Consistency Assessment and Regional Sample Size Calculation for MRCT Under Random Effects Model.

Area of Science:

• Clinical Trials Methodology
• Biostatistics
• Pharmaceutical Development

Background:

• Multi-regional clinical trials (MRCTs) are standard for drug development and global registration.
• Demonstrating regional consistency in treatment effects is crucial for regulatory approval.
• Existing methods based on fixed effects models may not fully capture treatment effect heterogeneity.

Purpose of the Study:

• To propose a random effects model for designing MRCTs and assessing regional consistency.
• To develop methods for calculating overall sample size and regional sample fractions.
• To provide theoretical properties for consistency probability assessment using an empirical shrinkage estimator.

Main Methods:

• Utilizing a random effects model to account for treatment effect heterogeneity across regions.
• Developing sample size determination methods for overall and regional allocation.
• Applying an empirical shrinkage estimator for consistency probability assessment, based on MHLW Method 1.
• Elaborating on applications for normal, binary, and survival endpoints.

Main Results:

• The proposed random effects model effectively retains the desired consistency probability.
• Simulation studies validate the method's performance across different endpoint types.
• The empirical shrinkage estimator aids in determining regional sample sizes of interest.

Conclusions:

• The random effects model offers a more effective approach for MRCT design and inference compared to fixed effects models.
• The presented methodology provides a robust framework for ensuring regional consistency in drug development.
• The R package facilitates the practical implementation of these advanced statistical methods in real-world trials.