Some notable properties of the standard oncology phase I design

TL;DR

This paper explores fundamental properties of the standard 3+3 oncology phase I trial design, revealing its implicit use of isotonic regression, its relation to Bayesian methods, and its behavior in worst-case scenarios.

Contribution

It uncovers the mathematical and statistical properties of the standard design, including its connection to Bayesian approaches and its performance limits.

Findings

Standard design implicitly uses isotonic regression.

A Bayesian design can replicate the standard design with slight modifications.

The standard design's behavior in worst-case scenarios is characterized and compared.

Abstract

We identify three properties of the standard oncology phase I trial design or 3 + 3 design. We show that the standard design implicitly uses isotonic regression to estimate a maximum tolerated dose. We next illustrate the relationship between the standard design and a Bayesian design proposed by Ji et al. (2007). A slight modification to this Bayesian design, under a particular model specification, would assign treatments in a manner identical to the standard design. We finally present calculations revealing the behavior of the standard design in a worst case scenario and compare its behavior with other 3 + 3-like designs.