Designing superiority trials with window mean survival time as a primary endpoint

Area of Science:

• Clinical Trials Methodology
• Biostatistics
• Survival Analysis

Background:

• Restricted mean survival time (RMST) is a valuable metric, but its extension, Window Mean Survival Time (WMST), remains underutilized.
• Current clinical trial designs often lack familiarity with mean survival time comparisons and supporting software tools.

Purpose of the Study:

• To introduce Window Mean Survival Time (WMST) as a primary endpoint for clinical trials.
• To provide researchers with practical insights and software for designing trials utilizing WMST.
• To highlight the advantages of WMST, particularly in scenarios with non-proportional hazards.

Main Methods:

• The study focuses on the statistical properties and application of WMST.
• It introduces the 'survWMST' R package, available on GitHub, for trial design.
• The package facilitates power and sample size calculations for WMST-based trials.

Main Results:

• WMST provides interpretable estimates of treatment effect, especially beneficial when hazards are non-proportional.
• The developed R package enables robust sample size and power calculations for WMST endpoints.
• This facilitates the practical implementation of WMST in clinical trial design.

Conclusions:

• WMST is a powerful and interpretable endpoint for clinical trials, offering advantages over traditional methods.
• The availability of the 'survWMST' package addresses the need for accessible tools for trial design with WMST.
• This work aims to increase the adoption of WMST in clinical research for more effective survival outcome evaluation.