Novel Response-Adaptive Randomization Designs for Clinical Trials with Time-to-Event Outcomes

• Activity Number: 592
• Type: Topic Contributed
• Date/Time: Wednesday, August 12, 2015 : 2:00 PM to 3:50 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Oleksandr Sverdlov*
• Companies: EMD Serono
• Keywords: ethical concern ; efficiency ; multi-arm trial ; optimal design ; Response-adaptive randomization ; survival trial

Abstract

Response-adaptive randomization (RAR) designs modify treatment randomization probabilities based on accumulating data from patients in the trial to achieve some pre-specified experimental objectives while maintaining validity and integrity of the trial results. In this presentation I will give an exposition to some novel optimal RAR designs for multi-arm trials with time-to-event outcomes. It will be shown that these designs can achieve modest reduction in treatment failures without loss in statistical efficiency. Therefore, these designs can be attractive in clinical trials where the outcome is severe or life threatening and even modest reduction in treatment failures in the trial can be viewed as a significant contribution from an ethical endpoint. Some examples will be given to illustrate the application of the methodology in multi-arm survival trials.