A Bayesian Adaptive Dose Selection Procedure with a Count Endpoint

• Activity Number: 228
• Type: Topic Contributed
• Date/Time: Monday, August 1, 2011 : 2:00 PM to 3:50 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Luca Pozzi*+ and Heinz Schmidli and Mauro Gasparini and Amy Racine
• Companies: University of California at Berkeley and Novartis Pharma AG and Politecnico di Torino and Novartis
• Keywords: Clinical Trials ; Bayesian Model Averaging ; Predictive Probability ; Adaptive Design ; Dose Selection

Abstract

In clinical drug development, a sequence of studies is carried out to identify an efficacious and safe dose of a newly developed pharmaceutical drug. Adaptive designs Can considerably improve upon traditional designs, by modifying design aspects of the ongoing trial, including early stopping, adding or dropping doses, or changing the sample size. In the present work we propose a two-stage Bayesian adaptive design for a Phase II study aimed at selecting the lowest e?ective dose for Phase III. In the first stage patients are randomized to placebo, maximal tolerated dose, and one or more additional doses within the dose range. Based on an interim analysis, the study is either stopped for futility or success, or enters the second stage, where newly recruited patients are allocated to placebo, maximal tolerated dose, and one additional dose chosen based on interim data. Assuming a monotone dose-response relationship, at interim, criteria based on the Predictive Probability of Success are used to decide on whether to stop or to continue the trial, and, in the latter case, which dose to select for the second stage.