The Ci3+3 Design for Dual-Agent Combination Dose-Finding Clinical Trials

TL;DR

The paper introduces the Ci3+3 design, a simple, rule-based method for dual-agent dose-finding trials that simplifies the process while maintaining comparable accuracy to more complex model-based approaches.

Contribution

It extends the i3+3 design to combination trials with a two-stage, rule-based approach using a beta-binomial model for dose evaluation.

Findings

Ci3+3 shows similar operating characteristics to complex model-based designs.

The design simplifies the trial process with fixed decision rules.

Simulation results support its effectiveness in practice.

Abstract

We propose a rule-based statistical design for combination dose-finding trials with two agents. The Ci3+3 design is an extension of the i3+3 design with simple decision rules comparing the observed toxicity rates and equivalence intervals that define the maximum tolerated dose combination. Ci3+3 consists of two stages to allow fast and efficient exploration of the dose-combination space. Statistical inference is restricted to a beta-binomial model for dose evaluation, and the entire design is built upon a set of fixed rules. We show via simulation studies that the Ci3+3 design exhibits similar and comparable operating characteristics to more complex designs utilizing model-based inferences. We believe that the Ci3+3 design may provide an alternative choice to help simplify the design and conduct of combination dose-finding trials in practice.