Digital Twins in Oncology: From Concept to Implementation
AI has become deeply embedded in modern medicine and oncology, supporting tasks ranging from early detection and diagnosis to prognosis, treatment planning, patient monitoring, and clinical decision support.1 Within this broader AI ecosystem, the concept of a digital twin—a high-fidelity virtual representation of a real-world individual that can simulate biological processes and clinical trajectories—has begun to show significant promise.2 In oncology, where cancer heterogeneity and patient variability are the norm, digital twins offer a pathway toward truly individualized care.3 However, like any transformative technology introduced into health care, digital twins bring both opportunities and challenges.4 A balanced perspective is essential to understand their potential impact and to design a roadmap for responsible and effective implementation.
What Digital Twins Would Enable in Oncology
Digital twins could reshape oncology across several key dimensions. First, they could enable identification of optimal treatment strategies by integrating diverse data sources, including genetic, molecular, clinical, environmental, and social determinants of health.5 Rather than relying on population-level evidence alone, clinicians could run in silico experiments on a patient’s twin—comparing drug combinations, dosing strategies, and sequencing approaches, which could reduce trial and error in treatment selection, minimize toxicity, and improve outcomes.
Second, digital twins could improve quality of life by predicting outcomes and treatment-related side effects along a patient’s health trajectory.6 By forecasting toxicity risks, disease progression, and survivorship challenges, digital twins would support proactive symptom management and personalized supportive care, ultimately shifting oncology toward a more anticipatory model.
Third, digital twins could enable virtual clinical trials and benchmarking of clinical performance through the use of virtual control arms.7 This would reduce the need for large control groups, accelerate drug development, and allow more patients to receive potentially beneficial experimental therapies. In addition, health-care systems could use digital twins to compare real-world performance against simulated optimal outcomes, identifying gaps in care delivery.
Fourth, digital twins hold promise for early intervention and prevention.8 By continuously updating with longitudinal data, a digital twin could detect subtle deviations from a healthy baseline, enabling early diagnosis or even prediction of cancer risk before clinical manifestation. This opens the door to precision prevention strategies tailored to the individual.
Major Challenges in Digital Twin Implementation
Despite their promise, several major challenges must also be addressed.
(1) Data Acquisition, Integration, Standardization, and Quality
Digital twins depend on vast amounts of multimodal data, including imaging, genomics, clinical records, and lifestyle information. Collecting, integrating, and curating such heterogeneous data streams is nontrivial. Moreover, inconsistencies in data standards, variable quality, and measurement errors can compromise model reliability. Establishing robust data governance, interoperability frameworks, and quality control processes is essential.
(2) Multiscale Modeling and Simulations
Human biology operates across multiple spatial and temporal scales—from molecular interactions to organ systems and behavioral dynamics. Capturing these layers with usable models is extraordinarily complex. Furthermore, cancer progression involves nonlinear, dynamic processes influenced by both biological and environmental factors. Developing models that can accurately represent these multiscale, time-evolving phenomena remains a significant scientific challenge.
(3) Responsible AI
The integration of AI into digital twins raises critical concerns about fairness, transparency, accountability, robustness, safety, privacy, and security. Bias in training data can lead to inequitable outcomes, while lack of explainability may hinder clinical trust. Ensuring that digital twin systems are interpretable, validated, and aligned with ethical standards is not optional—it is foundational.
(4) Computing Infrastructure
Digital twin models require substantial computational resources, including high-performance computing (HPC) and potentially emerging paradigms, such as quantum computing. Ensuring equitable access to such infrastructure, along with the ability to maintain and scale these systems, poses both technical and economic challenges.
Six Pillars for Realizing Digital Twins in Oncology
Addressing these challenges requires coordinated efforts by stakeholders across multiple domains.
A robust regulatory framework must evolve to ensure data security and privacy. Digital twins rely on sensitive, longitudinal data from sources such as electronic health records, wearable devices, and genomic sequencing. Compliance with regulations like HIPAA and the EU General Data Protection Regulation (GDPR), along with strong consent processes and cybersecurity measures, is essential.
