All Times EDT
Keywords: Dynamic borrowing, bayesian statistics, historical data, meta-analytic-predictive prior
Clinical trials often encounter enrollment challenges. Borrowing historical data to the current trial may alleviate these challenges by reducing the number of participants and shortening trial duration while increasing statistical efficiency. However, naive borrowing can result in biased estimates and inflate false positives rate (FPR). We proposed a machine learning-based dynamic borrowing approach – Regularized Bayes (RB) to ensure efficient borrowing via minimizing mean square error (MSE) to optimize the balance between bias and uncertainty. A regularization term was used to link the current study with historical data to dynamically calibrate the degree of borrowing. The final estimate based on RB is anticipated to improve the MSE based on independent analysis (IND) of the current study data (i.e., no borrowing). The simulation results showed that the RB had better adaptivity of discounting historical information with higher between-trial heterogeneity. Thus, the RB had the optimal balance between bias and uncertainty, leading to the lowest MSE, achieving up to 55% lower MSE relative to IND. In addition, the RB improved true positive rate (TPR) over Meta-analytic prior and IND, at about 15% to 30%; and simultaneously provided added protection against trial heterogeneity similarly to conservative approaches Robust meta-analytic prior and Commensurate prior.