A Bayesian Approach to Incorporate Prior Information in Interim Decisionmaking with Internal DMC

• Activity Number: 127
• Type: Contributed
• Date/Time: Monday, August 4, 2014 : 8:30 AM to 10:20 AM
• Sponsor: Biopharmaceutical Section
• Author(s): Gang Jia*+ and Bruce Binkowitz and Paul DeLucca
• Companies: Merck and Merck and Merck
• Keywords: Phase II Dose Study ; Interim Analysis ; Bayesian ; Data Monitoring Committee

Abstract

For phase II dose ranging studies with big sample size, long enrollment period and quick to evaluate endpoints, it is often practical and beneficial to have an internal interim analysis to assess the study results early. The benefit is two-fold: 1. early detection of efficacy enables the company to move to Phase III development sooner; 2. early decision on futility leads to stopping the study on a non-effective drug which enables the reallocation of resources to other programs. An internal Data Monitoring Committee is an ideal decision body to look at the un-blinded interim results and make the decision based on pre-defined rules to protect the integrity of the study and objectiveness of decision making. However, the go/no-go decision making in such scenario often involves evaluating the totality of evidence which usually go beyond the interim results of the phase II dose ranging study. We propose a strategy that incorporates prior evidence on efficacy endpoints with the interim results of phase II study using a Bayesian analysis approach.