Sample Size Considerations When Using the Synthesis Method for Non-inferiority Trials

• Activity Number: 522
• Type: Contributed
• Date/Time: Wednesday, August 1, 2012 : 10:30 AM to 12:20 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Huei Wang*+ and Guanghui Wei
• Companies: Amgen, Inc. and Amgen, Inc.
• Keywords: Non-inferiority trial ; Synthesis Method ; Preservation Rate ; Fixed Margin Method ; Non-inferiority Margin

Abstract

Non-inferiority (NI) trials are widely used in drug development. The choice of NI margin has important practical consequences, e.g. a smaller margin requires a larger sample size and a large margin may lead to false conclusion of drug effectiveness. In NI trials comparing test drug to active control, one may consider two margins (1) the margin based on that whole active control effect (M1) (2) the largest clinically acceptable difference of the test drug compared to the active control (M2). Showing the effect size of M1 would only provide assurance that the test drug has an effect greater than placebo. Fixed margin and synthesis approaches are the two conventional strategies to show NI to M2. For situations in some therapeutic areas, it is challenge what preservation rate for M2 should be chosen based on synthesis approach in order to sufficiently demonstrate the test drug effect over placebo and active control. In addition, it is mathematically possible that sample size required for 2nd stage (M2) is less than N required for 1st stage (M1). This poster presentation discusses the issues and potential solutions by using a real example for phase 3 trial planning.