Recent applications of artificial intelligence in cancer radiotherapy and immunotherapy: current status and future directions

Artificial intelligence (AI) enhances the precision, personalization, and efficiency of cancer treatment through deep learning and machine learning techniques. This review comprehensively examines the evolution of AI and its expanding applications in cancer radiotherapy, immunotherapy, and drug discovery and repurposing. In radiotherapy, AI enables automated medical image segmentation, thereby facilitating the accurate delineation of tumor targets. Furthermore, AI-driven feedback systems support clinicians in developing individualized treatment plans by offering real-time assessment of treatment safety and potential efficacy. In the context of cancer immunotherapy, AI integrates multi-omics data to advance the discovery of novel biomarkers, analyze the tumor immune microenvironment, and accurately predict responses to immune checkpoint inhibitors. Moreover, AI accelerates drug discovery and repurposing through virtual screening, protein structure prediction, and the identification of novel therapeutic targets. However, the true clinical value of these AI models depends heavily on their generalizability across diverse patient cohorts and their performance compared to standard clinical baselines. Despite these promising prospects, AI still faces challenges in clinical applications, such as insufficient data standardization, poor model interpretability, and a lack of ethical oversight, delaying its formal inclusion into standardized clinical guidelines. With the rapid growth of data volume and computational power, AI is expected to play an increasingly central role in cancer management, holding immense promise for improving patient outcomes and advancing precision oncology.