Efficient dose-finding for drug combination studies involving a shift in study populations

Area of Science:

• Oncology
• Clinical Trial Design
• Pharmacology

Background:

• Early-phase clinical trials are increasingly complex, investigating novel drug combinations.
• There is a need for adaptive trial designs to accommodate patient heterogeneity and evolving research questions.
• Evaluating drug combinations requires flexible methods for dose escalation and identifying maximum tolerated doses.

Purpose of the Study:

• To describe a novel model-based adaptive design for early-phase drug combination trials.
• To evaluate the safety and tolerability of entinostat in combination with capecitabine.
• To efficiently identify the maximum tolerated dose (MTD) of the combination in different breast cancer populations.

Main Methods:

• Prospective, two-part early-phase clinical trial design.
• Adaptation of a model-based approach for dose finding.
• Evaluation of entinostat plus capecitabine in metastatic breast cancer and post-neoadjuvant therapy residual disease populations.
• Utilizing operating characteristics to assess design performance.

Main Results:

• The model-based adaptive design efficiently transitions from an initial to a target patient population.
• The design accurately predicts true maximum tolerated dose combinations with high probability.
• The method allows for treatment of participants at or near optimal dose combinations using reasonable sample sizes.

Conclusions:

• The proposed adaptive design is a practical and efficient method for early-phase drug combination dose finding.
• This flexible approach addresses the need for novel designs in contemporary oncology trials.
• The design facilitates the investigation of challenging research questions, including drug combinations and patient heterogeneity.