Statistical Methodologies to Detect Ineffective Regional Treatment Effect in a Multiregional Trial

• Activity Number: 247 - Clinical Trial Design- 2
• Type: Contributed
• Date/Time: Monday, July 30, 2018 : 2:00 PM to 3:50 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Hsiao-Hui Tsou* and Yu-Chieh Cheng and Chin-Fu Hsiao
• Companies: National Health Research Institutes and National Health Research Institutes and National Health Research Institutes
• Keywords: multiregional clinical trial; Discrete Random-Effects Model; regional heterogeneity; ineffective treatment effect

Abstract

In recent years, multiregional clinical trials (MRCTs) has become a popular strategy in the development of new medicines. By incorporating subjects from many geographical regions around the world under the same protocol, MRCTs can speed up drug development and obtain faster approval of a drug globally. Traditionally, a common treatment effect across regions are assumed for the design and evaluation of MRCTs. However, regional variability has been observed and may have impact upon a medicine's effect. We may also observe an ineffective regional treatment effect during the interim analysis. The challenge is how to verify if the observed ineffective regional treatment effect is truly ineffective or by chance. In this research, we use a discrete random effects model to combine evidence of treatment effects from different regions. We explore how to incorporate the possibility of regional differences into trial planning. We propose some methodologies to determine if the observed ineffective regional treatment effect is truly ineffective or by chance. Numerical examples are given to illustrate applications of the proposed approach in different scenarios.