A hierarchical Bayesian design for randomized Phase II clinical trials with multiple groups.

Area of Science:

• Oncology
• Biostatistics
• Clinical Trial Design

Background:

• Advances in cancer biology and genetics are refining treatment targets.
• Targeted therapies in oncology often address common signaling pathways across diverse malignancies.
• Information sharing across tumor types is crucial for developing novel therapeutic agents.

Purpose of the Study:

• To propose a hierarchical Bayesian design (HBD) for simultaneously evaluating novel therapeutic agents in multiple cancer groups.
• To enhance efficiency and reduce financial costs in randomized Phase II clinical trials.

Main Methods:

• Utilized a hierarchical Bayesian design (HBD) for multi-group randomized Phase II trials with binary endpoints.
• Developed an R package (hbdct) to implement the HBD and facilitate sample size calibration.

Main Results:

• The HBD demonstrated a significant reduction in sample size compared to traditional parallel designs for individual tumor groups.
• The proposed design improves the overall efficiency of randomized Phase II clinical trials.

Conclusions:

• The hierarchical Bayesian design offers a more efficient and cost-effective approach for testing novel oncology agents across multiple tumor types.
• The hbdct R package provides a practical tool for implementing this advanced clinical trial methodology.