Modified Robust Meta-Analytic-Predictive Priors for Incorporating Historical Controls in Clinical Trials

Abstract

Incorporating historical information in clinical trials has been of much interest recently because of its potential to reduce the size and cost of clinical trials. Data-conflict is one of the biggest challenges in incorporating historical information. In order to address the conflict between historical data and current data, several methods have been proposed including the robust meta-analytic-predictive (rMAP) prior method. In this article, we propose to modify the rMAP prior method by using an empirical Bayes approach to estimate the weights for the two components of the rMAP prior. Via numerical calculations, we show that this modification to the rMAP method improves its performance regarding multiple key metrics.

KEYWORDS:

• Clinical trials
• Dynamic borrowing
• Empirical bayes
• Historical control
• Mixture distribution

Supplementary Materials

S1. Weight parameter(s) in the posterior distribution given a mixture of Beta prior distributions.

S2. Variance of a parameter regarding to its posterior distribution given a mixture prior distribution.

Acknowledgments

The authors would like to thank the reviewers for their valuable comments which have helped improve this manuscript.

Disclosure Statement

The authors report there are no competing interests to declare.

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.