1 Most common adverse event: lymphocytopenia Lymphocytopenia was - - PDF document

1

A Semi-mechanistic Model of Lymphocyte Dynamics in Patients with Multiple Sclerosis Treated with Cladribine Tablets A.L. Quartino (1), P. Girard (2), M.O. Karlsson (1), A. Munafo (2)

(1) Dept of Pharmaceutical Biosciences, Uppsala University, Uppsala, Sweden; (2) Modeling and Simulation, Merck Serono S.A. – Geneva, Switzerland*

*An affiliate of Merck KGaA, Darmstadt, Germany

2

Cladribine tablets: effective in multiple sclerosis (MS)

Cladribine selectively reduces peripheral lymphocyte counts Evaluated in the 96-week, parallel group, double-blind phase III CLARITY trial (1,326 patients)

– Placebo – Cumulative dose 3.5 mg/kg 4 courses (Weeks 1, 5, 48 and 52) – Cumulative dose 5.25 mg/kg 6 courses (Weeks 1, 5, 9, 13, 48 and 52)

Giovannoni G, et al. N Engl J Med 2010;362(5):416–26 100 75 50 25 Patients with no relapse (%) 12 24 36 48 60 72 84 96 Weeks Placebo Cladribine tablets 5.25 mg/kg P<0.001 Cladribine tablets 3.5 mg/kg P<0.001 300 250 200 150 100 50 Cumulative no. of relapses 12 24 36 48 60 72 84 96 Weeks Treatment course (4–5 days) 252 114 109 Placebo Cladribine tablets 5.25 mg/kg p<0.001 Cladribine tablets 3.5 mg/kg p<0.001

2

3

Most common adverse event: lymphocytopenia

Time after dose (weeks) Mean and observed ALC values in each patient over time

The dotted blue line represents the lower limit of CTCAE Grade 1 lymphopenia Treatment course (4–5 days)

Lymphocytopenia was mostly graded as mild or moderate Infection rates may increase with severe lymphocytopenia (below CTCAE Grade 2)

ALC, absolute lymphocyte count; CTCAE, Common Terminology Criteria for Adverse Events

Lymphocyte count (*109/L) Placebo 5.25 mg/kg group 3.5 mg/kg group 10 5 4 3 2 1 0.5 0.1 12 24 36 48 60 72 84 96 12 24 36 48 60 72 84 96 0 12 24 36 48 60 72 84 96

4

Rationale and objectives of ALC model and simulator

Treatment delay may be necessary in a few cases to allow recovery

• f ALC to above CTCAE Grade 2 and prevent the development of

severe lymphopenia

– Need for a tool to evaluate the potential impact of treatment delay

Identify a predictive model for ALC Build a simulator for evaluation of

– Lymphocyte dynamics in the rare cases of patients requiring treatment delay – Proportion of patients showing Grade 2 and over, and their recovery time – Percentage of patient completing treatments

3

5

Initial model for ALC

Based on myelosuppression model1

– Cladribine myelosuppression reversible effect – + Cumulative cladribine non-reversible effect – + Gender, body weight and creatinine clearance (CRCL) covariate influence 1 Blood circulation ALC kcirc = LN(2) / ktr ktr MMT=(n+1)/k tr 3 ktr ktr SLOPE * Cp E = Non-mitotic compartments T½ cell pool Proliferating 2

Clad =

Non-mitotic compartments

EClad, cum

Amax *

50 +

=

max

A

• 1. Friberg LE, et al. J Clin Oncol 2002;20:4713–21

ALC 0 γ

γ γ γ

Feedback = ALCt

0 γ

γ γ γ

Feedback = Dosecum/CRCL Dosecum/CRCL

6

Typical prediction ALC versus time (typical 67 kg female)

Lymphocyte count (*109/L)

5.25 mg/kg group 3.5 mg/kg group

Time (weeks) 12 24 36 48 60 72 84 96 2 1 0.5 0.25 12 24 36 48 60 72 84 96 Cumulative drug effect Combined drug effect Transient drug effect Cumulative drug effect Combined drug effect Transient drug effect 2 1 0.5 0.25

4

7

VPC looks OK ...

Time after dose (weeks) Lymphocyte count (*109/L) Placebo 5.25 mg/kg group 3.5 mg/kg group 10 5 4 3 2 1 0.5 0.1

12 24 36 48 60 72 84 96 108 0 12 24 36 48 60 72 84 96 108 0 12 24 36 48 60 72 84 96 108

The model seems to adequately describe the data and its variability: 2.5, 50 and 97.5 percentiles of observed data and their model-derived 95% CI

CI, confidence interval; VPC, Visual Predictive Check

8

But NPC revealed over-estimation of lymphopenia after 1 year

NPC, Numerical Predictive Check

Patients (%) Observed and simulated % of lymphopenia grade=0; initial model 20 40 60 80 100 80 60 40 20 Patients (%) Observed and simulated % of lymphopenia grade=1; initial model Patients (%) Observed and simulated % of lymphopenia grade=2; initial model Weeks 100 80 60 40 20 Patients (%) Observed and simulated % of lymphopenia grade=3; initial model Weeks 100 80 60 40 20

