Integrated Multiple Adaptive Clinical Trial Design Involving Sample Size Re-Estimation and Response-Adaptive Randomization for Continuous and Binary Outcomes

• Activity Number: 341 - Topics in Adaptive Designs: Sample Size, Randomization and Related Topics
• Type: Contributed
• Date/Time: Wednesday, August 5, 2020 : 10:00 AM to 2:00 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Christine M Orndahl*
• Companies: Virginia Commonwealth University Dept of Biostatistics
• Keywords: multiple adaptive design; sample size re-estimation; response-adaptive allocation; clinical trial design; weighted sum optimization; integrated adaptive design

Abstract

Recent interest has emerged in the area of multiple adaptive designs, where more than one adaptive design is utilized within a single clinical trial. Usually, multiple adaptive components are performed independently and sequentially, limiting the flow of information between adaptive designs. Our proposed methods integrate two adaptive designs, sample size re-estimation (SSR) and response-adaptive randomization (RAR), into a clinical trial with continuous or binary outcomes. Weighted sum multi-objective optimization is used to simultaneously optimize two objective functions: the first minimizes the sample size and the second minimizes 1) a function of the total expected response (assuming a smaller response is preferable) for continuous outcomes and 2) a function of the total expected number of treatment failures for binary outcomes. This allows the aims of SSR and RAR to be considered concurrently, resulting in sample sizes and allocation ratios that inform each other and are adaptively adjusted throughout the trial. Performance of these methods in simulated clinical trials is presented and compared to singly adaptive, sequentially adaptive, and non-adaptive designs.