Defining Regions in Multiregional Clinical Trials: An Analytical Approach to Considering Impact of Intrinsic and...

This article describes follow-up work extended from Tanaka et al (2011) in defining the region in multiregional clinical trials (MRCTs). The previous paper advocated a systematic approach to defining regions and recommended the sponsor to think through carefully, prespecify, and justify any regional definitions as well as obtain regulatory concurrence prior to study conduct. Particular attention was advised for intrinsic and extrinsic factors such as race/ethnicity, disease epidemiology, medical practice, and geographic proximity. An analytical approach such as a cluster method to define region was introduced, and references on observed regional differences were provided.

A closer review of those references and others from a more comprehensive literature search was conducted to gain a deeper understanding of intrinsic and extrinsic factors in relation to the observed regional differences in the treatment effect. The application of a k-means cluster method to define region was explored, as well as the implication to regional sample size for MRCTs.

Of extrinsic factors, diagnostic capabilities, medical/clinical practice such as aspirin use, patient care such as site's usual practice, intervention, capability, and concomitant medications were considered potential contributors to differences in clinical outcomes or measures. Commonly reported intrinsic factors were underlying etiology, epidemiology, genetics, and patient characteristics (age, ethnicity, race, risk factors).

Using a checklist to identify intrinsic/extrinsic factors that might lead to differences in treatment effect allows one to scientifically define more meaningful regions from the identified factors, which will help with estimating the relative treatment outcome as well as exploring appropriate regional sample size.