AI Is Transforming How Patients Find Clinical Trials
New AI tools are making it faster and easier to match patients with clinical trials, a breakthrough that could help solve the industry's long-standing recruitment problem. That is, 80% of clinical trials face delays due to recruitment challenges, slowing down medical breakthroughs and limiting patient access to new treatments. Institutions like the NIH and Mass General Brigham have built and begun to implement patient-trial matching AI tools: TrialGPT and RECTIFIER, respectively.
TrialGPT has developed an AI algorithm that takes a patient summary containing relevant medical and demographic information, identifies clinical trials in ClinicalTrials.gov for which the patient is eligible, then returns a list of those trials ranked by relevance and eligibility. Similarly, RECTIFIER uses generative AI to screen a patient’s electronic health record and determine if they are eligible for a clinical trial.
Real-world results are proving the tools work. In a pilot user study, researchers from the NIH’s National Library of Medicine (NLM) and National Cancer Institute compared clinicians who manually assessed patients’ eligibility for clinical trials with those using TrialGPT. They found that clinicians using TrialGPT spent 40% less time screening patients while maintaining the same level of accuracy in determining patients’ eligibility. Additionally, in a randomized study of about 4,500 patients by JAMA, RECTIFIER nearly doubled enrollment rates for an ongoing heart failure (HF) implementation trial (NCT05734690) compared to manual screening, with no significant differences across race, gender, or ethnicity.
These tools are scaling quickly. On one hand, the NLM was awarded the 2024 Director’s Challenge Innovation Award to continue assessment on the TrialGPT model’s performance and usability in real clinical settings. On the other, RECTIFIER is currently being used not only to determine patients’ eligibility for clinical trials, but also to accelerate patient identification and reduce the burden of manual chart review in clinical and population health in cardiology, oncology, gastroenterology, neurology, pathology, psychiatry, and more.
At FindMyClinicalTrial, we applaud researchers and developers at the NIH and Mass General Brigham for their contribution to speeding up patient-trial matching and, more largely, streamlining recruitment. With tools that make it easier and quicker for clinicians to find relevant clinical trials for their patients, the difficulty that trials face to recruit patients in limited time may be alleviated.
Further, we seek to facilitate the connection between tools like TrialGPT and RECTIFIER and clinical trial enrollment and participation. We aim to make patient–trial matching accessible to all stakeholders – clinicians, patients, and researchers – while also delivering AI tools that streamline enrollment and simplify participant monitoring.
Most importantly, FindMyClinicalTrial seeks to provide these kinds of tools through a healthcare equity lens. While we have a goal to streamline recruitment for trials, we also have a goal to increase the representation of historically underserved populations in clinical trials. By democratizing access to tools like TrialGPT and RECTIFIER and by providing specialized tools for underrepresented populations, we hope to expand equitable access to promising therapies.
Sources: