Bayesian Adaptive Design for Evaluating Treatment with Predefined Subgroup

• Activity Number: 555
• Type: Topic Contributed
• Date/Time: Wednesday, August 1, 2012 : 2:00 PM to 3:50 PM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Ying Yang*+
• Companies: FDA
• Keywords: Bayesian Adaptive ; pre-defined subgroup

Abstract

In this presentation, we consider clinical trials with the objective of establishing efficacy claim for the total population or a targeted subgroup in case the trial did not meet its objective for the total population. We propose an efficient development tool that combines a test for overall treatment effect in all randomly assigned patients with a test for the targeted subgroup in a single clinical trial. The procedure provides prospective tests of the hypotheses that the new treatment is beneficial for the entire patient population or that it is beneficial for a pre-defined subset of patients. Bayesian decision tool will be used to determine whether the new treatment is beneficial for the pre-defined subpopulation at the end of study time for all randomized patients. The simulation study demonstrates that the proposed adaptive design maintains the power to detect the overall effect when the new treatment is effective for the entire patient population and controls the study-wise Type I error rate.