An optimal adaptive design to address local regulations in global clinical trials

Pharm Stat. 2010 Jul-Sep;9(3):179-89. doi: 10.1002/pst.456.

Abstract

After multi-regional clinical trials (MRCTs) have demonstrated overall significant effects, evaluation for a region-specific effect is often important. Recent guidance from regulatory authorities regarding evaluation for possible country-specific effects has led to research on statistical designs that incorporate such evaluations in MRCTs. These statistical designs are intended to use the MRCTs to address the requirements for global registration of a medicinal product. Adding a regional requirement could change the probability for declaring positive effect for the region when there is indeed no treatment difference as well as when there is in fact a true difference within the region. In this paper, we first quantify those probability structures based on the guidance issued by the Ministry of Health, Labour and Welfare (MHLW) of Japan. An adaptive design is proposed to consider those probabilities and to optimize the efficiency for regional objectives. This two-stage approach incorporates comprehensive global objectives into an integrated study design and may mitigate the need for a separate local bridging study. A procedure is used to optimize region-specific enrollment based on an objective function. The overall sample size requirement is assessed. We will use simulation analyses to illustrate the performance of the proposed study design.

MeSH terms

  • Drug Approval / legislation & jurisprudence
  • Drug Approval / statistics & numerical data
  • Geography
  • Guidelines as Topic
  • Humans
  • Internationality / legislation & jurisprudence*
  • Multicenter Studies as Topic / legislation & jurisprudence*
  • Multicenter Studies as Topic / statistics & numerical data*
  • Randomized Controlled Trials as Topic / legislation & jurisprudence*
  • Randomized Controlled Trials as Topic / statistics & numerical data*
  • Research Design
  • Sample Size
  • Small-Area Analysis