On Post Hoc Justification of Noninferiority Margin: A Case Study

• Activity Number: 100
• Type: Invited
• Date/Time: Monday, July 30, 2012 : 8:30 AM to 10:20 AM
• Sponsor: Biopharmaceutical Section
• Author(s): Sudhakar Rao*+ and Steven Sun
• Companies: Janssen Pharmaceuticals R&D and Johnson & Johnson
• Keywords: non-inferiority ; clinical trial ; margin justification

Abstract

For selection of non-inferiority (NI) margin, although the new FDA draft guidance on the use of non-inferiority trials gives a very good explanation on what the margin means, and the combined statistical and clinical argument that must be made to justify it, it does not give specific guidance on the selection of the margin. Because of the ambiguous nature in the NI margin selection, the sponsor typically needs to get agreement from the regulatory agency before conducting a NI trial. In this talk, we will present a case study where the justification of a NI margin is made after the completion of the study. More specifically, we first used meta-analyses to estimate the control effect. Then we applied both the classical fixed margin approach using the standard 95/95 rule (based on the lower 95% confidence limit of the control effect) and the synthesis approach to demonstrate that the test treatment is not only non-inferior to the active control agent, but also superior to placebo.