UNITED: A Unified Transparent and Efficient Phase I/II Trial Design for Dose Optimization Accounting for Ordinal...

A primary objective of oncology dose-finding trials for novel therapies is to determine an optimal biological dose (OBD) that is both tolerable and therapeutically beneficial for patients in subsequent clinical trials. These new therapeutic agents are more likely to induce multiple low- or moderate-grade toxicities rather than dose-limiting toxicities. Additionally, efficacy is evaluated comprehensively, differentiating between complete remission and partial remission, as well as incorporating continuous efficacy endpoints. This important issue was highlighted in the American Statistical Association (ASA) Biopharmaceutical (BIOP) Section open forums and was a significant consideration of the FDA's "Project Optimus." We proposed the UNITED design, a unified, transparent, and efficient Phase I/II trial design to incorporate toxicity and efficacy grades and types, as well as continuous efficacy responses, into dose-finding and optimization. The UNITED design can handle binary, quasi-binary, continuous, and mixed toxicity and efficacy endpoints. We further extended the UNITED design, referred to as TITE-UNITED, to accommodate delayed toxicity and efficacy outcomes. Simulation studies showed that the UNITED and TITE-UNITED designs have desirable operating characteristics, performing comparably to or better than existing designs. A user-friendly software is available for practical implementation.