Matching Patients to Clinical Trials with Large Language Models

TL;DR

TrialGPT is a comprehensive large language model-based framework that significantly improves the accuracy and efficiency of matching patients to clinical trials through retrieval, matching, and ranking modules, reducing screening time.

Contribution

This paper introduces TrialGPT, a novel end-to-end LLM-based system for zero-shot patient-trial matching, combining retrieval, matching, and ranking to outperform existing methods.

Findings

Recall over 90% of relevant trials with minimal initial data

Achieves 87.3% accuracy in patient-criterion matching

Reduces patient screening time by 42.6%

Abstract

Patient recruitment is challenging for clinical trials. We introduce TrialGPT, an end-to-end framework for zero-shot patient-to-trial matching with large language models. TrialGPT comprises three modules: it first performs large-scale filtering to retrieve candidate trials (TrialGPT-Retrieval); then predicts criterion-level patient eligibility (TrialGPT-Matching); and finally generates trial-level scores (TrialGPT-Ranking). We evaluate TrialGPT on three cohorts of 183 synthetic patients with over 75,000 trial annotations. TrialGPT-Retrieval can recall over 90% of relevant trials using less than 6% of the initial collection. Manual evaluations on 1,015 patient-criterion pairs show that TrialGPT-Matching achieves an accuracy of 87.3% with faithful explanations, close to the expert performance. The TrialGPT-Ranking scores are highly correlated with human judgments and outperform the best-competing models by 43.8% in ranking and excluding trials. Furthermore, our user study reveals that TrialGPT can reduce the screening time by 42.6% in patient recruitment. Overall, these results have demonstrated promising opportunities for patient-to-trial matching with TrialGPT.