Commensurate Priors for Incorporating Historical Information in Clinical Trials

• Activity Number: 70
• Type: Topic Contributed
• Date/Time: Sunday, July 29, 2012 : 4:00 PM to 5:50 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Brian Hobbs*+ and Daniel J. Sargent and Bradley P Carlin
• Companies: MD Anderson Cancer Center and Mayo Clinic and University of Minnesota
• Keywords: Meta-analysis ; Bayesian analysis ; Clinical trials

Abstract

Assessing between-study variability in the context of conventional random-effects meta-analysis is notoriously difficult when incorporating data from only a small number of historical studies. In order to borrow strength, historical and current data are often assumed to be fully homogeneous a priori, but this can have drastic consequences for power and Type I error if the historical information is biased. In this paper, we propose empirical and fully Bayesian modifications of the commensurate prior model (Hobbs et al. 2011) extending Pocock (1976), and evaluate their frequentist and Bayesian properties for incorporating patient-level historical data. Our proposed commensurate prior models lead to admissible estimators that facilitate alternative bias-variance trade-offs than those offered by pre-existing methodologies for incorporating historical data from a small number of historical studies. We also provide a sample analysis of a colon cancer trial comparing time-to-disease progression rates using a Weibull regression model.