Efficacy Lead-In Phase 2 Single-Arm Trial Designs

• Activity Number: 320 - Methods Tailored to Unique Data and Trial Features
• Type: Contributed
• Date/Time: Wednesday, August 5, 2020 : 10:00 AM to 2:00 PM
• Sponsor: Section on Medical Devices and Diagnostics
• Author(s): Seongho Kim* and Weng-Kee Wong and Nithya Ramnath and Manohar Ratnam
• Companies: Wayne State University and UCLA and Department of Medicine, University of Michigan and Department of Oncology, Wayne State University
• Keywords: Cross-Entropy; Efficacy; Lead-in cohort; Phase 2; Run-in cohort; Single-arm

Abstract

One often desires to include an efficacy lead-in cohort for the higher probability of detecting a treatment effect, but no design is available for an efficacy lead-in cohort yet. Thus, we develop one- and two-stage single-arm designs for efficacy lead-in cohorts. The objectives of an efficacy lead-in cohort are to optimize regimens for a relatively safe or established treatment in terms of efficacy as well as to spare study duration and patient by allowing the optimized lead-in cohort to be included in the post-efficacy cohort. An efficacy lead-in cohort shares the same endpoint with the corresponding efficacy cohort and so requires a correction for multiple testing. Its sample size will not only depend on an investigator’s pre-knowledge but also require being satisfied with criteria for the post-efficacy phase of the trial. To satisfy requirements, we use two-stage and three-stage single-arm designs along with Cross-Entropy (CE) optimization resolving computational burden and so making the develop designs efficient. We further analytically derive the minimum sample size for the efficacy lead-in cohorts and use simulation, real-life application to validate the developed algorithms.