AI-Powered Digital Twins Offer a New Window Into Tumor Metabolism in Brain Cancer

Advances in precision oncology increasingly emphasize the importance of understanding tumor metabolism, particularly in aggressive cancers such as glioblastoma. Brain tumors are known for their metabolic adaptability, enabling them to survive, resist therapy, and recur despite aggressive treatment. Historically, clinicians and researchers have faced significant barriers to directly measuring metabolic activity within human tumors, especially in real time. This has been overcome by an innovative approach used in the latest research study at the University of Michigan (UM), which uses the digital twin technique based on artificial intelligence.^1-3

The study, published in Cell Metabolism, describes the first success in using the power of machine learning-based "digital twins" to noninvasively measure metabolic flux in actual patients' tumors. This breakthrough has the potential to open the door to personalized metabolic therapies by predicting the treatment success rate for each patient and avoiding treatment failures.²

The Challenge of Measuring Tumor Metabolism in Real Time

Metabolism in tumors can change dynamically in accordance with environmental and pressure-related factors. Yet most current metabolic analyses rely on static tissue samples collected during surgery. This approach fails to grasp the dynamic nature of metabolism in the tumor.

“Typically, metabolic measurements during surgeries to remove tumors can't provide a clear picture of tumor metabolism—surgeons can't observe how metabolism varies with time, and labs are limited to studying tissues after surgery. By integrating limited patient data into a model based on fundamental biology, chemistry, and physics, we overcame these obstacles,” said Deepak Nagrath, UM professor of biomedical engineering and co-corresponding author of the study.¹

This has posed a challenge for personalizing any metabolic treatment, whether related to diet or metabolic inhibitors. This implies that, because metabolic flux cannot be measured, any treatment given to patients is generalized.

Digital Twin Technology Meets Oncology

To address this challenge, the research team applied digital twin technology, a concept borrowed from engineering and manufacturing in which virtual replicas simulate real-world systems. In medicine, a digital twin integrates patient-specific data into computational models to predict biological behavior under different conditions.⁴

To achieve digital twins of brain tumors, the research team used the subject’s intraoperative patient data to generate digital twins using models developed in biochemistry and physics. Machine learning models were applied to simulate metabolic fluxes and infer metabolism.²

“This is the first time a machine learning and AI-based approach has been used to measure metabolic flux directly in patient tumors,” said Baharan Meghdadi, a doctoral student in chemical engineering and co-first author of the study.¹

The approach allows researchers to test how tumors might respond to different metabolic therapies virtually, before exposing patients to potentially ineffective or toxic treatments.

Predicting Therapeutic Response and Resistance

The most promising implication of this technology in the art of predicting therapeutic resistance is that many tumors intrinsically acquire resistance to metabolic therapies, rendering treatments ineffective and exposing patients to unnecessary side effects.

“These results are exciting. The ability to measure metabolic activity in patient tumors could allow us to predict which metabolic therapies might work best for each patient,” said Daniel Wahl, the Achtenberg Family Professor of Radiation Oncology and a co-corresponding author of the study.¹

Similarly, Wajd N. Al-Holou, assistant professor of neurosurgery and co-first author, emphasized the potential clinical impact of this predictive capability. “This amazing tool could help doctors avoid prescribing treatments that a specific tumor is already equipped to resist and is a way for us to move towards more targeted and personalized treatments for our patients,” Al-Holou said.¹

Implications for Diet-Based and Metabolic Therapies

The study has considerable implications for new metabolic therapies, such as dietary therapies aimed at depriving tumor cells of a certain nutrient. Experimental work has shown that a new diet could affect tumor metabolism, but results for patients have been variable.³

The simulation of the metabolic response using the digital twin could, in principle, help clinicians identify patients who would most benefit from dietary and/or metabolic management. This aligns with the broader efforts to integrate nutrition, metabolism, and oncology into more comprehensive management plans.

“This work moves us closer to truly personalized cancer care—not just for brain cancer, but eventually for a variety of tumors. By simulating different therapies virtually, we hope to spare patients from unnecessary treatments and focus on those likely to help,” said Costas Lyssiotis, the Maizel Research Professor of Oncology and co-corresponding author of the study.¹

What Does This Mean for Pharmacists?

For pharmacists, especially those working in cancer and outpatient settings, the use of digital twins may play an increasingly important role in treatment decision-making and medication management. As metabolic therapies and other treatments become more personalized, pharmacists will play an important role in interpreting metabolic information and counseling patients.

Pharmacists may also help detect medication-nutrient interactions, manage the adverse effects of metabolic therapy, and communicate the logic and rationale of the treatment regimen to patients. As AI-driven tools such as digital twins move closer to clinical adoption, pharmacists’ expertise in pharmacokinetics, metabolism, and patient-centered care will become increasingly valuable.

Although this technology remains in the research phase, its implications are far-reaching. By enabling direct measurement of tumor metabolism without invasive sampling, AI-powered digital twins could redefine how clinicians approach cancer treatment. As validation studies continue and clinical integration advances, this approach may extend beyond brain cancer to other metabolically driven malignancies.

For now, the study represents a significant step toward precision oncology—one that leverages artificial intelligence not just to analyze data, but to meaningfully guide individualized patient care.

REFERENCES

1. Brain cancer digital twin predicts treatment outcomes. EurekAlert!. Published January 12, 2026. Accessed January 13, 2026. https://www.eurekalert.org/news-releases/1112339

2. Meghdadi B, Mittal A, Nagrath D, et al. Digital twins for in vivo metabolic flux estimations in patients with brain cancer. Zenodo (CERN European Organization for Nuclear Research). Published January 6, 2026. Accessed January 13, 2026. doi:10.5281/zenodo.17373726

3. Sen A. Dietary changes could provide a therapeutic avenue for brain cancer. Michigan Medicine. Published September 3, 2025. Accessed January 13, 2026. https://www.michiganmedicine.org/health-lab/dietary-changes-could-provide-therapeutic-avenue-brain-cancer

4. What is digital-twin technology? McKinsey & Company. Published August 26, 2024. Accessed January 13, 2026. https://www.mckinsey.com/featured-insights/mckinsey-explainers/what-is-digital-twin-technology