UCLA Launches First Large-Scale U.S. Trial Testing AI in Breast Cancer Screening
Key Insights
UCLA and UC Davis (search) will co-lead the PRISM Trial, a $16 million randomized controlled study evaluating artificial intelligence support for mammogram interpretation across seven academic medical centers.
The study represents the first large-scale randomized trial of AI in breast cancer (search) screening in the United States, involving hundreds of thousands of mammograms across California, Florida, Massachusetts, Washington, and Wisconsin.
Researchers aim to determine whether AI tools help radiologists detect more cancers or simply increase false positives, addressing a critical evidence gap as AI-assisted mammograms become increasingly common in clinical practice.
UCLA and UC Davis (search) will co-lead a groundbreaking $16 million clinical trial to evaluate whether artificial intelligence can improve breast cancer (search) screening accuracy while reducing unnecessary patient callbacks and anxiety. The PRISM Trial (Pragmatic Randomized Trial of Artificial Intelligence for Screening Mammography), funded by the Patient-Centered Outcomes Research Institute (search) (PCORI), represents the first large-scale randomized controlled trial of AI in breast cancer screening in the United States.
The study will analyze hundreds of thousands of mammograms across seven academic medical centers in California, Florida, Massachusetts, Washington, and Wisconsin. The AI support tool being evaluated is Transpara (search) by ScreenPoint Medical (search), with clinical workflow integration provided by the Aidoc (search) aiOS platform.
Addressing Critical Evidence Gaps
"This is the first large-scale randomized trial of AI in breast cancer (search) screening in the United States," said Dr. Joann G. Elmore, Dual Principal Investigator and professor of medicine at the David Geffen School of Medicine at UCLA. "We're looking carefully and objectively at whether AI helps or hinders — and for whom. Expert radiologists remain in the driver's seat for all interpretations."
The trial addresses a significant gap in evidence for technology already applied to millions of mammograms annually. While AI-assisted mammograms are becoming more common in breast imaging practices, clinicians struggle to understand their impact on patients. European prospective trials have suggested AI tools can increase cancer detection rates, but those results cannot be easily extrapolated to the U.S., where mammograms are typically read by a single radiologist instead of Europe's two-radiologist standard.
"Companies are pushing hard to incorporate AI into practice," said Diana Miglioretti, co-lead of the U.S. Breast Cancer (search) Surveillance Consortium, who will lead the trial's data coordinating center at UC Davis (search) Health. However, most evidence for the tools is based on "artificial settings" where results don't impact women directly.
Patient-Centered Research Design
The PRISM trial distinguishes itself through its emphasis on patient-centered research, developed in close partnership with patient advocates, clinicians, health system leaders, and policymakers. Each participating facility will continue routine screening procedures with no changes to the patient experience.
Mammograms will be randomly assigned for interpretation either by a radiologist independently or with assistance from the FDA-cleared AI support tool. In all cases, a radiologist reads the exam and makes the final decision, ensuring human expertise remains central to the process.
"There's never been a trial of this scope looking at AI in breast cancer (search) screening in the U.S.," said Dr. Hannah Milch, Co-Principal Investigator and UCLA Site PI and assistant professor of radiology at UCLA. "The results will help inform not just clinical practice, but also insurance coverage, technology adoption, and patient communication."
Evaluating Real-World Impact
The study aims to resolve critical questions about AI's role in breast cancer (search) screening. Breast cancer remains one of the leading causes of cancer death among women in the U.S. While routine mammography screening reduces mortality through early detection, it also has drawbacks including false positives that can lead to unnecessary testing, anxiety, and costs, as well as missed cancers.
"AI has great promise, but it also raises real questions," said Elmore, who is also an investigator at the UCLA Health Jonsson Comprehensive Cancer Center. "We want to know whether AI helps radiologists find more cancers, or just flags more exams that ultimately turn out to be normal."
Beyond analyzing cancer detection and recall rates, the study will include focus groups and surveys to capture how patients and radiologists perceive and trust AI-assisted care.
Multi-Institutional Collaboration
The PRISM trial brings together seven leading academic medical centers: UCLA (Administrative Coordinating Site), UC Davis (search) (Data Coordinating Center), Boston Medical Center, UC San Diego Health (search), University of Miami, University of Washington – Fred Hutchinson Cancer Center, and University of Wisconsin–Madison (search).
"There is a lot of hope that AI will make care better, but very few rigorous trials have actually evaluated its real-world effects," said Elmore. "This is our opportunity to generate trustworthy evidence, with the patient perspective front and center."
The trial is expected to inform future policy decisions, best practices in screening, and effective integration of emerging technologies into patient care, with results potentially influencing clinical practice standards nationwide.