Targeting the Uniformly Most Powerful Unbiased Test in Sample Size Reassessment Adaptive Clinical Trials with Deep Learning
share
In recent pharmaceutical drug development, adaptive clinical trials become more and more appealing due to ethical considerations, and the ability to accommodate uncertainty while conducting the trial. Several methods have been proposed to optimize a certain study design within a class of candidates, but finding an optimal hypothesis testing strategy for a given design remains challenging, mainly due to the complex likelihood function involved. This problem is of great interest from both patient and sponsor perspectives, because the smallest sample size is required for the optimal hypothesis testing method to achieve a desired level of power. To address these issues, we propose a novel application of the deep neural network to construct the test statistics and the critical value with a controlled type I error rate in a computationally efficient manner. We apply the proposed method to a sample size reassessment confirmatory adaptive study MUSEC (MUltiple Sclerosis and Extract of Cannabis), demonstrating the proposed method outperforms the existing alternatives. Simulation studies are also performed to demonstrate that our proposed method essentially establishes the underlying uniformly most powerful (UMP) unbiased test in several non-adaptive designs.
READ FULL TEXT