Comparisons of a multi-regional trial for four or five regions by fixed effect model and random effect model about allocating sample size rationally into individual regions for a multi-regional trial

Abstract

Recently, multi-regional trials have attracted much attention from sponsors as well as regulatory authorities. For researchers, it is always concerned about how to design a multi-regional trial. The fixed effect model is usually used to design a multi-regional trial. These methods were based on the assumption that true effect size is uniform across regions. However, this assumption is based on no ethnic or environmental factor among regions. In practice, it is necessary to consider the existence of ethnic or environmental factor among regions for a multi-regional trial. The random effect model is employed to deal with the heterogeneous effect size among regions. However, Most papers focus on multi-regional trials for three regions. In this paper, the multi-regional trials by a fixed effect model and a random effect model for four or five regions are designed and evaluated. The design of the multi-regional trials is evaluated by the proposed criteria considering the number of the regions. The assurance probability of proposed criteria considering the number of the regions is employed to determine if the clinical design is suitable for a multi-regional trial. More regions in a multi-regional trial can detect the variation among the regions more easily.