A Bayesian Noninferiority Approach to Evaluation of Bridging Studies

Abstract

A bridging study defined by The International Conference on Harmonization E5 is usually conducted in the new region only after the test product has been approved for commercial marketing in the original region due to its proven efficacy and safety. In this paper, we address the issue of analysis of clinical data generated by the bridging study conducted in the new region to evaluate the similarity for extrapolation of the foreign clinical data to the population of the new region. Information on efficacy, safety, dosage, and dose regimen of the original region cannot be concurrently obtained from the local bridging studies but are available in the trials conducted in the original region. A Bayesian noninferiority approach is therefore proposed to incorporate the data generated in the original region to evaluate bridging evidence by the local bridging studies and assess similarity between the new and original regions. Methods for sample size determination for the bridging study are also proposed.