MEXA-CTP: Mode Experts Cross-Attention for Clinical Trial Outcome Prediction

TL;DR

This paper introduces MEXA-CTP, a lightweight multi-modal attention model that effectively predicts clinical trial outcomes by integrating diverse data types without relying on prior knowledge or wet lab data, outperforming previous methods.

Contribution

The paper presents a novel mode experts cross-attention model that leverages multi-modal data for clinical trial outcome prediction, avoiding human biases and improving accuracy over existing approaches.

Findings

MEXA-CTP achieves up to 11.3% higher F1 score than previous models.

The model improves PR-AUC by 12.2% on the TOP benchmark.

Ablation studies confirm the effectiveness of each component.

Abstract

Clinical trials are the gold standard for assessing the effectiveness and safety of drugs for treating diseases. Given the vast design space of drug molecules, elevated financial cost, and multi-year timeline of these trials, research on clinical trial outcome prediction has gained immense traction. Accurate predictions must leverage data of diverse modes such as drug molecules, target diseases, and eligibility criteria to infer successes and failures. Previous Deep Learning approaches for this task, such as HINT, often require wet lab data from synthesized molecules and/or rely on prior knowledge to encode interactions as part of the model architecture. To address these limitations, we propose a light-weight attention-based model, MEXA-CTP, to integrate readily-available multi-modal data and generate effective representations via specialized modules dubbed "mode experts", while avoiding human biases in model design. We optimize MEXA-CTP with the Cauchy loss to capture relevant interactions across modes. Our experiments on the Trial Outcome Prediction (TOP) benchmark demonstrate that MEXA-CTP improves upon existing approaches by, respectively, up to 11.3% in F1 score, 12.2% in PR-AUC, and 2.5% in ROC-AUC, compared to HINT. Ablation studies are provided to quantify the effectiveness of each component in our proposed method.