The rapid advancement of artificial intelligence (AI) technologies has opened new avenues for advancing personalized immunotherapy in cancer treatment. This review highlights current research progress in applying AI to optimize the use of immunotherapy for patients with cancer. Recent studies demonstrate that AI models can accurately diagnose cancers and discover biomarkers by integrating multi-omics and imaging data, establish predictive models to estimate treatment responses and adverse reactions, formulate personalized treatment plans integrating multiple modalities by considering various factors, and achieve precise patient stratification and clinical trial matching, thereby addressing specific obstacles throughout processes from diagnosis to treatment in personalized immunotherapy. Furthermore, this review also discusses the challenges and limitations faced by AI in clinical applications, such as difficulties in data acquisition, low quality of data, poor interpretability of models, and insufficient generalization ability. Finally, we outline future research directions, including optimizing data management, developing explainable AI, and improving the generalization ability of models. These efforts aim to optimize the role of AI in personalized immunotherapy and promote the development of precision medicine. To ensure the clinical applicability of these AI models, large-scale studies, multi-omics integration, and prospective clinical trials are necessary.

An abstract is unavailable. This article is available as a PDF only.

Background: No robust biomarkers have been identified to predict the efficacy of programmed cell death protein 1 (PD-1) inhibitors in patients with locoregionally advanced nasopharyngeal carcinoma (LANPC). We aimed to develop radiomic models using pre-immunotherapy MRI to predict the response to PD-1 inhibitors and the patient prognosis. Methods: This study included 246 LANPC patients (training cohort, n = 117; external test cohort, n = 129) from 10 centers. The best-performing machine learning classifier was employed to create the radiomic models. A combined model was constructed by integrating clinical and radiomic data. A radiomic interpretability study was performed with whole slide images (WSIs) stained with hematoxylin and eosin (H&E) and immunohistochemistry (IHC). A total of 150 patient-level nuclear morphological features (NMFs) and 12 cell spatial distribution features (CSDFs) were extracted from WSIs. The correlation between the radiomic and pathological features was assessed using Spearman correlation analysis. Results: The radiomic model outperformed the clinical and combined models in predicting treatment response (area under the curve: 0.760 vs. 0.559 vs. 0.652). For overall survival estimation, the combined model performed comparably to the radiomic model but outperformed the clinical model (concordance index: 0.858 vs. 0.812 vs. 0.664). Six treatment response-related radiomic features correlated with 50 H&E-derived (146 pairs, | r |= 0.31 to 0.46) and 2 to 26 IHC-derived NMF, particularly for CD45RO (69 pairs, | r |= 0.31 to 0.48), CD8 (84, | r |= 0.30 to 0.59), PD-L1 (73, | r |= 0.32 to 0.48), and CD163 (53, | r | = 0.32 to 0.59). Eight prognostic radiomic features correlated with 11 H&E-derived (16 pairs, | r |= 0.48 to 0.61) and 2 to 31 IHC-derived NMF, particularly for PD-L1 (80 pairs, | r |= 0.44 to 0.64), CD45RO (65, | r |= 0.42 to 0.67), CD19 (35, | r |= 0.44 to 0.58), CD66b (61, | r | = 0.42 to 0.67), and FOXP3 (21, | r | = 0.41 to 0.71). In contrast, fewer CSDFs exhibited correlations with specific radiomic features. Conclusion: The radiomic model and combined model are feasible in predicting immunotherapy response and outcomes in LANPC patients. The radiology–pathology correlation suggests a potential biological basis for the predictive models.

Autonomous clinical artificial intelligence (AI) agents powered by large language models (LLMs), meaning systems that can complete a diagnostic workflow without continuous human input, are increasingly capable of supporting complex reasoning and decision-making. Clinical translation, however, remains limited by two unmet requirements: institutionally governed deployment and reliable decision-time uncertainty estimation. Here we developed and evaluated a fully on-premise clinical agent that couples local operational control with a multi-perspective reliability framework to support selective autonomy. Across two Medical Information Mart for Intensive Care IV (MIMIC-IV)-derived benchmarks, the agent achieved 90.04% accuracy on a seven-disease task and 83.8% accuracy on a four-disease task, approaching a cloud baseline on the primary benchmark. To assess decision-time reliability, we quantified internal-likelihood, language-based and behavioral-stability measures for diagnosis and reasoning. Diagnostic behavioral consistency provided the strongest discrimination of correctness (area under the curve (AUC) = 0.860) and remained informative under stress testing (AUC = 0.875). At a consistency threshold of 0.90, 49.4% of cases were retained at 98.9% diagnostic accuracy. These findings support a practical framework for institutionally governed clinical agents in which decision-time reliability signals identify a lower-risk subset for autonomous handling and defer the remainder for review.

Multidisciplinary tumor boards integrate longitudinal treatment histories, molecular profiling and rapidly evolving evidence to guide decisions in hematological malignancies, yet access to this level of subspecialty deliberation is increasingly uneven. Here we develop HemaGuide, a locally deployable, modular large language model agent that converts unstructured clinical documents into structured case representations, autonomously routes cases to specialized decision modes (‘guideline’, ‘advanced’ and ‘molecular’) and grounds recommendations in disease-specific guideline flowcharts and a clinical decision memory of >2,000 real-world tumor board cases. In expert-blinded benchmarking on 45 high-complexity cases across six foundation models, HemaGuide substantially improved concordance with tumor board decisions. A systematic ablation study across 11 layers confirmed that performance gains were routing-type-dependent, with no single component sufficient across case types. Automated classification of 70 clinically relevant missense variants showed high concordance with expert standards; no oncogenic variant was downgraded to benign and the whole workflow was completed under real-time conditions on commodity hardware with a median latency of 39 s rather than the hours typically required for manual molecular board workflows. In a simulated practice study, agent-assisted resident physicians achieved near-senior concordance and partially outperformed senior physicians in their subspecialty. External validation on 555 independent cases from a second academic center yielded 81.8% concordance across 47 entities, and a prospective 1-month silent trial on 64 consecutive, unselected cases achieved 82.8% concordance. Hallucinations occurred in 2 of 664 evaluated cases (0.3%). Together these data provide evidence that locally deployable, case-grounded large language model agents can deliver auditable clinical decision support across hematological malignancies, with concordance maintained across institutions and under real-time conditions on commodity hardware.