Adaptive randomized phase II design for biomarker threshold selection and independent evaluation

Area of Science:

• Clinical Trial Design
• Biostatistics
• Translational Medicine

Background:

• Biomarker validation for patient selection in phase II trials is often incomplete at study initiation.
• Retrospective biomarker analysis can lead to over-enrollment of non-responders, underpowered studies, and suboptimal patient care.

Purpose of the Study:

• Propose an adaptive randomized phase II study design for prospective biomarker threshold identification.
• Incorporate early futility stopping, mid-trial accrual adjustments based on biomarker status, and focused final analyses.

Main Methods:

• An interim analysis guides decisions on futility stopping, continuation without a marker, or adaptive accrual/resizing based on a promising biomarker.
• Final efficacy analyses are conducted in the identified target population most likely to benefit.
• Simulation studies assess error rates, power, and sample size efficiency.

Main Results:

• The adaptive design effectively identifies predictive biomarkers and restricts accrual to biomarker-positive patients.
• Error rates (Type I and Type II) are controlled through threshold-based prevalence restriction and optimal interim analysis timing.

Conclusions:

• This adaptive design provides a solution for identifying and validating continuous biomarkers within randomized phase II trials.
• It enhances the precision of identifying patient populations most likely to benefit from experimental therapies.