Digital Twins: A Tool for Risk Mitigation in the Era of COVID-19

Enrollment challenges, timelines, and high failure rates burden clinical trials. COVID-19 has only compounded these issues. Nearly 200 companies have stopped or delayed their trials due to the pandemic (1). Interest in trial participation, as well as perception of the safety of clinical trials, has decreased (2). Research in disease areas like Alzheimer’s, with vulnerable populations, a history of unsuccessful trials, and no disease-modifying treatments, is especially at risk. An estimated 150 to 300 experimental Alzheimer’s treatments have failed to show clinical benefits (3). 

The world has now seen how difficult it is to conduct clinical research - limited resources, the logistical complexity of setting up a randomized controlled trial, and the timelines that seem to crawl, despite the immense pressure in the race to find new treatments as quickly as possible. A devastating trial failure is a technical one - for example, a trial is unable to be carried out because of insufficient enrollment, or dropout. Without the pre-specified, required number of patients (sample size), the statistical power required to analyze the trial results cannot be reached. 

In what ways can we mitigate risk and apply innovative solutions to unstable trials in the wake of COVID-19? What if there was a way to recover statistical power in a trial, enabling the study to move forward, despite attrition? Incorporating Digital Twins can salvage trials with low enrollment, due to COVID-19, or other external factors. 

A Digital Twin is a comprehensive, longitudinal, and computationally-generated clinical record that describes what would likely have happened to a specific subject if they had received a placebo. Leveraging historical trial data and generative machine learning models, these statistically indistinguishable, virtual patients are generated using baseline data, and are precisely matched to patients in an actual trial. 

A simple, hypothetical scenario can demonstrate how Digital Twins could restore a trial impacted by the pandemic. A research team determines that they will need 200 patients to achieve a statistical power of 80% in a randomized controlled trial in Alzheimer’s disease. 50 patients demonstrate decreased adherence - they are unable to visit sites due to fear, illness, limited public transportation, or other reasons exacerbated by COVID-19. The trial can now only be powered at 65%. Running a trial with low power increases the risk of a negative result, even if the treatment is actually effective. 

Adding Digital Twins to the trial to make up for the loss of 50 actual patients recovers the statistical power, returning the percentage to 80%. 

Digital Twins have enormous potential to rescue trials in jeopardy of failure due to COVID-19, while managing regulatory risk and limiting timeline delays. Craig Lipset, former head of Innovation at Pfizer and Strategic Advisor for Unlearn, recognizes the marked opportunity to apply novel technology to support trials during these unprecedented times. "Incorporating Digital Twins in an ongoing clinical trial is a feasible and low-burden way to reduce risk for clinical trials running during the pandemic. Building a 'third arm' of Digital Twins provides some mitigation should today's unpredictable environment prevent patient visits to a site and accelerate data loss while slowing enrollment."

The FDA is encouraging alternative forms of evidence and innovative trial designs, and has engaged directly with the Unlearn team regarding the use of Digital Twins. Technology companies and research organizations must rise to the challenge of finding creative ways to advance discovery despite chronic and new barriers to trials. Unlearn is seeking partners to help drive innovation in clinical research - inventive approaches are posed to be the unique relief for vulnerable trials.


1.More than two-thirds of trials hit by COVID-19 enrollment halts, with midstage tests the worst affected

2.Patient Willingness to Join Clinical Trials Drops Dramatically, New Data Show

3.'Crushing': Another Alzheimer's treatment trial has failed. What's next?

Enter your email address to download paper.

Click the link to begin download.
Oops! Something went wrong while submitting the form.

Enter your email address to watch the webinar.

Click the link to watch webinar.
Oops! Something went wrong while submitting the form.

Why can’t we agree on how to define digital twins in healthcare?

White Papers

Summary of the EMA September 2022 Qualification Opinion for PROCOVA™


Charles Fisher, Unlearn.AI: “now is the time to adopt AI-based solutions”

The potential for AI implementation in healthcare can barely be measured, as it can already do what humans do, just countless times better and more efficiently.
The European Medicines Agency has qualified Unlearn’s AI-powered method for running smaller, faster clinical trials.
Digital twins seem to be everywhere in healthcare now, but no one agrees on a single definition for them.