Artificial intelligence in immunotherapy: revolutionizing diagnostic and therapeutic applications in cancer and autoimmune diseases

Artificial intelligence (AI) is increasingly advancing precision immunotherapy by integrating high-dimensional biomedical data to support diagnosis, treatment selection, and longitudinal monitoring in both cancer and autoimmune diseases. This review summarizes AI applications in biomarker discovery, prediction of immune checkpoint inhibitor (ICI) response and toxicity, neoantigen prioritization, CAR-T cell optimization, and therapeutic antibody engineering. In oncology, multimodal models combining multi-omics, medical imaging, and clinical variables improve patient stratification and non-invasive response assessment, with several imaging- and pathology-based prediction tasks reporting clinically meaningful performance (frequently AUC ~ 0.70–0.95 across tumor types and endpoints). In autoimmune diseases, AI enables earlier diagnosis, molecular subtyping, treatment-response prediction, and real-time disease activity tracking using EHR, laboratory, imaging, and wearable data—supporting precision management in conditions such as rheumatoid arthritis and type 1 diabetes. Key challenges include data heterogeneity, model interpretability, and governance; however, explainable AI, federated learning, and digital twin frameworks offer practical routes toward trustworthy clinical translation. Overall, AI is emerging as a foundational technology for next-generation, patient-specific immunotherapy across oncology and autoimmune medicine. Artificial intelligence enhances cancer and autoimmune disease immunotherapy through biomarker discovery, response prediction, neoantigen and antibody optimization, and real-time treatment management, enabling precision medicine across clinical applications.