Group-sequential three-arm noninferiority clinical trial designs

Area of Science:

• Clinical Trials Methodology
• Biostatistics
• Pharmaceutical Research

Background:

• Noninferiority clinical trials are essential for evaluating new treatments against existing ones.
• Three-arm trials with active and placebo controls are increasingly used to assess both treatment efficacy and noninferiority.
• Group-sequential designs allow for early stopping and adaptive decision-making in clinical trials.

Purpose of the Study:

• To extend existing fixed margin and fraction approaches for noninferiority trials to a group-sequential setting.
• To develop and evaluate two distinct decision-making frameworks for group-sequential three-arm noninferiority trials.
• To investigate the impact of various design factors on trial operating characteristics, including power, Type I error, and sample size.

Main Methods:

• Adaptation of fixed margin and fraction noninferiority approaches into a group-sequential framework.
• Development of two decision-making frameworks for group-sequential three-arm trials.
• Simulation studies to analyze operating characteristics (power, Type I error, sample size) under varying design factors.
• Assessment of sample size recalculation's effect on trial outcomes.

Main Results:

• The study details the operating characteristics of the proposed group-sequential designs.
• Analysis reveals how design factors influence power, Type I error rates, and expected/maximum sample sizes.
• Simulation results demonstrate the impact of sample size recalculation on trial efficiency and accuracy.

Conclusions:

• The proposed group-sequential designs offer flexible frameworks for three-arm noninferiority trials.
• Understanding the interplay of design factors is crucial for optimizing trial operating characteristics.
• Sample size recalculation can be effectively integrated to improve trial efficiency while maintaining statistical integrity.