Overview of Adaptive Design Zoran Antonijevic Regulatory SIAC 16 Oct, 2013

Presentation Contents FDA Guidance Exploratory Stage More efficient drug development Confirmatory Stage (A&WC Trials) Different Types FDA approvals based on adaptive designs in Phase 3 DMC for Adaptive Clinical Trials

2010 FDA Guidance

Well Understood Adaptive Designs A. Adaptation of Study Eligibility Criteria Based on Analyses of Pretreatment (Baseline) Data B. Adaptations to Maintain Study Power Based on Blinded Interim Analyses of Aggregate Data C. Adaptations Based on Interim Results of an Outcome Unrelated to Efficacy D. Adaptations Using Group Sequential Methods and Unblinded Analyses for Early Study Termination Because of Either Lack of Benefit or Demonstrated Efficacy E. Adaptations in the Data Analysis Plan Not Dependent on Within Study, Between-Group Outcome Differences

Less Well Understood (LWU) Adaptive Designs Exploratory A. Adaptations for Dose Selection Studies B. Adaptive Randomization Based on Relative Treatment Group Responses Confirmatory (but it could be done at Exploratory Stage as well) C. Adaptation of Sample Size Based on Interim-Effect Size Estimates D. Adaptation of Patient Population Based on Treatment-Effect Estimates E. Adaptation for Endpoint Selection Based on Interim Estimate of Treatment Effect

Exploratory Stage

Adaptation for Dose Selection Value of Design Easily under-appreciated, trial designs have a major impact on the value of a pharmaceutical product Adaptive Phase 2 Designs FDA AD Guidance recommends use of AD in Phase 2 AD is the most efficient approach for this stage of development Development Program Optimization Phase 2 most critical for optimal drug development Development Scenario Simulation Comparing relative efficiency of various development options. Accomplished only with simulations

Traditional Phase 2 Program . A Single Dose-Adaptive Design can replace Traditional PoC trial and Ph2 Trials Traditional Phase 2 Program . Phase 3 PoC (1b/2a) (High Dose vs. Placebo) Dose- Finding Definitive Dose-Response (if needed) Phase 2 with Dose-Adaptive PoC Trial Phase 3 PoC + Dose-Finding Definitive Dose-Response (if needed)

Phase 3: Seamless Phase 2/3 Adaptive Design Single Dose-Adaptive Design can replace Typical PoC and Ph2a and Ph2b Trials Traditional Phase 2 Program . Phase 3 PoC (Ib/IIa) (High Dose vs. Placebo) Dose- Finding Definitive Dose-Response (if needed) Phase 2 with Dose-Adaptive PoC Trial Phase 3: Seamless Phase 2/3 Adaptive Design PoC + Dose-Finding

Confirmatory Stage

Novel (LWU) Adaptive Designs at Confirmatory Stage Unblinded Sample Size Re-Assessment (Adaptation of Sample Size) Two stage design with dose selection (Seamless Phase 2b/3 design) Adaptive Enrichment Design (Adaptation of Patient Population)

Selected Approvals Based on Adaptive Designs PROCYSBI™ ; Approved in 2013 Raptor Pharmaceuticals Nephrophatic Cystinosis Unblinded Sample Size Re-assessment Fulyzaq™; Approved in 2012 Napo/Salix HIV-associated diarrhea Two-Stage (non-stop) Phase 2/3 Design

Unblinded SSR Two stage slow horiz reveal

Two Stage Design With Dose Selection

Predictive Enrichment Identification of patients more likely to respond to a given treatment Adaptive Enrichment Design validates predictive marker and confirms efficacy within a single trial.

Phase 2 to 3 Biomarker Development Paths Whole Population Biomarker Subpopulation _ + Biomarker Subpopulation _ = + + ? Phase III in Full population. Drug works, no differentiation between subpopulations. Phase III in Enriched population only. Drug works in biomarker subpopulation only. Phase III with an Adaptive Enrichment Design. Phase 2 data inconclusive. Phase III in Enriched population only. Drug works in biomarker subpopulation only. Stop development. Drug does not work.

Adaptive Enrichment Design Enroll subjects from full population Interim Analysis at … PFS events Favorable Zone: CPFull> X% Inconclusive: Z % ≤ CPFull ≤ X% Futility: CPFull≤ Z% Enrichment Zone: CPSub-CPFull> Y% Unfavorable Zone: CPSub-CPFull≤ Y% Continue as Planned: enrollment with full population Restrict enrollment to biomarker sub. Increase Nsub Continue as Planned: enrollment with full population

DMC for Adaptive Clinical Trials

Pro-activity of the DMC The FDA Adaptive Design Guidance: “Because the DMC is unblinded to interim study results it can help implement the adaptation decision according to the prospective adaptation algorithm, but it should not be in a position to otherwise change the study design except for serious safety-related concerns that are the usual responsibility of a DMC”

Novel (LWU) Adaptive Designs Basics; Unblinded Sample Size Re-assessment Complex Adaptive Designs at Confirmatory Stage Two-Stage with Dose-Selection Adaptation of Patient Population Adaptive Designs in Exploratory Studies Dose Selection Studies

Unblinded Sample Size Reassessment Similarities with GSD: - Both address the same objective/concern which is uncertainty regarding the treatment effect - The adaptation algorithm can be pre-specified in a relatively straightforward manner Differences: GSD; concerns re overflow SSR; design carefully not to reveal treatment effect

Complex Adaptive Designs at Confirmatory Stage Decisions involve parameters beyond just ethical considerations. Include decisions that traditionally have been sponsor responsibilities. May have major commercial implications, for example dose selection or sub-population selection. Complex algorithms DMC may desire to consult with the steering committee that includes a sponsor representative not involved in trial management.

Adaptive Designs in Exploratory Studies Not intended to demonstrate confirmatory evidence; less rigor in restricting knowledge of interim results Lesser regulatory implications, though avoiding operational bias should be recommended in any trial May require frequent updates based on interim data Adaptations are often driven by a pre-specified algorithm automated through an integrated response system Therefore, it may be preferable that recommendations are endorsed by an Internal Review Committee

References US Food and Drug Administration. Draft Guidance for Industry: Adaptive Design Clinical Trials for Drugs and Biologics, February, 2010. Zoran Antonijevic, Jose Pinheiro, Parvin Fardipour, Roger J. Lewis. Impact of Dose Selection Strategies Used in Phase II on the Probability of Success in Phase III. Statistics in Biopharmaceutical Research 2010 2:4, 469-486 Mehta CR, Pocock SJ. Adaptive increase in sample size when interim results are promising: A practical guide with examples. Stat Med 2011; 30:3267–84. Zoran Antonijevic, Paul Gallo, Christy Chuang-Stein, Vlad Dragalin, John Loewy, Sandeep Menon, Eva Miller, Caroline Morgan, Matilde Sanchez. Views on Emerging Issues Pertaining to Data Monitoring Committees for Adaptive Trials. Therapeutic Innovation & Regulatory Science. 2013 July 2013 47: 495-502

Back-Ups

Case Study Scenarios Results (7) different statistical approaches (2) Phase 2 strategies (3) Phase 3 strategies For a total of 42 scenarios. Same assumptions for safety and efficacy used for all scenarios Cost and development time calculated using real data from existing databases Results The expected NPV for different scenarios ranged from $0.5B to $1.2B Adaptive Designs far outperformed traditional approaches

Expected NPV Method