Designing cancer immunotherapy trials with random treatment time-lag effect

Area of Science:

• Clinical Trials Methodology
• Biostatistics
• Cancer Research

Background:

• Cancer immunotherapy trials may exhibit treatment time-lag effects where patient response varies.
• Individual heterogeneity in lag duration necessitates specialized study designs and analysis.

Purpose of the Study:

• To develop a robust statistical test and design for clinical trials accounting for random time-lag effects.
• To enhance study power and provide accurate sample size/power calculations for delayed treatment effects.

Main Methods:

• Introduction of the Generalized Piecewise Weighted Logrank (GPW-Logrank) test.
• Development of analytic and numeric approaches for sample size and power calculations.
• Derivation of asymptotic properties and evaluation of finite sample efficiency via simulations.

Main Results:

• The GPW-Logrank test demonstrates substantial efficiency gains over standard methods that ignore delayed effects.
• Proposed approaches are more robust than fixed time-lag methods when the lag model is misspecified.
• An R package (DelayedEffect.Design) is available for practical implementation.

Conclusions:

• The GPW-Logrank test offers a powerful and robust solution for analyzing trials with random time-lag effects.
• This methodology is crucial for optimizing study design and analysis in cancer immunotherapy and similar fields.
• The developed R package facilitates the application of these advanced statistical techniques.