Improved inference for MCP-Mod approach for time-to-event endpoints with small sample sizes

TL;DR

This paper enhances the MCP-Mod approach for small sample size time-to-event data by developing improved ML estimators that reduce bias and better control type I error, validated through simulation studies.

Contribution

It introduces improved ML estimators for censored Weibull regression, tailored for small samples in MCP-Mod, ensuring accurate inference and error control.

Findings

Improved ML estimators are less biased than standard ML estimators.

Wald-type statistics with improved estimators maintain nominal type I error.

Recommended for sample sizes 5 to 25 subjects per dose.

Abstract

The Multiple Comparison Procedures with Modeling Techniques (MCP-Mod) framework has been recently approved by the U.S. Food and Administration and European Medicines Agency as fit-per-purpose for phase II studies. Nonetheless, this approach relies on the asymptotic properties of Maximum Likelihood (ML) estimators, which might not be reasonable for small sample sizes. In this paper, we derived improved ML estimators and correction for their covariance matrices in the censored Weibull regression model based on the corrective and preventive approaches. We performed two simulation studies to evaluate ML and improved ML estimators with their covariance matrices in (i) a regression framework (ii) the Multiple Comparison Procedures with Modeling Techniques framework. We have shown that improved ML estimators are less biased than ML estimators yielding Wald-type statistics that controls type I error without loss of power in both frameworks. Therefore, we recommend the use of improved ML estimators in the MCP-Mod approach to control type I error at nominal value for sample sizes ranging from 5 to 25 subjects per dose.