AI-powered breast cancer risk tool earns FDA Breakthrough Device designation

A new artificial intelligence-powered technology that analyzes mammograms to predict a woman’s five-year risk of developing breast cancer has received Breakthrough Device designation from the U.S. Food and Drug Administration (FDA). Developed by researchers at Washington University School of Medicine in St. Louis, the system is designed to enhance early detection and personalized prevention strategies. The software, named Prognosia Breast, uses machine learning to assess mammogram images and generate an individualized five-year risk score. It works with both traditional 2D digital mammograms and synthetic 3D images from digital breast tomosynthesis. Unlike conventional risk assessments that rely on questionnaires about age, race, and family history, Prognosia Breast analyzes visual patterns in the mammogram itself, producing a more accurate estimate of future risk. According to the developers, the system predicts five-year breast cancer risk 2.2 times more accurately than standard methods. It provides an absolute risk score that can be compared to national average rates, aligning with U.S. guidelines for risk reduction. A score of 3% or higher is considered elevated, triggering recommendations for additional screening or preventive measures such as more frequent imaging or medications like tamoxifen. The technology was created by Graham A. Colditz, MD, DrPH, and Shu (Joy) Jiang, