A response-adaptive covariate-balanced randomization for multi-arm clinical trials

Randomization is a key characteristic of clinical trials which makes them the gold standard for determining treatment effectiveness. Response-adaptive randomization is desirable because it allows more patients to receive the best treatment; however, compared to traditional equal randomization, response-adaptive randomization is more likely to allow imbalance in prognostic baseline covariates. We propose a simple yet flexible two-stage randomization for multi-arm trials which marries response-adaptative and covariate- balancing designs. The operating characteristics of the proposed methods were assessed via simulation for a variety of scenarios in which values of treatment success probability and patient response delay time were varied.