A New Bayesian Sample Size in Exploratory Clinical Trials with Two Coprimary Endpoints

• Activity Number: 633
• Type: Topic Contributed
• Date/Time: Thursday, August 13, 2015 : 8:30 AM to 10:20 AM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Wei Zhong* and Xin Wang and Yifan Huang
• Companies: Takeda Pharmaceuticals and AbbVie and AstraZeneca
• Keywords: sample size determination ; Bayesian sample size ; exploratory clinical trials ; co-primary endpoints ; searching algorithm

Abstract

Sample size determination is a key component of clinical trial designs in drug development. The frequentist approaches work well in most confirmatory trials where sufficient knowledge on the investigated drug has been gained from early studies. In exploratory clinical trials, many new Bayesian approaches have been implemented to expedite drug development recently. The Bayesian sample size proposed by Whitehead et al justifies the acceptable minimum sample size on one primary endpoint by a "conclusiveness" condition, which is quite interpretable and meaningful for decision making. In this work, we expanded their concept and proposed a new Bayesian sample size for the two co-primary endpoints case. Two searching algorithms were introduced with theoretical proofs of their validity. Our approach can also be flexibly extended to a multi-stage trial design with sample size re-estimation. Examples of sample size selection were discussed for various correlations, decision probability boundaries and standardized effect sizes.