Bayesian order constrained adaptive design for phase II clinical trials evaluating subgroup-specific treatment...

Area of Science:

• Biostatistics
• Clinical Trial Design
• Biomarker Research

Background:

• Biotherapies often exhibit heterogeneous treatment effects across patient subgroups defined by biomarkers.
• Traditional phase II trial designs are ill-suited for evaluating these subgroup-specific effects.
• Accurate identification of treatment effects in specific subgroups is crucial for personalized medicine.

Purpose of the Study:

• To propose a novel Bayesian adaptive phase II clinical trial design for biotherapies.
• To enhance the detection of subgroup-specific treatment effects guided by biomarkers.
• To improve the efficiency and cost-effectiveness of phase II biotherapy trials.

Main Methods:

• Introduced the Bayesian-order constrained adaptive design (BOCAD), integrating enrichment and sequential design features.
• The BOCAD design incorporates an initial "all-comers" stage followed by a biomarker-guided enrichment stage based on interim analysis.
• Developed go/no-go enrichment criteria using posterior probabilities and extended the design to handle missing biomarker data.

Main Results:

• Comprehensive simulations demonstrated the superior performance of the BOCAD compared to existing and conventional designs.
• The BOCAD design effectively detected subgroup-specific treatment effects.
• The design demonstrated a favorable balance between trial efficiency and cost-effectiveness.

Conclusions:

• The Bayesian-order constrained adaptive design is a highly effective and efficient approach for phase II biotherapy trials.
• This adaptive design optimizes the evaluation of subgroup-specific treatment effects in biomarker-guided studies.
• The proposed methodology offers a robust framework for personalized medicine development.