Optimal sequential enrichment designs for phase II clinical trials

Area of Science:

• Clinical trial design
• Biostatistics
• Pharmacology

Background:

• Molecularly targeted agents (MTAs) often show differential efficacy across biomarker subgroups.
• Determining treatment benefit in marker-negative patients after initial success in marker-positive patients is clinically important.

Purpose of the Study:

• To propose optimal sequential enrichment (OSE) designs for phase II clinical trials.
• To address the challenge of evaluating MTAs in both biomarker-positive and marker-negative patient subgroups.

Main Methods:

• OSE designs sequentially evaluate treatment effects, starting with marker-positive patients and proceeding to marker-negative patients if necessary.
• An efficient optimization algorithm was developed to determine optimal design parameters.
• Go/no-go interim decision rules are pre-specified for practical application.

Main Results:

• OSE designs minimize expected or maximum sample size under the null hypothesis of MTA futility.
• Simulation studies demonstrate that OSE designs perform well and are ethically preferable to marker-stratified designs.
• The OSE design was successfully applied to an endometrial carcinoma clinical trial.

Conclusions:

• OSE designs provide an optimal and ethically sound framework for evaluating MTAs in phase II trials.
• These designs facilitate efficient decision-making regarding treatment benefit across biomarker subgroups.
• OSE designs enhance the practical utility and ethical considerations in early-phase drug development.