Table of Contents
Artificial intelligence is becoming increasingly integrated into oncology care, with applications ranging from administrative tasks to clinical decision support. A newer evolution of this technology, agentic AI, is designed to go beyond simply generating responses or retrieving information, drawing on multiple sources to help users move through a series of decisions or actions, an approach now shaping how agentic AI clinical trial enrollment tools function in real-world oncology practice.
What Makes Agentic AI Different for Clinical Trial Enrollment
In an interview with Targeted Oncology, Shaalan Beg, MD, MBA, FASCO, chief medical officer at ConcertAI and a gastrointestinal medical oncologist, described agentic AI as tools developed to solve specific problems using the totality of available evidence, additional patient context, and general data available outside both. “What we’ve seen over the evolution of the last few years as AI tools continue to develop is that general-purpose AI doesn’t always understand the nuances of a task,” Beg said, pointing to elements like interpreting molecular profiles and oncology guidelines that require specialized training data.
How CancerLinQ Synthesizes Data for Trial Matching
Beg explained that ConcertAI’s CancerLinQ tool synthesizes data from electronic medical records, both structured and unstructured elements like notes, pathology reports, and radiology notes, supplemented by social determinants of health data and claims data. These inputs feed agents designed to determine a patient’s molecular characterization, its association with national oncology guidelines and FDA-approved medications, and clinical trials the patient may currently be eligible for.
Why Timing Matters for Agentic AI Clinical Trial Enrollment
A patient may not be eligible for a clinical trial until their cancer reaches a specific state, since many trials are designed for a first line of therapy while others target a third line. “The tools we have developed based on the data that we have access to are designed to identify when a patient’s cancer journey meets the point that a clinical trial is interested in,” Beg said, distinguishing patients eligible today from those who might become eligible in six months or two years.
Reducing Cognitive Load for Coordinators and Clinicians
For a coordinator screening patients, Beg said the tools provide information in the sequence needed and prioritize which patients are more likely to be eligible, without removing the human prescreening and screening activities themselves. For clinicians, the tools flag care gaps, such as an untested biomarker, surfacing missing information rather than requiring physicians to remember and check every relevant biomarker manually.
The Role of Human Oversight in Agentic AI Clinical Trial Enrollment
“It’s not taking away the need that a human will have to look at those records and talk to a patient, explain the study, and have to get a signature on the trial, but it reduces the screening activity from 1 hour or 90 minutes down to a few minutes,” Beg said. He does not currently see these tools as fully automating eligibility determination or executing consent without human oversight for most oncology use cases, though he noted more pragmatic trial designs are beginning to employ greater automation for certain outcome-focused studies.
What Gets Missed Without These Tools
Beg described a common scenario: a new patient with records scattered across multiple hospitals, progress notes scanned under a media tab, and molecular reports still arriving by fax. Collecting that information manually takes significant time, and things get missed or misinterpreted, he said, which is precisely the gap these tools are designed to close by reviewing and ranking patients by likelihood of matching.
How Agentic AI Clinical Trial Enrollment Could Address Access Disparities
Asked whether these tools could help address existing enrollment disparities for patients treated outside academic centers, Beg said “one hundred percent,” pointing to the challenge many academic centers face in staffing community satellite facilities with clinical research staff. By delivering these tools, research sites can extend their reach to oversee activity at satellite centers, screening patients seen closer to home with community clinicians at the same time as those coming to the main academic site.
Distance as a Documented Enrollment Barrier
Beg noted that disparities in trial enrollment involve race, ethnicity, insurance status, and zip code, and that distance from a clinical trial site remains one of the biggest known barriers to enrollment. Tools that extend screening capability to community settings directly target this specific, well-documented obstacle rather than addressing disparities in the abstract.
What’s Needed for Agentic AI Clinical Trial Enrollment to Scale
Beg said “success begets success,” predicting that as more of these implementation stories become public, trust in the tools will increase alongside continued improvement from user feedback. He identified implementation itself, not just tool development, as one of the central challenges, noting that many health centers are still defining their governance for informatics, AI, and agentic AI tools, with some states also introducing regulations affecting deployment.
Why Consolidated Tool Suites May Win Out
Beg predicted that tools offering multiple solutions through a single implementation process are most likely to succeed, drawing a comparison to how bundling Word, Excel, and PowerPoint into one Microsoft Office deployment simplified adoption decades ago. He noted that ConcertAI and CancerLinQ support both clinical care and clinical research simultaneously, reflecting how in oncology, unlike many other diseases, the same physician often manages both a patient’s clinical care and their potential trial participation.
What This Agentic AI Clinical Trial Enrollment Approach Means Going Forward
Given Beg’s emphasis that implementation and governance, not just tool capability, represent the primary barrier to scaling these systems, health systems and community oncology practices considering agentic AI trial-matching tools may need to prioritize building clear informatics and AI governance frameworks before deployment. Given the tools’ demonstrated ability to extend research site oversight into satellite and community locations, this approach could meaningfully narrow the well-documented distance-based enrollment gap that has historically limited trial access for patients treated outside major academic centers.
What to Watch Going Forward
As more health systems adopt agentic AI tools for clinical trial screening, industry observers will likely watch whether measurable increases in community-site trial enrollment follow, particularly given Beg’s specific framing of distance as one of the largest known barriers to participation. Given his prediction that consolidated, multi-solution platforms will outcompete single-purpose tools, this agentic AI clinical trial enrollment landscape may increasingly consolidate around vendors like ConcertAI that can support both clinical care and research workflows through a unified implementation, rather than requiring health systems to adopt separate point solutions for each distinct oncology AI use case.
For more healthcare industry updates, insights and news, visit DistilINFO. Click here to subscribe to stay informed.
