On selecting the critical boundary functions in group-sequential trials with two time-to-event outcomes

Area of Science:

• Biostatistics
• Clinical Trial Design
• Statistical Methods

Background:

• Group-sequential clinical trials often involve multiple endpoints.
• Selecting appropriate boundary functions is crucial for maintaining trial integrity and statistical power.
• Challenges arise when endpoints have unequal effect sizes or different evaluation durations.

Purpose of the Study:

• To investigate the selection of critical boundary functions for two time-to-event outcomes in group-sequential trials.
• To evaluate scenarios with unequal endpoint effect sizes or short-term versus long-term evaluations.
• To provide guidance on statistical power considerations.

Main Methods:

• The study considers Bonferroni-Holm and fixed-sequence procedures.
• It assesses the impact of hazard ratios and endpoint correlation on statistical power.
• Simulation or theoretical analysis of boundary function selection was performed.

Main Results:

• The choice of boundary function significantly affects statistical power, especially with unequal endpoint characteristics.
• Hazard ratio magnitudes and inter-endpoint correlation are key factors influencing power.
• Specific procedures (Bonferroni-Holm, fixed-sequence) demonstrate varying performance.

Conclusions:

• Careful selection of boundary functions is essential for robust group-sequential trials with multiple time-to-event outcomes.
• Guidance is provided for optimizing trial design based on endpoint characteristics and desired statistical power.
• Understanding the interplay between effect sizes, evaluation periods, and correlations is critical for successful trial planning.