Predicting Targeted Therapy Resistance in Non-Small Cell Lung Cancer Using Multimodal Machine Learning

2025-03-31

AI Summary

Late-stage EGFR-mutant non-small cell lung cancer (NSCLC) patients lack reliable predictive tools for osimertinib resistance. Method: We propose the first interpretable multimodal machine learning framework integrating histopathological images, NGS-based genomic variants, clinical phenotypes, and demographic data. Our approach combines CNN and Transformer-based feature extraction, cross-modal alignment, survival modeling, and SHAP-based interpretability analysis. Results: Evaluated on a multicenter cohort, the model achieves a concordance index (c-index) of 0.82—significantly outperforming unimodal baselines (c-index: 0.75–0.77)—demonstrating robustness and clinical utility of heterogeneous data integration. Contribution: This work introduces the first interpretable multimodal paradigm specifically designed for tyrosine kinase inhibitor (TKI) resistance prediction, uniquely balancing high predictive accuracy with clinical traceability, thereby enabling more informed, individualized targeted therapy decisions.

Technology Category

• Machine Learning: Multimodal Learning
• Computer Vision: Multi-modal Vision
• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics

Abstract

Lung cancer is the primary cause of cancer death globally, with non-small cell lung cancer (NSCLC) emerging as its most prevalent subtype. Among NSCLC patients, approximately 32.3% have mutations in the epidermal growth factor receptor (EGFR) gene. Osimertinib, a third-generation EGFR-tyrosine kinase inhibitor (TKI), has demonstrated remarkable efficacy in the treatment of NSCLC patients with activating and T790M resistance EGFR mutations. Despite its established efficacy, drug resistance poses a significant challenge for patients to fully benefit from osimertinib. The absence of a standard tool to accurately predict TKI resistance, including that of osimertinib, remains a critical obstacle. To bridge this gap, in this study, we developed an interpretable multimodal machine learning model designed to predict patient resistance to osimertinib among late-stage NSCLC patients with activating EGFR mutations, achieving a c-index of 0.82 on a multi-institutional dataset. This machine learning model harnesses readily available data routinely collected during patient visits and medical assessments to facilitate precision lung cancer management and informed treatment decisions. By integrating various data types such as histology images, next generation sequencing (NGS) data, demographics data, and clinical records, our multimodal model can generate well-informed recommendations. Our experiment results also demonstrated the superior performance of the multimodal model over single modality models (c-index 0.82 compared with 0.75 and 0.77), thus underscoring the benefit of combining multiple modalities in patient outcome prediction.

Problem

• Predicting osimertinib resistance in NSCLC patients
• Lack of standard tools for TKI resistance prediction
• Multimodal ML model integrates diverse data for precision

Innovation

• Multimodal machine learning predicts resistance
• Integrates histology, NGS, clinical data
• Achieves high c-index of 0.82