Standardization is equally critical. Without common data models, interoperability frameworks, and benchmarking protocols, the scalability and reproducibility of digital twin systems will be limited. Developing global standards will enable broader adoption and ensure that advances benefit diverse populations.
Ethical considerations must remain central. The risk of exacerbating health disparities through unequal access to digital twin technologies is real. Governance structures must safeguard against discrimination, ensure equitable distribution of benefits, and uphold individual rights over digital representations.
Community engagement is another cornerstone. Incorporating patient and public perspectives through inclusive and culturally sensitive approaches—such as focus groups, town halls, and digital platforms—will improve trust, relevance, and adoption.
Sustainable funding strategies are also necessary. Funding agencies play a pivotal role in supporting interdisciplinary research, infrastructure development, education, and long-term maintenance of digital twin systems. Investment should extend beyond technology to include policy development and public outreach.
Finally, cross-disciplinary collaboration is indispensable. Initiatives such as the European Virtual Human Twins (VHT) Project9 and the Digital Twins for Health Society10 exemplify the kind of global, collaborative networks needed to advance the science and application of digital twins in health and oncology. By bringing together academia and industrial researchers, clinicians, engineers, data scientists, patient advocates, policymakers, and all other stakeholders, such efforts can accelerate innovation while maintaining scientific and ethical rigor.
Conclusions
The promise of digital twins in oncology is not just technological. It represents a fundamental shift toward truly individualized care, where each patient’s disease is understood and treated as a unique, evolving system. Realizing this vision depends on a deep understanding of human biology, robust multimodal data integration, and sophisticated multiscale modeling. Ensuring verification, validation, and uncertainty quantification (VVUQ)—alongside responsible AI principles—is critical for safe and effective implementation. Ultimately, sustained progress will depend on strong cross-disciplinary collaborations, global standards, and an unwavering commitment to ethical AI.
REFERENCES
1. Rajpurkar P, Chen E, Banerjee O, Topol EJ: AI in health and medicine. Nat Med 28:31-38, 2022.
2. National Academies of Sciences, Engineering, and Medicine. Foundational research gaps and future directions for digital twins. Washington, DC: The National Academies Press, 2024.
3. Stahlberg EA, Abdel-Rahman M, Aguilar B, et al: Exploring approaches for predictive cancer patient digital twins: Opportunities for collaboration and innovation. Front Digit Health 4:1007784, 2022.
4. Katsoulakis E, Wang Q, Wu H, et al: Digital twins for health: A scoping review. NPJ Digit Med 7:77, 2024.
5. Björnsson B, Borrebaeck C, Elander N, et al: Digital twins to personalize medicine. Genome Med 12:4, 2019.
6. Mulder ST, Omidvari AH, Rueten-Budde AJ, et al: Dynamic digital twin: Diagnosis, treatment, prediction, and prevention of disease during the life course. J Med Internet Res 24:e35675, 2022.
7. Moingeon P, Chenel M, Rousseau C, Voisin E, Guedj M: Virtual patients, digital twins and causal disease models: Paving the ground for in silico clinical trials. Drug Discov Today 28:103605, 2023.
8. Kamel Boulos MN, Zhang P: Digital twins: From personalised medicine to precision public health. J Pers Med 11:745, 2021.
9. European Virtual Human Twins (VHT) Initiative. Available at https://digital-strategy.ec.europa.eu/en/policies/virtual-human-twins. Accessed May 4, 2026.
10. Digital Twins for Health Society (DT4HS). Available at https://dt4hs.org/. Accessed May 4, 2026.
DISCLOSURES: Dr. Deng serves as a Scientific Advisor for Apex Health for Life LLC. Dr. Deng is the President of the Digital Twins for Health Society (DT4HS). Dr. Deng has received grant fundings from Yale ASCEND Initiative and Poorvu Center for Teaching and Learning.
Dr. Deng is a Professor of Therapeutic Radiology and Biomedical Informatics and Data Science at Yale University School of Medicine.
Disclaimer: This commentary represents the views of the author and may not necessarily reflect the views of ASCO, Conexiant, or ASCO AI in Oncology.