Group sequential design for time-to-event data using the concept of proportional time

Area of Science:

• Clinical Trials Methodology
• Biostatistics
• Survival Analysis

Background:

• Sequential monitoring is crucial in clinical trials for efficacy and safety.
• Existing group sequential designs often assume proportional hazards or exponential lifetimes, limiting their use with complex time-to-event data.
• Traditional sample size calculations for survival data are restrictive when underlying assumptions are violated.

Purpose of the Study:

• To propose a novel group sequential design for time-to-event outcomes that addresses limitations of existing methods.
• To develop efficient sample size calculation methods for clinical trials with survival data.
• To provide a flexible design accommodating various survival data complexities.

Main Methods:

• Introduced a novel group sequential design based on the concept of Proportional Time.
• Utilized the generalized gamma ratio distribution for calculating efficacy and safety boundaries.
• Employed an error spending approach compatible with various distributions within the generalized gamma family.

Main Results:

• The proposed Proportional Time method offers a flexible framework for group sequential designs with time-to-event data.
• The design effectively handles complexities such as loss to follow-up, censoring, and varying accrual patterns.
• Demonstrated applicability through three practical examples, highlighting its advantages over traditional methods.

Conclusions:

• The novel group sequential design using Proportional Time provides a robust and flexible alternative for clinical trials with time-to-event outcomes.
• This method improves the efficiency and applicability of sample size calculations when standard assumptions are not met.
• The design's features enhance ethical, financial, and administrative considerations in trial conduct.