Clinical trials will always need human participants. But that’s not to say that technologies like digital twins can’t make the trials run better.
Digital twins are virtual models of patients built from molecular and clinical data that can be updated in real time and simulate the outcome of different disease or treatment scenarios. Generative AI has made it easier to forecast how a patient’s condition will develop based on previous data, making way for more accurate digital twins that can be deployed across various applications in healthcare, like health management and disease treatment, according to a review in the journal npj Digital Medicine.
Health tech company Aitia uses “causal generative AI,” which can determine underlying cause and effect relationships between molecular variables and disease outcome to “reverse engineer models of disease” using molecular and genetic patient data they receive through partners as well as public and private biobanks, CEO and Co-founder Colin Hill told Healthcare Brew.
The main challenges for running clinical trials are the cost, lengthy timelines, and recruiting and retaining patients, Jun Deng, professor of therapeutic radiology at Yale University School of Medicine and an author of the review, said.
Digital twins can address each of these problems, he added.
For every physical patient, their data can be used to “simulate a lot of what-if scenarios” and create slightly different versions of the original patient, Deng explained. This can be used to enhance the power of the results by rounding out the sample size, especially in the control arm.
Additionally, he said digital twins could also be used to improve patient recruitment by helping researchers identify target populations that would have the best treatment outcome.
Healthcare consulting firm Towards Healthcare reported that the digital twin market for clinical trials was worth $1.17 billion in 2022 and estimated it will grow to $38.43 billion by 2032.
Navigate the healthcare industry
Healthcare Brew covers pharmaceutical developments, health startups, the latest tech, and how it impacts hospitals and providers to keep administrators and providers informed.
And government agencies like the FDA are interested in supporting new projects exploring the use of digital twins in preclinical and clinical research. In October, the federal agencies awarded a total of $6 million to seven different projects, including digital twin-based studies of neurodegenerative diseases and virtual clinical trials of cardiovascular medical devices.
Names in the game
Big pharma companies such as AstraZeneca, Bayer, Sanofi, and Merck KGaA are exploring the use of digital twins to improve trial efficiency.
Digital twin developer Unlearn, which has worked with Merck KGaA, Roche, and J&J, has said its technology could accurately simulate placebo outcomes and reduce control arm sizes by 33%–35%. “For a trial with 1,000 patients, reducing the control group by 25% can cut enrollment time by four to five months,” Unlearn’s CEO Steve Herne wrote for life science research firm the Conference Forum.
In an online pamphlet, Bayer stated its use of digital twins helped select doses for an anticoagulant drug and reduced adverse side effects like risk of stroke, heart attack, and thrombosis. Digital twins also served as “external control arms” in studies where recruiting human controls was “not feasible or ethical.” Bayer noted that barriers for implementing digital twins more widely include the fragmentation of healthcare data and a lack of regulatory guidance around acceptance for computer-generated evidence.
Simulating the drug at the individual patient level before the trial enables researchers to discover “markers of response to the drug,” Hill said. Then “you preferentially recruit patients that have those characteristics.”
This can speed up clinical trials or reshape how they’re run.
“You’ve now essentially run the clinical trial ahead of time in the digital twins. You’ve used it to discover patient responders versus nonresponders,” Hill said. “Now you can go forward in the trial with the responding group.”