DrugReasoner: Interpretable Drug Approval Prediction with a Reasoning-augmented Language Model

2025-08-25

AI Summary

Drug approval prediction faces challenges due to the poor interpretability of conventional AI models, limiting their utility in clinical decision support. To address this, we propose a reasoning-augmented large language model built upon the LLaMA architecture, which jointly incorporates molecular descriptors and contrastive reasoning over structurally similar compounds. We introduce group-relative policy optimization (GRPO) for fine-tuning and generate interpretable predictions accompanied by stepwise reasoning chains and confidence scores. This approach overcomes the black-box limitation of prior methods. It achieves robust performance on the validation set (AUC = 0.732, F1 = 0.729) and test set (AUC = 0.725, F1 = 0.718), and generalizes effectively to an external cohort (AUC = 0.728, F1 = 0.774), significantly outperforming baseline models such as ChemAP. The method demonstrates strong generalizability and translational potential for real-world drug development applications.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: (Large) Language Models
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Drug discovery is a complex and resource-intensive process, making early prediction of approval outcomes critical for optimizing research investments. While classical machine learning and deep learning methods have shown promise in drug approval prediction, their limited interpretability constraints their impact. Here, we present DrugReasoner, a reasoning-based large language model (LLM) built on the LLaMA architecture and fine-tuned with group relative policy optimization (GRPO) to predict the likelihood of small-molecule approval. DrugReasoner integrates molecular descriptors with comparative reasoning against structurally similar approved and unapproved compounds, generating predictions alongside step-by-step rationales and confidence scores. DrugReasoner achieved robust performance with an AUC of 0.732 and an F1 score of 0.729 on the validation set and 0.725 and 0.718 on the test set, respectively. These results outperformed conventional baselines, including logistic regression, support vector machine, and k-nearest neighbors and had competitive performance relative to XGBoost. On an external independent dataset, DrugReasoner outperformed both baseline and the recently developed ChemAP model, achieving an AUC of 0.728 and an F1-score of 0.774, while maintaining high precision and balanced sensitivity, demonstrating robustness in real-world scenarios. These findings demonstrate that DrugReasoner not only delivers competitive predictive accuracy but also enhances transparency through its reasoning outputs, thereby addressing a key bottleneck in AI-assisted drug discovery. This study highlights the potential of reasoning-augmented LLMs as interpretable and effective tools for pharmaceutical decision-making.

Problem

• Predicts small-molecule drug approval likelihood accurately
• Enhances interpretability in AI-assisted drug discovery
• Generates step-by-step rationales with confidence scores

Innovation

• LLM with reasoning for drug approval prediction
• Integrates molecular descriptors and comparative analysis
• Generates step-by-step rationales and confidence scores