Robust group sequential designs for trials with survival endpoints and delayed response.

Area of Science:

• Oncology
• Biostatistics
• Immunology

Background:

• Randomized clinical trials in oncology commonly use time-to-event endpoints like progression-free survival and overall survival.
• The log-rank test is the standard statistical method for analyzing these endpoints, assuming proportional hazards.
• Immuno-oncology drugs, such as cancer vaccines, often exhibit delayed treatment effects, violating the proportional hazards assumption and potentially making the log-rank test suboptimal.

Purpose of the Study:

• To investigate the max-combo test as a robust alternative to the log-rank test for analyzing immuno-oncology clinical trials.
• To implement the max-combo test within a two-stage group sequential design, allowing for early stopping due to overwhelming efficacy.
• To evaluate the performance of the max-combo test using Fleming and Harrington (1981) weights and Magirr and Burman (2019) modest weights.

Main Methods:

• The study proposes and investigates the max-combo test, which combines two weighted log-rank statistics.
• The max-combo test is applied to two-stage group sequential trial designs.
• Two distinct weighting schemes, Fleming-Harrington and Magirr-Burman modest weights, are examined for the max-combo test.

Main Results:

• The max-combo test is presented as a more appropriate statistical approach when the proportional hazards assumption is violated.
• The implementation allows for early efficacy stopping in clinical trials, which is desirable for immuno-oncology agents known to prolong survival.
• The investigation explores the characteristics of the max-combo test under different weighting strategies.

Conclusions:

• The max-combo test provides a statistically sound and robust alternative for analyzing time-to-event data in immuno-oncology randomized clinical trials.
• This approach addresses the challenges posed by delayed treatment effects common in immuno-oncology.
• The developed methodology supports efficient clinical trial design by enabling early stopping for efficacy.