The following is a guest article by Thomas Kluz, Managing Director at Niterra Ventures
The most consequential shift in modern medicine isn’t happening inside hospitals or biotech labs. It’s happening inside medical simulation labs/training centers.
Drug development has long been marked by high costs, inefficiency, and human trial-and-error. Indeed, the cost of developing a single therapeutic exceeds $2.6 billion according to the Tufts Center for the Study of Drug Development. Much of this expense can be attributed to the clinical trial process.
Clinical trials are a true bottleneck in pharmaceutical innovation. Not only are they slow and expensive, but they often fail to produce data that is broadly applicable. Securing patients is perhaps one of the greatest challenges in conducting a clinical trial. Even with modern recruitment platforms and global trial networks, nearly half of all trial sites enroll less than two participants. This enrollment challenge is exacerbated by the need to secure patients representative of the diversity and complexity of real-world patients. Women, minorities, and patients with multiple chronic conditions have routinely been underrepresented in research cohorts. Efforts to secure an appropriate mix of patients result in delays, compounding the expense of bringing a Phase I drug study to fruition.
Add to all this, clinical trials are difficult to scale efficiently. They cannot quickly adapt to emerging threats or novel therapies. And they certainly cannot simulate how a treatment would behave in an ultra-rare disease population that’s difficult, or impossible, to assemble in the real world.
Imagine then, if we could conduct drug testing, simulate adverse effects, and refine trial protocols without recruiting a single human subject.
Enter the Virtual Patient
“Virtual patients” are the answer to the clinical trial bottleneck conundrum.
Virtual patients are computational models trained on real-world clinical and genomic data to reflect plausible human physiology, pathology, and behavioral responses. These synthetic stand-ins can be used to optimize trial design, predict biological outcomes across subpopulations, and train clinicians in diagnostic reasoning using interactive, lifelike avatars.
A virtual patient is not a digital puppet. It is an evolving system of interconnected variables governed by physiological constraints. These patients can be used to test hypotheses, simulate rare interactions, and model treatment pathways under a range of clinical conditions.
In the hands of drug developers, they can shorten timelines and reduce reliance on large trial populations. In medical education, they offer opportunities to train future physicians on realistic cases that would otherwise be too dangerous, rare, or ethically complex to recreate. For regulators, they can serve as an adjunct source of evidence, particularly in early-phase evaluations or for therapies targeting rare diseases.
Most importantly, virtual patients offer an opportunity to build healthcare systems that are inclusive from the start, not just by design, but by simulation.
Why Now?
Though the idea has been around for years, the convergence of three forces has made virtual patients newly viable.
First, large generative models, originally designed for natural language or image generation, have demonstrated the capacity to capture dynamic human phenomena, including disease progression and response to therapy.
Second, federated learning methods allow for the training of models across distributed, privacy-protected datasets, preserving patient confidentiality while enabling large-scale insight generation.
Third, both the U.S. Food and Drug Administration (FDA) and the European Medicines Agency (EMA) have taken meaningful steps toward recognizing simulation data in support of medical device and drug evaluations. This does not mean virtual trials will replace real ones, but they will increasingly shape them, refine them, and de-risk them.
In short, the technical, regulatory, and infrastructural barriers that once made virtual patients feel speculative are collapsing. What remains is execution and ambition.
The Risks Are Real and So Are the Responsibilities
The promise of virtual patients does not absolve us of the duty to scrutinize. Models trained on incomplete, biased, or outdated data will simply reproduce the failings of the systems they’re meant to improve. Explainability, data provenance, and biological fidelity must be prioritized over superficial resemblance or visual realism.
Investors and builders alike have an obligation to avoid shallow optimism. The value of a virtual patient lies not in its believability, but in its accuracy. This requires input from across disciplines: bioinformatics, clinical medicine, regulatory science, ethics, and systems engineering.
Without this commitment to rigor, the same tools that could revolutionize medicine could also erode its legitimacy.
Where the Smart Capital Should Flow
The investment opportunity in virtual patients is not hypothetical, it is infrastructural.
Consider three domains where capital and innovation can converge to produce durable value:
- Digital Twin Platforms: These tools create individual-specific models for use in trial optimization, risk prediction, and personalized treatment planning
- Synthetic Data Engines: Broader model training requires diverse, high-quality data; generating synthetic health data that preserves statistical validity while maintaining privacy is a critical enabler of progress
- Medical Education Applications: Virtual patients can be embedded in clinical training programs, equipping healthcare professionals with realistic diagnostic and therapeutic scenarios that reflect the spectrum of patient variation
The Future Isn’t Virtual… But It Will Be Simulated
Virtual patients will not eliminate clinical trials or replace real-world testing. But they will transform how we prepare for them, design them, and learn from them. By modeling complexity rather than simplifying it, they can elevate medicine from its reactive posture to a more predictive, inclusive, and efficient paradigm.
We owe it to the patients who wait, and the ones we will never meet, to get this right.
About Thomas Kluz
Thomas Kluz is a distinguished venture capitalist with over a decade of experience. He’s the Managing Director at Niterra Ventures, where his investments focus on energy, mobility, and healthcare. With deep expertise in healthcare-focused venture capital, he has a proven track record of success with various organizations, such as Qualcomm Ventures and Providence Ventures.
