Biased Interim Results Due to Regional Difference

• Activity Number: 427
• Type: Contributed
• Date/Time: Tuesday, August 6, 2013 : 2:00 PM to 3:50 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Jun Zhao*+ and Gang Li
• Companies: Merck & Co, Inc and Johnson & Johnson
• Keywords: MRCT ; futility ; interim analysis ; probability of study success ; resampling

Abstract

In late stage clinical development, interim analysis may be applied to evaluate the probability of success in order to stop a study early for either efficacy or futility, or adjust sample size when design elements are uncertain. The timing of interim analysis is often determined by the number of patients entered the study. However, the interim patient population may not reflect the final one. For example, regions may participate in the study at different time in multiregional trials. Those late participating regions are less or no representation in the interim analysis. Thus, the interim data is biased and the decision based the interim result is not robust. In this research work, we calculated the expected probability of study success for different scenarios. Data from an anti-psychotic drug is used for illustration. Resampling methods are used to calculate the expected probability of study success at interim.