All Times EDT
Keywords: External Trial Data, Hybrid Approach, Hybrid Control Arm, Borrowing Weights
Utilizing the external data has received increasing interest in drug development. Incorporating external data into clinical trials has mainly been using various matching methods for baseline characteristics to establish an external control arm or to augment a control arm of a randomized controlled trial (RCT). However, matching the baseline characteristics between the trial subjects and the external subjects can only guarantee that the external subjects are comparable on the baseline characteristics. The difference in outcome between the two data sources may still remain due to difference in contemporaneous and operational characteristics. Such differences are usually reflected in the outcomes data rather than in the subjects’ baseline characteristics. Similarity between the outcomes in the trial control subjects and the external subjects who have similar baseline characteristics can play a critical role in leveraging the external subjects in the clinical trials. In this paper a resampling method is proposed to establish the empirical distribution for the difference between the trial and the external outcomes. The borrowing decision is derived from this distribution and the borrowing weights for the external subjects are also established and incorporated in the weighted partial likelihood. A simulation study is conducted to evaluate the performance of the proposed borrowing method in type I error, bias, and power. An illustrative example using a simulated data is also presented.