Sample Size Allocation to Regions in a Multiregional Trial

Hiroyuki Uesakaa* a The Center for Advanced Medical Engineering and Informatics , Osaka University , Osaka, Japan * Department of Biomedical Statistics, Osaka University, Graduate School of Medicine, 2-2 Yamadaoka, Suita, Osaka 565-0871, Japan E-mail: uesaka@medstat.med.osaka-u.ac.jp The objective of a multiregional bridging trial is to show the efficacy of a drug in various global regions, and at the same time to evaluate the possibility of applying the overall trial results to each region. However, to apply overall results to a specific region, the result in that region should be consistent with either the overall results or the results of other regions. This article discusses methods of sample size allocation to regions by introducing statistical criteria for consistency between regional and overall results. Specifically, three rules of sample size allocation are discussed: (1) allocating equal size to all regions, (2) minimizing total sample size, and (3) minimizing the sample size of a specific region. Some total and regional sample sizes calculated under each allocation rule are illustrated.