Application of Resampling Method in Futility Analysis Based on Multiple Studies

• 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): Jun Zhao*+ and Annpey Pong and Jingshan Zhang
• Companies: Merck and Merck and Celgene Corporation
• Keywords: futility ; interim analysis ; probability of success ; Bayesian methods ; resampling

Abstract

In late stage drug development, it is common to conduct multiple studies simultaneously in a development program in order to fulfill regulatory filing requirements as early as possible. However, the design elements such as sample sizes and corresponding powers are determined based on information from past studies at earlier phases. In addition, the assumptions may vary in terms of timing or conditions of conducting these new studies. Considering limited resource, the futility analysis is usually to be considered at interim to stop a study or a program early if the expected probability of success is relatively low. In this research work, a resampling method is proposed to assess the expected probability of success when a futility evaluation of multiple studies is performed in an interim analysis in a development program.