Great Wall: A Generalized Dose Optimization Design for Drug Combination Trials Maximizing Survival Benefit

Area of Science:

• Clinical Trials
• Pharmacology
• Biostatistics

Background:

• Many phase I-II drug-combination trials assume early optimal dose selection ensures long-term survival.
• This assumption is frequently contradicted by high relapse rates post-initial response in clinical studies.

Purpose of the Study:

• To introduce the Great Wall design, a novel dose optimization strategy for drug-combination trials.
• To address the challenge of maximizing long-term survival benefits by optimizing dose combinations.

Main Methods:

• The Great Wall design uses a 'divide-and-conquer' algorithm for partial order of toxicity.
• It employs early outcomes to eliminate suboptimal dose combinations and dose randomization to create a candidate set.
• Patients in the candidate set are monitored for survival outcomes to select the final optimal dose combination.

Main Results:

• Simulation studies demonstrated favorable operating characteristics of the Great Wall design across diverse clinical scenarios.
• The design effectively balances toxicity and early efficacy to identify dose combinations maximizing survival benefits.

Conclusions:

• The Great Wall design offers a practical and modular approach for drug-combination trials prioritizing long-term survival.
• Its utility is particularly relevant when extended patient follow-up for survival assessment is feasible.