Testing multiple primary endpoints in clinical trials with sample size adaptation.

Area of Science:

• Clinical trial design
• Biostatistics
• Pharmaceutical research

Background:

• Accelerated approval pathways require robust statistical designs.
• Balancing interim and final analysis endpoints is crucial for drug development.
• Sample size adaptation methods are essential for efficient clinical trials.

Purpose of the Study:

• To propose a novel clinical trial design for drug approval using dual endpoints.
• To compare sample size adaptation rules for improved trial efficiency and bias control.
• To evaluate different testing procedures for controlling family-wise error rate.

Main Methods:

• A two-stage trial design with interim and final analyses.
• Implementation of sample size adaptation based on long-term endpoints.
• Comparison of two adaptation rules: conditional power maintenance and step-function.
• Evaluation of three testing procedures: alpha splitting and two alpha exhaustive methods.

Main Results:

• The proposed designs strongly control the family-wise error rate across dual endpoints and sample size adaptation.
• Alpha exhaustive designs significantly improve statistical power when both endpoints are effective.
• The choice of sample size adaptation rule has minimal impact on power.

Conclusions:

• The proposed dual-endpoint design with sample size adaptation offers a flexible and powerful approach for drug approval.
• Alpha exhaustive testing procedures are recommended for maximizing power in such designs.
• The methodology is extendable to more complex clinical trial settings.