AI Accelerates Rare Disease Clinical Trial Enrollment, Improves Diversity

A new study demonstrates that artificial intelligence (AI) can significantly speed up and improve the accuracy of identifying potential participants for rare disease clinical trials, while also increasing diversity in enrollment. The research, a collaboration between Cleveland Clinic and Dyania Health, was published in The Journal of Cardiac Failure and highlights the potential of medically trained large language models (LLMs) to overcome longstanding barriers to clinical trial recruitment.

The Challenge of Clinical Trial Enrollment

Clinical trial recruitment is a major bottleneck in medical research. Approximately 80% of clinical trials fail to meet enrollment timelines, and 50% of trial sites don’t enroll any patients at all. This delay can hinder the development of new therapies and limit access to potentially life-saving treatments.

How AI is Transforming the Process

The study focused on DepleTTR-CM, a Phase 3 trial for transthyretin amyloid cardiomyopathy (ATTR-CM), a type of heart failure primarily affecting older adults. Dyania Health’s AI system, Synapsis AI, was used to pre-screen electronic medical records (EMRs) for eligible patients across 25 hospitals and 250 outpatient centers in Ohio, Florida, and Nevada.

The AI system analyzed both structured data and complex clinical notes using natural language processing. It provided detailed justifications for its inclusion or exclusion decisions, allowing clinicians to verify eligibility with confidence. The system achieved 96.2% accuracy when answering 7,700 trial-specific questions.

Key Findings of the Study

• Speed: The AI-assisted screening process enrolled seven patients in just six days, compared to 10 patients over 90 days using traditional screening methods.
• Accuracy: The AI system correctly excluded 198 out of 200 ineligible patients, achieving a 99% negative predictive value.
• Increased Identification of Eligible Patients: 29 of the 30 patients identified as trial matches through the AI system and clinician review had not been identified through traditional recruitment methods.
• Improved Diversity: 36.6% of patients identified by the AI were Black, significantly higher than the 7.1% identified through routine screening.
• Expanded Access: Only 60% of AI-identified patients were already connected to a heart failure specialist, compared to 92.8% identified through traditional methods, suggesting AI can reach underrepresented populations.

Expert Perspectives

“This study shows how medically trained AI can support chart review at scale, transforming what has traditionally been a labor-intensive process,” said Trejeeve Martyn, M.D., lead study investigator and director of Heart Failure Population Health at Cleveland Clinic. “By rapidly identifying high-quality trial candidates across a large health system, we can increase enrollment efficiency and increase enrollment of patients from different backgrounds and from a broader geographical area.”

Eirini Schlosser, CEO and Co-founder of Dyania Health, added, “Clinical research is often limited by how efficiently and equitably we can match patients to trials. This study provides compelling evidence that AI can help solve that bottleneck – not just by improving workflow efficiency, but by helping surface eligible patients who may otherwise be missed, especially those from historically underrepresented groups.”

Future Implications

Cleveland Clinic and Dyania Health are exploring expanding the leverage of AI-enabled tools for clinical trial matching, population health registries, and real-time quality reporting. This collaboration reflects a growing trend toward leveraging AI to accelerate medical research and improve patient care. Cleveland Clinic has invested in Dyania and may benefit financially from the sale of this technology.