0, 1, 2, 3 Observed % of patients with CTCAE grade Cladribine courses Simulated % of patients ALC check for 3rd course

20 40 60 80 100 80 60 40 20 20 40 60 80 20 40 60 80

0 00 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 22 2 2 2 2 2 2 2 2 2 2 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3

5

9

Model update for ALC

Cladribine myelosuppression reversible effect + Cumulative cladribine (formerly non-reversible) effect + Gender, body weight and CRCL covariate influence + Recovery on cumulative effect

1 Blood circulation ALC kcirc = LN(2) / ktr ktr MMT=(n+1)/k tr 3 ktr ktr SLOPE * Cp E = Non-mitotic compartments T½ cell pool Proliferating 2

Clad =

Non-mitotic compartments

EClad, cum

Amax *

50+

=

max

A 50 ALC 0 γ

γ γ γ

Feedback = ALCt

0 γ

γ γ γ

Feedback =

Σ Σ Σ Σ[Dosecum /CRCL*e-K(t-Tdose)] Σ Σ Σ Σ[Dosecum /CRCL*e-K(t-Tdose)]

10

• Patients (%)

Observed and simulated % of lymphopenia grade=0; updated model

Patients (%)

Observed and simulated % of lymphopenia grade=1; updated model

Patients (%)

Observed and simulated % of lymphopenia grade=2; updated model

Weeks Patients (%)

Observed and simulated % of lymphopenia grade=3; updated model

Weeks

Model validation: with cumulative AUC recovery

• 0, 1, 2, 3 Observed % of patients with CTCAE grade

Cladribine courses Simulated % of patients ALC check for 3rd course

20 40 60 80 100 80 60 40 20 100 80 60 40 20 100 80 60 40 20 20 40 60 80 100 80 60 40 20 20 40 60 80 20 40 60 80

0 00 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 22 2 2 2 2 2 2 2 2 2 2 2 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3

6

11

Simulation of ALC model*

Patients may in rare cases need to delay additional courses while their ALC recovers 10,000 patients resampled from CLARITY study database for gender, body weight, creatinine clearance and baseline ALC One course of treatment = 5 days of body-weight, dose-adjusted cladribine tablets No more than 4 courses of treatment per 48 weeks

*TS2 (Pharsight, under Windows XP)

Time after dose (weeks) 3.5 mg/kg group 12 24 36 48 60 72 84 96 Lymphocyte count (*109/L) 10 5 4 3 2 1 0.5 0.1

Lower limit of CTCAE Grade 1

Mean and

• bserved ALC

values in each patient over time

12

Model + patient resampling in TS2

Lymphocyte count (1,000/mmˆ3) 4 3 2 1 200 simulated patients and nominal dose timings Time (weeks) 2.0 4.0 6.0

7

13

Week 5 Week 49 Week 53 Week after ≥Grade 2 observation Patients with ≥Grade 2 lymphopenia, % 4 8 12 16 20 24 36 48 30 25 20 15 10 5

Simulation to estimate percentage of patients in CTCAE Grade 2 at start of courses 2 to 4, and their time to recovery

NB This simulation assumes that if ALC <800 / mm3 at starting of 2nd, 3rd or 4th course, corresponding to weeks 5, 49 and 53, respectively, no more treatment is given

14

80 70 60 50 40 30 20 10

Proportion of patients achieving 1 to 4 courses (simulated data)

No postponing Patients % 1 2 3 4 Number of courses +11%

• 8%
• 4%

Postpone course when ALC <800/mm3

8

15

Conclusions

A model has been developed of lymphocyte dynamics with transient and slowly-recovering effect in patients with multiple sclerosis receiving cladribine tablets treatment The proposed model can be used to predict ALC dynamics in patients receiving cladribine tablets in a clinical trial or real-life setting Simulations allow exploration of

– % of patients recovering to CTCAE Grade 1 and their recovery time – % of patients completing the full treatment (4 courses over 96 weeks)

The ALC model is being coupled with an efficacy model It will be further refined with longer observation data

16

Acknowledgements and disclosures

Uppsala University Merck Serono

– Pharmacometrics – Marianne Ekblom, PharmD, PhD – Dan Mikol, MD

This study was funded by Merck Serono S.A. – Geneva, Switzerland* M Karlsson is a paid consultant for Merck Serono S.A.* A Munafo and P Girard are employees of Merck Serono S.A.*

Cladribine tablets treatment is not approved in the USA. Marketing authorization for the use

• f cladribine tablets in patients with RRMS has been granted in Russia and Australia (2010).

Please refer to full prescribing information for further details on use

*An affiliate of Merck KGaA, Darmstadt, Germany