilly Diabetes Alliance. Date of preparation: May 2015 | UK/GLA/00036 Data from BI-Lilly Diabetes Alliance Date of preparation: May 2015 |