The i3+3 design for phase I clinical trials

ABSTRACT

The traditional rule-based design, 3 + 3, has been shown to be less likely to achieve the objectives of dose-finding trials when compared with model-based designs. We propose a new rule-based design called i3 + 3, which is based on simple but more advanced rules that account for the variabilities in the observed data. We compare the operating characteristics for the proposed i3 + 3 design with other popular phase I designs by simulation. The i3 + 3 design is far superior than the 3 + 3 design in trial safety and the ability to identify the true MTD. Compared with model-based phase I designs, i3 + 3 also demonstrates comparable performances.

KEYWORDS:

• 3+3
• dose finding design
• model-based design
• rule-based design
• toxicity

Supplementary material

Supplemental data for this article can be accessed on the publisher’s website.