AI Tools Enhance Immunotherapy Outcome Prediction in Lung Cancer Patients
Artificial intelligence (AI) tools are demonstrating significant potential in predicting treatment and survival outcomes for patients with advanced non-small cell lung cancer (NSCLC) undergoing immunotherapy. A new study, published in Nature Medicine and stemming from the I3LUNG project, highlights the development of AI-based predictive models designed to optimize therapeutic approaches for individual patients. Immunotherapy, while transformative for 20% to 30% of lung cancer patients, often faces resistance, and current prediction methods, largely based on PD-L1 expression, have limitations. The I3LUNG project integrated diverse patient data, including clinical, imaging, pathology, and genomic information, to train and validate two families of AI models. These models consistently outperformed standard clinical biomarkers, with one model achieving an AUC score of 0.88 when incorporating clinical, blood, imaging, and digital pathology data. The study also found that AI support improved physicians' ability t...