AI predicts immunotherapy response in lung cancer patients

Researchers involved in the international i3Lung project have developed an artificial intelligence model capable of predicting how patients with metastatic non-small cell lung cancer will respond to immunotherapy. The study, published in Nature Medicine, utilized data from 2,396 patients, integrating CT images, digital pathology, genomics, and clinical laboratory results into machine learning models.

Led by oncologist Arsela Prelaj from the Fondazione IRCCS Istituto Nazionale dei Tumori in Milan, the project aims to move beyond current methods that often only measure immunotherapy effectiveness after treatment has begun. By providing a way to stratify patients more accurately, the algorithm seeks to enable personalized treatment plans and improve the overall efficacy of standard care for this common subtype of lung cancer.