BaySize: Bayesian Sample Size Planning for Phase I Dose-Finding Trials

TL;DR

BaySize is a novel Bayesian sample size calculator designed for phase I dose-finding trials, incorporating effect size and equivalence intervals, with practical lookup tables for diverse trial designs.

Contribution

It introduces the first versatile Bayesian sample size tool applicable to a wide range of phase I trial designs, using a dual-prior decision framework.

Findings

Provides practical lookup tables for trial planning

Enables sample size calculation with desired statistical power

Applicable to various phase I dose-finding designs

Abstract

We propose BaySize, a sample size calculator for phase I clinical trials using Bayesian models. BaySize applies the concept of effect size in dose finding, assuming the MTD is defined based on an equivalence interval. Leveraging a decision framework that involves composite hypotheses, BaySize utilizes two prior distributions, the fitting prior (for model fitting) and sampling prior (for data generation), to conduct sample size calculation under desirable statistical power. Look-up tables are generated to facilitate practical applications. To our knowledge, BaySize is the first sample size tool that can be applied to a broad range of phase I trial designs.