Sequential monitoring of response-adaptive randomized clinical trials with sample size re-estimation

Response-adaptive randomization can be used in phase III confirmatory trials to achieve various aims related to ethics and efficiency such as minimizing the total number of failures and maximizing the power by targeting different optimal allocation proportions. Sequential monitoring and sample size re-estimation are also popular and are often required in modern clinical trials. In this paper, we propose to combine these three adaptive procedures to sequentially monitor response-adaptive randomized clinical trials with sample size re-estimation. We obtain the asymptotic results for this sequential procedure under certain conditions and find that the type I error rate can be controlled asymptotically with commonly used stopping boundaries from traditional group sequential designs. Assuming continuous or binary endpoints, we show in simulations that the proposed method can achieve various objectives while maintaining the type I error rate on finite samples. In addition, we include a case study to demonstrate how the method can be used in practice to permit ethical considerations while allowing sample size re-estimation to account for uncertainties in the treatment effect. (C) 2019 Elsevier B.V. All rights reserved.