Interim Analysis: Some Statistical and Regulatory Challenges

• Activity Number: 13
• Type: Topic Contributed
• Date/Time: Sunday, August 3, 2014 : 2:00 PM to 3:50 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Pabak Mukhopadhyay*+ and Satrajit Roy Choudhury
• Companies: Novartis and Novartis

Abstract

Decision to stop a clinical trial for superior efficacy before its completion require complex combination of ethical, statistical, and practical considerations. An Independent Data Monitoring Committee (IDMC), made up of experienced clinicians and statistician is often entrusted with making the decision to stop for overwhelming efficacy based on predefined stopping boundaries and other relevant information. This mechanism to stop early is a very important as it allows a potential life changing drug patients sooner. However it well known from statistical literature that there is possibility for the observed benefit and the 95% CI to be biased upwards when a trial is stopped early for superiority. This creates challenges for both the regulators and the sponsor in making a proper risk benefit assessment. Often the consequence is to wait for more data to come in either from the trial under consideration or from a separate trial, with the risk of delaying treatments to patients. In this talk we will consider some statistical methods to mitigate the above concerns