|
Abstract:
|
Bayesian methods can help in the analysis of clinical trials, by adding proper prior information into the analysis thereby potentially decreasing required sample size. We develop proper prior information for the analysis of a phase III trial for showing that a proposed biosimilar is similar to a reference biologic. For the reference product, we use a meta-analysis of published results to set a prior, and we propose priors for the proposed biosimilar informed by the strength of the evidence generated in the earlier steps of the approval process. As part of our development, we propose methods to aid in the interpretation of our priors for (i) understanding the accuracy of the approximations and (ii) assessing the influence of the priors on the posteriors. A simulation study demonstrates the impacts of the priors as compared to a non-Bayesian analysis and shows that except in extreme combinations, the Bayesian relative risk analysis provides 90% credible intervals that have more than 90% frequentist coverage, are much shorter than the 90% frequentist intervals and the Bayesian posterior mean has substantially smaller mean squared error than the usual frequentist estimate.
|
