Only 7% of Cancer Patients Join Clinical Trials—New AI Tool Could Change That

A groundbreaking artificial intelligence (AI) platform launched by Mount Sinai Tisch Cancer Center could dramatically increase cancer patient participation in clinical trials, which currently sits at just 7% despite thousands of available studies each year. The new system, called PRISM, uses specialized AI to automatically scan electronic health records and match patients with relevant clinical trials, eliminating the time-consuming manual process that has historically limited access.

Why Do So Few Cancer Patients Join Clinical Trials?

Despite approximately 10,000 new cancer patients being seen across the Mount Sinai Health System annually, enrollment in clinical trials remains surprisingly low. Several significant barriers prevent patients from accessing these potentially life-saving studies.

• Time and Transportation: Many patients face logistical challenges getting to trial locations and managing appointment schedules
• Lack of Awareness: Patients often don't understand what clinical trials are or how they work, missing opportunities for cutting-edge treatments
• Physician Knowledge Gaps: Doctors may be unaware of ongoing trials that their patients could qualify for, limiting referral opportunities
• Manual Screening Process: The traditional method of reviewing patient records to identify trial candidates is labor-intensive and inconsistent

"Many patients and physicians are not aware that they might be eligible for a clinical trial," said Dr. Karyn Goodman, professor and vice chair of Clinical Research in the Department of Radiation Oncology at the Icahn School of Medicine at Mount Sinai.

How Does the New AI Platform Work?

The PRISM platform, developed by Triomics and powered by their OncoLLM AI framework, represents a significant technological advancement in clinical trial matching. The system uses an oncology-specific large language model to automatically review electronic medical records stored in EPIC systems, identifying patients who meet specific trial criteria without human intervention.

"This automated system will reduce the manual work to review patient records to find those who might be eligible for a clinical trial," explained Dr. Goodman. The AI considers multiple patient factors including age, demographics, medical history, and comorbidity situations to create more comprehensive matches than traditional methods.

What Could This Mean for Cancer Treatment Access?

The implications of improved clinical trial matching extend far beyond simple enrollment numbers. Clinical trials have historically led to major therapeutic breakthroughs in cancer treatment, and increased participation could accelerate the development of new therapies while providing patients access to cutting-edge treatments not yet available to the general public.

Dr. Goodman emphasized that the platform aims to "enhance equity and access to studies, increase participant diversity, reduce staff burden, and lower recruitment costs." The system could particularly benefit underserved populations who have traditionally faced greater barriers to clinical trial participation.

Medical oncologist Dr. Nilesh Vora from MemorialCare Todd Cancer Institute noted that this approach builds on existing capabilities, comparing it to next-generation sequencing tests that already provide molecular tumor profiling with matching clinical trial suggestions. However, the Mount Sinai system goes further by incorporating comprehensive patient demographics and medical history for more precise matching.

The success of this AI-driven approach at Mount Sinai could pave the way for broader implementation across healthcare systems nationwide, potentially transforming how cancer patients access experimental treatments and accelerating the pace of medical breakthroughs in oncology research.