Past

Robust estimates of treatment effects in MRCT via semi-parametric models

Multiregional randomized clinical trials (MRCT) are increasingly common in drug development. With China joining ICH in 2016 and the publication of E17, it is of current and long-term interest to resolve statistical issues in MRCT. In this talk I will highlight some key statistical issues, e.g., increased heterogeneity in trials involving different regions in the world. So accurate estimate of variance would be important to obtain more accurate sample size estimate. In addition, despite of randomization, there may be a differential treatment effect among different regions, potentially due to confounding region-specific factors. Most current methods for the assessment of the consistency or similarity of the treatment effect between different ethnic groups based on some subjectively specified model. We then propose a novel semi-parametric model and show that it can give robust estimates of the (regional) treatment effects. The model is estimated by maximizing profile likelihood using EM algorithm. The profile likelihood ratio statistic is used to test the existence of regional differences. We derived the asymptotic properties of the estimate and show that semiparametric model performs well by simulation. We then discuss applications to two clinical trials

(This work is in collaboration with Ao Yuan, Yizhao Zhou and Shuxin Wang).