uTPI: A utility-based toxicity probability interval design for phase I/II dose-finding trials.

Area of Science:

• Clinical Trials
• Biostatistics
• Pharmacology

Background:

• Traditional chemotherapy dose-finding relies on maximum tolerated dose (MTD), which may not apply to molecularly targeted agents and immunotherapies.
• Phase I/II trials integrating toxicity and efficacy can identify more clinically meaningful doses than toxicity-based Phase I trials.
• Current methods for dose optimization in early-phase trials can be complex and may not fully capture the risk-benefit tradeoff.

Purpose of the Study:

• To introduce a novel utility-based toxicity probability interval (uTPI) design for optimizing biological dose selection in clinical trials.
• To provide a simplified and robust method for identifying optimal doses that balance toxicity and efficacy for novel therapeutics.
• To develop a design that avoids parametric assumptions about dose-response relationships and directly models dose desirability.

Main Methods:

• The utility-based toxicity probability interval (uTPI) design uses a numerical utility to summarize bivariate toxicity and efficacy outcomes.
• It employs a quasi-binomial likelihood to model dose desirability without parametric dose-response assumptions.
• Dose escalation/de-escalation decisions are adaptive, based on posterior desirability distributions and screening of overly toxic doses using toxicity probability intervals.

Main Results:

• The uTPI design demonstrates desirable and robust performance across various simulation scenarios.
• It offers a flexible framework accommodating diverse dose desirability formulations with minimal design parameters.
• The design provides a clear decision structure, enabling pre-calculation of a dose-assignment table for simplified trial implementation.

Conclusions:

• The proposed uTPI design offers a strengthened and simplified approach to phase I/II clinical trial dose finding.
• It effectively identifies optimal biological doses by integrating efficacy and toxicity, crucial for targeted agents and immunotherapies.
• The uTPI design enhances the practical implementation of clinical trials by providing a clear, adaptive decision-making process.