InSilicoTrials has joined a research team led by Sage Bionetworks that has received an award from the Advanced Research Projects Agency for Health (ARPA-H) under its Critical Illness Immunological Reprogramming and Control Point Learning Engine, or CIRCLE, program. The collaboration brings InSilicoTrials’ computational modeling and simulation capabilities into a government-backed effort focused on developing patient-specific digital twins for critical care.
The project targets a difficult problem in intensive care medicine. Severe infections, traumatic injuries and complications associated with chronic disease can trigger immune and inflammatory responses that become harmful rather than protective. When these responses progress toward organ failure, physicians may have limited time to determine how an individual patient will respond to treatment.
The work also places digital health technology closer to the clinical validation process by combining computational models with patient-specific information and simulated clinical testing. As part of the TWIN-TECC project, InSilicoTrials will develop a secure environment in which digital twin models created through the CIRCLE program can be evaluated, compared and combined before they are considered for broader clinical applications.
InSilicoTrials to Focus on Validation and In Silico Trials
Digital twins are increasingly being explored as a way to represent an individual patient’s condition using data and computational models. In critical care, the concept could eventually help clinicians understand how a patient’s immune response is changing and evaluate potential interventions more quickly.
However, models intended for medical decision-making require more than sophisticated algorithms. Their performance, limitations and uncertainty need to be assessed systematically. InSilicoTrials’ role in the TWIN-TECC project will therefore focus heavily on establishing the credibility of these models.
The company will work on automated verification, validation and uncertainty quantification for digital twin models. These processes are intended to establish whether a model functions as designed, whether it accurately represents the biological or clinical system it is intended to simulate, and how uncertainty could affect its results.
InSilicoTrials will also support co-simulation involving mechanistic models, artificial intelligence models and hybrid approaches. This is significant because critical illness involves multiple biological processes that may not be adequately represented by a single modeling technique.
Another component of the work will involve in silico trials conducted using validated digital cohorts. Instead of relying exclusively on conventional clinical testing for every stage of model assessment, researchers can use computational patient populations to examine how models perform under defined conditions.
The project will also incorporate documentation designed to support regulatory review. InSilicoTrials said its work will include FDA-ready documentation intended to help move validated computational tools toward potential clinical use.
Luca Emili, executive chairman of InSilicoTrials, said the company’s participation represents recognition of its work on making computational models sufficiently credible for use in healthcare. Mario Torchia, CEO of InSilicoTrials, added that critical care presents a particularly demanding environment for digital twin technology because clinical decisions can need to be made within hours and patients can respond differently to the same intervention.
That combination of speed and individual variability is central to the CIRCLE program. Rather than treating critical illness as a uniform process, the initiative is designed around the idea that patient-specific models could help physicians better understand changing immune and inflammatory responses.
ARPA-H Program Connects AI, Critical Care and Regulatory Science
The CIRCLE program is focused on developing AI-enabled approaches that can help track and potentially modulate immune responses during critical illness. Its broader objective is to create tools that could provide clinicians with more detailed information while a patient’s condition is changing.
The TWIN-TECC project brings together expertise spanning artificial intelligence, model evaluation, critical care, immunology and regulatory science. Sage Bionetworks leads the team, with InSilicoTrials contributing its computational modeling platform and experience in digital model assessment.
For the healthcare technology industry, the project highlights a growing requirement for digital tools to demonstrate reliability before they can influence clinical decisions. Building a digital twin is only one part of the development process. Researchers also need methods for determining whether the model is sufficiently accurate for its intended purpose and whether its predictions remain reliable when applied to different patient populations or clinical circumstances.
Dr. Vincenzo Carbone, head of Research & Innovation at InSilicoTrials, said the ability to automate verification, validation and uncertainty assessment could help researchers refine and evaluate promising models more efficiently. He also pointed to the potential role of in silico clinical trials using validated digital cohorts.
The regulatory component could prove particularly important as computational models become more closely integrated with medical research and clinical workflows. Documentation that clearly establishes how a model was developed, tested and assessed can help researchers communicate its performance to regulators and other stakeholders.
For InSilicoTrials, participation in the ARPA-H initiative also creates an opportunity to apply its existing platform within a high-priority area of healthcare research. The company will provide infrastructure for evaluating digital twin models developed by teams participating in the CIRCLE program, potentially allowing different approaches to be tested within a common environment.
The initiative comes as healthcare researchers and technology companies continue to explore how computational models can complement conventional clinical research. In critical illness, where treatment decisions can change rapidly, the ability to simulate patient-specific responses could offer researchers another way to investigate potential interventions before or alongside clinical studies.
The CIRCLE program is funded in whole or in part by ARPA-H. The project team has stated that its views and conclusions should not be interpreted as representing the official policies of ARPA-H or the U.S. Government.
As the TWIN-TECC work progresses, the focus will remain on whether digital twin models can meet the level of evidence, validation and documentation required for real-world healthcare applications. For the digital health sector, that emphasis could be as important as the underlying modeling technology itself, particularly as computational tools move from research environments toward decisions involving individual patients.
