Adaptive Sample Size Re-estimation When Interim Results Are Promising

• Activity Number: 221
• Type: Topic Contributed
• Date/Time: Monday, August 2, 2010 : 2:00 PM to 3:50 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Cyrus R. Mehta*+
• Companies: Cytel Inc.
• Keywords: conditional power ; interim analysis ; adaptive ; trial design

Abstract

The sample size of a clinical trial is usually determined by statistical power calculations. Often there is uncertainty and debate as to what magnitude of alternative hypothesis treatment difference and between-patient variability it is appropriate to pre-specify. Hence there exist methods to facilitate re-estimation of the sample size in the light of interim efficacy results from the ongoing trial. Mos methods require the use of a weighted statistic for the final analysis in order to preserve the type-1 error. This is non-intuitive and confusing to medical practioners, involving the down-weighting of data contributed by one set of patients relative to the others. Chen, DeMets and Lan (2004) showed that the conventiona statistic may be used for the final analysis when the interim results are promising. We extend their results and provide some additional insights into their working.