07/04/2025
Ottawa, Ontario — Monday April 7, 2025
For researchers leading clinical trials, recruiting enough participants can be a challenge. When recruitment targets for clinical trials are not met, due to cuts in funding, studies being shut down, or difficulty finding eligible patients, the trial can be abandoned altogether or redesigned to accommodate the smaller data sets, resulting in less impactful conclusions than originally intended.
Abandoning clinical trials can be seen as unethical and unfair to the individuals who give their time and active participation in a trial, only to have no results or conclusions reached.
But what if there was a way to salvage clinical trials that may not meet recruitment targets? Dr. Khaled El Emam, Senior Scientist at the CHEO Research Institute and Canada Research Chair (Tier 1) in Medical AI at the University of Ottawa, sought to answer this exact question with the largest study to date that tested whether synthetic data generation (SDG) could accurately supplement recruitment gaps in clinical trials.
The study took nine completed breast cancer clinical trials from around the world, including some from The Ottawa Hospital Research Institute (OHRI), and re-created them by replacing a portion of the participants with synthetic data – also known as a digital twin. The analysis and results using the synthetic data were then compared to the original results from the clinical trials – in partnership with the original study teams.
The simulation worked “incredibly well” shared Dr. El Emam, with findings showing that as long as the original patient group remained at least 60% human, synthetic data was able to replicate findings with high accuracy.
“At this stage, it’s hard to say whether synthetic data will change the way we do clinical trials altogether or just be another tool in the toolbox we can depend on when recruitment targets cannot be achieved,” said Dr. El Emam. “I am hopeful of this becoming a trusted tool to mitigate common barriers and pitfalls in clinical trials, that is accepted not only by researchers but regulators alike. But there is more work to do – we need to continue to engage patient partners, clinicians, researchers, and regulators in discussions about the use of synthetic data to optimize clinical trials.”
Working closely with the medical oncology group and REaCT team at OHRI, Dr. El Emam gathered the largest international team thus far to successfully explore using synthetic data in clinical trials. They are now working on testing synthetic data on an even larger scale, with hopes to complete 100 trial simulations to see how it works across different diseases, patient populations, and study sizes.
“AI is revolutionizing society and the REaCT team is delighted to present this pivotal research in collaboration with CHEO and national and international groups,” said Dr. Mark Clemons, medical oncologist and scientist at The Ottawa Hospital and professor at the University of Ottawa. “This work is striving to improve clinical trials by making them more efficient so that better treatments reach patients sooner.”
Synthetic Data Summit 2025
On May 16, 2025, expert speakers from across industry, government, academia, and regulatory agencies are gathering for the 2025 Synthetic Data Summit at the PHI Centre in Montreal. The summit will explore how synthetic data is being used to address real-world data challenges and privacy problems of data access and sharing and examine its future applications in health care.
“By using AI to create high-quality datasets that do not contain personal information, synthetic data generation (SDG) enables data to be shared and analyzed, protecting privacy. SDG also helps amplify and fill gaps in datasets to improve research outcomes, as well as reduce bias in clinical trials and real-world evidence studies. The 2025 Synthetic Data Summit is a unique opportunity to gain a well-rounded perspective on how synthetic data is transforming health care now, and in the years ahead,” said Dr. El Emam, who as Director of the Electronic Health Information Laboratory, based at the CHEO Research Institute, is hosting the summit.
For additional summit details, including the full program and registration information, visit the 2025 Synthetic Data Summit Eventbrite page.
The Synthetic Data Summit is co-sponsored by Aetion, the global leader in real-world evidence (RWE) technology and analytics.