A Two-Stage Testing Strategy for Demonstrating Efficacy in Replicate Trials

• Activity Number: 599
• Type: Contributed
• Date/Time: Wednesday, August 7, 2013 : 2:00 PM to 3:50 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Chengxing Lu*+ and Tony Chen
• Companies: Novartis and Novartis
• Keywords: False positive rate ; Multiplicity adjustment ; Combined analysis ; Power

Abstract

As one of the regulatory requirements in most of the therapeutic areas, replicated evidence from at least two confirmatory trials are to be demonstrated for a successful registration in the United States. Taking various advantages from combined analysis given the similarity nature of the studies with replicated designs and trial conduction, we propose a two-stage testing strategy for such studies with the benefit of increased power, while the reproducibility concern from the individual studies is properly addressed. Optimal way of controlling false positive rate in each stage of the testing strategy while maximizing the power will be discussed. Power gained from this strategy will be presented with scenarios of either fixing sample size from traditional design or modifying sample size with the adjustment of the false positive rate. Simulation studies along with real data examples will also be provided.