The digital twin (DT) concept, originating from engineering disciplines, has emerged as a transformative technology in healthcare, particularly in oncology. A digital twin creates a dynamic, virtual replica of a patient's physiological and pathological state, integrating multi-dimensional data to enable personalised cancer care. Despite growing interest, comprehensive reviews examining the breadth of DT applications in oncology remain limited. This narrative review aims to synthesise current evidence on digital twin applications in oncology, evaluate their potential to transform cancer care delivery, and identify challenges hindering clinical translation. A comprehensive literature search was conducted across PubMed, Scopus, Web of Science, and IEEE Xplore databases from inception to September 2025. Studies describing DT development, validation, or application in any cancer type were included. Grey literature, conference proceedings, and expert commentaries were also reviewed to capture emerging trends. Digital twins demonstrate applications across the cancer care continuum, including precision treatment selection, radiotherapy optimisation, drug development, immuno-oncology modelling, surgical planning, and survivorship care. Integration of multi-omics data, imaging biomarkers, and artificial intelligence enables dynamic simulation of tumour behaviour and treatment response. However, challenges persist in data integration, model validation, computational scalability, and ethical governance. Digital twin technology holds substantial promise for advancing precision oncology through predictive, personalised, and adaptive care strategies. Addressing current limitations through interdisciplinary collaboration and regulatory framework development is essential for clinical implementation.

Checkpoint inhibitors have revolutionised cancer treatment, yet only 20-30 % of patients achieve durable responses, highlighting the critical need for predictive models. This review focuses on PD-1/PD-L1 pathway inhibitors as monotherapy, examining current prediction frameworks spanning biomarker-based approaches, multi-omics integration, mathematical modelling, and artificial intelligence applications. Recent advances include SCORPIO and LORIS machine learning systems demonstrating superior statistical performance compared to traditional biomarkers, with area under curve values of 0.763. However, critical analysis reveals significant limitations in external validation across diverse healthcare settings, with many promising models failing to maintain performance outside their development institutions. Traditional pathological assessment by expert pathologists, including standardised PD-L1 scoring and tumour-infiltrating lymphocyte quantification, continues to form the foundation of clinical decision-making and provides essential validation for emerging AI approaches. Despite extensive research, established biomarkers show limited predictive accuracy, with PD-L1 demonstrating predictive value in only 28.9 % of FDA approvals. Multi-feature models incorporating genomic and clinical data show improved accuracy but face substantial validation challenges. Integration of spatial biomarkers and digital pathology has enhanced capabilities, achieving area under curve values of 0.84 in select studies. The most critical challenge is the "validation gap", many models show excellent single-institution performance but fail external validation, limiting clinical translation. Current obstacles include inadequate standardisation, interpretability concerns, and healthcare system integration difficulties. Future directions must prioritise rigorous multi-institutional validation studies, development of clinically implementable frameworks, and addressing practical deployment challenges to realise precision immunotherapy's potential.

Immune organoids have emerged as a ground-breaking platform in immunology, offering a physiologically relevant and controllable environment to model human immune responses and evaluate immunotherapeutic strategies. Derived from stem cells or primary tissues, these three-dimensional constructs recapitulate key aspects of lymphoid tissue architecture, cellular diversity, and functional dynamics, providing a more accurate alternative to traditional two-dimensional cultures and animal models. Their ability to mimic complex immune microenvironments has positioned immune organoids at the forefront of cancer immunotherapy development, autoimmune disease modeling, and personalized medicine. This narrative review highlights the advances in immune organoid technology, with a focus on their applications in testing immunotherapies, such as checkpoint inhibitors, CAR-T cells, and cancer vaccines. It also explores how immune organoids facilitate the study of autoimmune disease pathogenesis with insights into their molecular basis and support in high-throughput drug screening. Despite their transformative potential, immune organoids face significant challenges, including the replication of systemic immune interactions, standardization of fabrication protocols, scalability limitations, biological heterogeneity, and the absence of vascularization, which restricts organoid size and maturation. Future directions emphasize the integration of immune organoids with multi-organ systems to better replicate systemic physiology, the development of advanced biomaterials that closely mimic lymphoid extracellular matrices, the incorporation of artificial intelligence (AI) to optimize organoid production and data analysis, and the rigorous clinical validation of organoid-derived findings. Continued innovation and interdisciplinary collaboration will be essential to overcome existing barriers, enabling the widespread adoption of immune organoids as indispensable tools for advancing immunotherapy, vaccine development, and precision medicine.

