On Sample Size Determination for Augmented Tests Based on Restricted Mean Survival Time in Randomized Clinical Trials

Area of Science:

• Biostatistics
• Clinical Trials
• Survival Analysis

Background:

• Restricted Mean Survival Time (RMST) is increasingly used to assess treatment effects in clinical trials.
• Existing sample size methods for RMST tests often require specifying full survival curves, which can be challenging.
• Augmented RMST tests offer improved efficiency and power by incorporating baseline covariates.

Purpose of the Study:

• To propose a sample size formula for augmented RMST tests.
• To address the need for sample size determination without assuming specific survival functions.
• To develop a sample size recalculation method for augmented RMST tests using blinded data.

Main Methods:

• Developed an approximated sample size formula for augmented RMST tests.
• Proposed a sample size recalculation approach to handle correlations between covariates and martingale residuals.
• The method does not require specifying the entire survival curve in the treatment group.

Main Results:

• The proposed formula provides a practical approach for sample size calculation in augmented RMST tests.
• The recalculation method allows for updating sample size estimates with blinded data.
• Enables studies to achieve target power for RMST differences even with unspecified survival functions.

Conclusions:

• The new methodology facilitates more robust sample size planning for augmented RMST tests.
• This approach enhances the efficiency and power of clinical trials utilizing RMST.
• It provides a valuable tool for researchers designing survival studies with complex covariate adjustments.