A Method for Ensuring a Consistent Dose-Response Relationship Between An Entire Population and One Region in Multiregional Dose-Response Studies Using MCP-Mod

Multiregional clinical studies are frequently conducted worldwide to accelerate drug approval. Conducting multiregional dose-response studies is useful to determine the effects of important ethnic factors on a drug during the early stages of development. Sample size allocation for regions should be based on a scientific rationale. Recently, Yamaguchi and Sugitani proposed a method for sample size allocation for multiregional dose-response studies using MCP-Mod. This method ensures that a consistent dose-response relationship in the contrast test statistic and the maximum dose-response curve difference between the entire population and one region are observed with sufficient probability. However, this method is not easy to use because of the inability to compute the assurance probability of consistency. In this study, we propose an analytical solution by defining the probability formulas for the existing method. The difficulty lies in formulating the probability of maximum curve differences between two nested populations: the entire population and one specific region. We addressed this difficulty by approximating the curve of the entire population by combination of the independent regional curves. The proposed formula enables the computation of the probability based on a multivariate normal distribution. The accuracy of the proposed formulas was confirmed through a simulation study.