Conditional power and predictive power based on right censored data with supplementary auxiliary information

Area of Science:

• Biostatistics
• Clinical Trial Methodology
• Statistical Inference

Background:

• Conditional power and predictive power are crucial for estimating clinical trial success at interim analyses.
• Early censoring of time-to-event endpoints can bias these power estimates, impacting trial decision-making.
• Accurate estimation is vital for effective trial management and resource allocation.

Purpose of the Study:

• To develop an enhanced statistical approach for estimating conditional and predictive power in clinical trials.
• To improve the accuracy of trial success probability estimates when dealing with right-censored time-to-event data.
• To provide a framework for more efficient futility interim analyses.

Main Methods:

• Incorporation of a fully observed auxiliary variable to augment information from the primary endpoint.
• Assumption of a bivariate normal distribution for the transformed primary variable and the auxiliary variable.
• Utilizing simulation studies to evaluate the performance of the proposed method.

Main Results:

• The proposed method significantly enhances conditional power and predictive power estimates.
• Simulation results demonstrate improved accuracy in estimators and reduced type II error inflation.
• The approach facilitates more precise determination of optimal timing for futility interim analyses.

Conclusions:

• Integrating auxiliary variables offers a robust solution to bias introduced by censoring in time-to-event trials.
• The method provides a more reliable basis for go/no-go decisions at interim analyses.
• This approach can lead to more efficient and accurate clinical trial designs and conduct.