Auto-Adaptive Alpha Allocation: A Strategy to Mitigate Risk on Study Assumptions

• Activity Number: 406
• Type: Topic Contributed
• Date/Time: Tuesday, August 11, 2015 : 2:00 PM to 3:50 PM
• Sponsor: Section on Medical Devices and Diagnostics
• Author(s): Yue Shentu* and Cong Chen and Lei Pang
• Companies: Merck and Merck Research Laboratories and Merck
• Keywords: adaptive design ; oncology ; targeted therapy ; multiplicity

Abstract

Potential biomarkers in clinical development programs may have promising but uncertain predictive effect. It is risky to study only the unselected population, or just the biomarker subpopulation. In 2009, Chen and Beckman proposed a Bayesian decision framework to optimize the type I error rate (alpha) allocation in a Phase III clinical study with possible predictive subset effect. The utilization of internal data in this framework is of particular interest because it provides an opportunity to mitigate the potential risk of mis-specified study assumptions using an auto-adaptive strategy. In this paper, we examine this auto-adaptive strategy in detail through extensive numerical case studies, and provide guidance on the appropriate use of partial internal data in this data-driven optimization framework. We show that given considerable uncertainty in the external data, internal data can be used to inform the alpha allocation to hypothesis testing in the unselected population and the subgroup. The resulting adaptive testing strategy is robust with respect to the uncertainty in the predictive subgroup effect and biomarker prevalence.