Digital twin technology has emerged as a transformative innovation in healthcare, offering virtual replicas of physical entities at patient-level, equipment-level, and departmental-level that enable real-time monitoring, prediction, and optimisation. This narrative review synthesizes current evidence on digital twin maturity and clinical translation in radiology and radiotherapy. A comprehensive literature search was conducted across PubMed, Scopus, IEEE Xplore, and Web of Science databases for peer-reviewed articles published from 2018 onwards. The review reveals that digital twin applications in radiology remain predominantly experimental, with equipment-focused implementations (predictive maintenance, workflow optimization) showing greater maturity than patient-level applications. In radiology, emerging applications include personalised imaging protocol optimisation, predictive equipment maintenance, dose management, and workflow enhancement. In contrast, radiotherapy demonstrates more advanced patient-level digital twin integration, facilitating individualised treatment planning, real-time dose adaptation, treatment response prediction, and quality assurance. DT-aligned adaptive radiotherapy report improved local/locoregional control in the low-teens to ∼18% relative range, alongside clinically meaningful toxicity-risk reductions in selected endpoints, while maintaining lower radiation dose to organs. Key benefits include improved patient outcomes, reduced radiation exposure, enhanced treatment precision, and optimised resource utilisation. However, critical gaps persist in standardized validation frameworks, interoperability standards, and regulatory guidance. Implementation faces challenges including data integration complexity, computational requirements, regulatory uncertainties, and domain-specific barriers differing between radiology and radiotherapy contexts. Successful clinical translation requires addressing technical infrastructure gaps, establishing evidence-based validation protocols, and developing reimbursement mechanisms that recognize digital twin value. Digital twin technology demonstrates substantial potential for advancing precision medicine in imaging and radiation oncology.

Artificial intelligence (AI) offers potential solutions to address critical challenges in oncology practice, particularly in resource-constrained settings like Nigeria. However, successful implementation requires understanding healthcare providers' perspectives, which remain largely unexplored in the Nigerian context. To explore Nigerian oncologists' perspectives on AI applications in oncology practice, identifying knowledge levels, perceived benefits, implementation barriers, and priority areas for AI integration. This qualitative study employed a descriptive exploratory design. Semi-structured interviews were conducted with 15 oncologists from nine major Nigerian healthcare institutions. All interviews were conducted in English. These institutions represent tertiary referral centres predominantly located in urbanised areas across different Nigerian geopolitical zones, including Southwest (OAUTH, LUTH, UCH, LASUTH, LAUTH), South-South (ISTH, UBTH), and North-Central (BSUTH, UATH). Participants represented various oncology specialties with experience ranging from 1 to 20 + years. Data were analysed using Braun and Clarke's six-phase thematic analysis approach with independent coding by multiple researchers to ensure inter-coder reliability. Nine key themes emerged: (1) Current Knowledge and Awareness of AI in Oncology; (2) Perceived Benefits of AI in Oncology Practice; (3) Perceived Barriers to AI Implementation; (4) AI in Oncology Research; (5) Data Management and Ethical Concerns; (6) Trust and Adoption Readiness; (7) Human-AI Interaction and Patient Dynamics; (8) Future Directions and Knowledge Requirements; and (9) Resource Allocation and Infrastructure Development. Participants demonstrated limited theoretical knowledge of AI applications, with most lacking practical implementation experience. Participants recognised AI's potential to address workforce shortages and improve diagnostic accuracy but identified significant barriers including financial constraints, infrastructure limitations, and insufficient technical expertise. Nigerian oncologists expressed cautious optimism about AI's potential to transform cancer care delivery despite substantial implementation challenges. Successful AI integration requires addressing infrastructure deficits, developing appropriate regulatory frameworks, and building technical capacity. A phased implementation approach focusing initially on diagnostic support applications is recommended, alongside sustained investment in digital infrastructure and workforce development.