A Bayesian phase I/II design for cancer clinical trials combining an immunotherapeutic agent with a chemotherapeutic agent

Immunotherapy is an innovative treatment approach that harnesses a patient’s immune system to treat cancer. It has provided an alternative and complementary treatment modality to conventional chemotherapy. Combining immunotherapy with cytotoxic chemotherapy agent has become the leading trend and the most active research field in oncology. To accommodate this growing trend, we propose a Bayesian phase I/II dose-finding design to identify the optimal biological dose combination (OBDC), defined as the dose combination with the highest desirability in the risk-benefit trade-off. We propose new statistical models to describe the relationship between the doses and treatment outcomes, including immune response, toxicity and progression-free survival (PFS). During the trial, based on accrued data, we continuously update model estimates and adaptively assign patients to dose combinations with high desirability. The simulation study shows that our design has desirable operating characteristics.