Application of Bayesian Approaches to Noninferiority Trials

• Activity Number: 307
• Type: Contributed
• Date/Time: Tuesday, August 6, 2013 : 8:30 AM to 10:20 AM
• Sponsor: Biopharmaceutical Section
• Author(s): Radha Railkar*+ and Mani Lakshminarayanan
• Companies: Merck & Co. and Merck Research Laboratories
• Keywords: Historical data ; Active control ; Non-inferiority margin ; Posterior probability

Abstract

Conventional active control non-inferiority trials are designed to show that the new treatment is not inferior to standard treatment by some clinical acceptable difference. Non-inferiority testing for therapeutic effectiveness in active controlled trials is often controversial despite the fact that there is considerable experience in using this approach to establish the effectiveness of a new treatment. Some of the controversial issues include choice of active control, statistical methods for estimation of the control effect, methods for determination of the non-inferiority margin, and others. As a result, these trials are unique in the sense that a meaningful interpretability of their results will need historical information. Application of a Bayesian approach presents a viable alternative as its paradigm is appropriate for sequential learning of current data in the light of historical information. In this presentation, we will explore the application of Bayesian methods in clinical trials involving non-inferiority hypotheses including a discussion on the selection of non-inferiority margin. Comparisons will be made between frequentist and Bayesian approaches via simulation.