How Digital Twins Are Rewriting Clinical Trials

By Denise N. Bronner, founder, Empactful Ventures

Once a tool reserved for optimizing manufacturing systems, digital twinning has entered a new frontier: clinical research. Pharma companies are increasingly exploring the use of digital twins to create “virtual patients”: data-rich simulations modeled from biological, clinical, behavioral, and pharmacological inputs. These virtual counterparts can be used to test interventions, predict drug responses, and optimize trial design, all before a single real-world patient is enrolled.

Pharma companies are piloting digital twins to streamline early-phase trials, potentially accelerating development timelines while lowering both financial and ethical costs. Their work illustrates what’s possible when simulation replaces speculation: fewer failed trials, faster go/no-go decisions, and less dependence on human testing during the riskiest stages of development.

While digital twins are powerful, they are not a replacement for traditional trials. They should be viewed as a tool to strengthen trial design and decision-making. Their value lies in helping researchers stress-test protocols, anticipate risks, and refine strategies before engaging with real patients, whose voices and lived experiences remain central to medical progress.

But while the benefits are undeniable, the road ahead is riddled with questions, and those questions go far beyond technology. They strike at the heart of what it means to simulate care, risk, and response in populations whose lives are anything but binary code.

The Case For Digital Twins

The traditional clinical trial model is expensive, slow, and often excludes the very patients most impacted by disease. With many trials failing in Phase 2 or 3 due to unforeseen safety signals, dosing errors, or lack of efficacy. Enter digital twins.

Key benefits include:

• Accelerated Timelines: Digital models can simulate drug responses across thousands of virtual patients in days, helping teams test scenarios and optimize trial protocols faster.
• Cost Reduction: Running simulations before human enrollment can reduce failed trials and the associated cost burden.
• Ethical Advantage: Fewer early-stage human tests mean less exposure to experimental treatments and fewer dropouts due to adverse effects.
• Precision Simulation: With enough data, digital twins can mimic real-world variability in metabolism, genetics, and comorbidities that guide dosage, delivery, and even recruitment criteria.

This isn’t theoretical. Sanofi’s internal studies have used digital twinning to refine early-phase oncology trial designs, exploring how tumor types might respond across modeled populations. The results helped guide patient selection and dosing strategies, potentially reducing the number of live trial participants needed in early testing.

Bias, Consent, And The Ethics Of Simulation

As pharma races toward digital-first research, a series of uncomfortable questions emerge, especially when digital twins are used for therapeutic evaluation, not just trial operations.

1. Is the Data Representative?

Digital twins are only as strong as the data used to build them. Many simulations rely on existing clinical trial data, EHRs, genomic repositories, and insurance claims — all of which skew toward white, insured, urban populations. If the model underrepresents Black, Indigenous, rural, LGBTQ+, disabled, or elderly patients, it risks compounding existing health disparities rather than reducing them.

2. How Current Is the Input?

Older data sets may not reflect current diagnostic criteria, evolving standard-of-care protocols, or newly recognized biomarkers. Using outdated data to simulate today’s patient outcomes could lead to false confidence in a therapy’s efficacy or safety.

3. Do Patients Know Their Data Is Being Used This Way?

Informed consent is murky when it comes to secondary data use. Patients who originally consented to share their data for treatment or traditional research may not have anticipated its reuse in digital twin modeling. Questions around broad consent, re-consent, and transparency remain unresolved — particularly when commercial incentives are involved.

4. Who Gets Access to the Results?

If digital twin findings influence trial design or lead to drug approvals, will the results be published? Will underserved communities whose data powered the models have visibility into the outcomes? Or will this remain a black box behind proprietary IP walls?

Can We Model Complexity Without Losing The Patient?

Digital twins promise a world where we can simulate thousands of patients, optimize trial designs, and even anticipate adverse events — without exposing a single human being to risk. But as pharma edges closer to this vision, we must confront an uncomfortable truth: human beings don’t behave like code.

Clinical trials are more than scientific exercises. They are social, emotional, and deeply personal experiences. Real patients show up with a web of invisible factors: fear of side effects, childcare needs, mistrust of institutions, co-occurring illnesses, transportation issues, cultural stigmas, fluctuating health literacy, and histories of medical harm. These are not bugs in the system — they are the system. And no matter how sophisticated our simulations become, these lived experiences don’t easily map onto a neural network.

Even the most data-rich digital twin will struggle to replicate what makes real-world participation so unpredictable and, in many ways, so valuable. We learn from a patient’s biological response as well as their lived response: the unexpected side effects; their rationale for quitting a trial; and their moments of trust, or doubt, or triumph. These insights help us build not just better drugs but better systems of care.

That’s why the future of clinical research must embrace a hybrid model. Digital twins should be viewed as an enhancement, not a replacement for patient participation. They can help optimize protocols, refine eligibility, and stress-test designs before launch. But they must be validated in real-world populations, and those populations must be diverse, engaged, and meaningfully informed.

To do otherwise is to fall into a dangerous trap: believing that the model is the medicine. In the rush to digitize, we risk flattening the very thing that makes medicine work: human complexity. You can simulate pharmacokinetics. But you can’t simulate a patient’s decision to skip a dose because they couldn’t afford the food they needed to take it with that day. In the end, no matter how accurate the simulation, drugs are not delivered to avatars. They’re delivered to people.

Final Takeaway

Digital twins could redefine the contours of clinical research, helping us test faster, design smarter, and serve patients more ethically. But to get there, we must do more than optimize models; we must interrogate them. Not just for accuracy, but for equity, transparency, and trust.

The real win isn’t a faster trial.

It’s a future where simulation enhances, not replaces, the deep complexity that only real patients can bring.

About The Author:

Denise N. Bronner, Ph.D., has roughly 15 years of organizational thought leadership experience within the global healthcare space and has held various roles in academia, consulting, pharma, and venture capital. During her career, she has specialized in health equity, data-driven global therapy program strategy development, pitch and storytelling refinement, and identifying business opportunities within pharma. Beyond her professional endeavors, she's passionate about enhancing diversity in STEM fields, serving on advisory boards, participating as a judge in pitch/business competitions, and mentoring young professionals. She holds a bachelor’s degree in biological sciences from Wayne State University, a Ph.D. in microbiology & immunology from the University of Michigan - Ann Arbor, and certification from the Venture Capital Executive Program from UC Berkeley Haas School of Business. She is the founder of Empactful Ventures, which currently consults healthcare-focused startups and venture funds, and she is a member of the Clinical Leader editorial board.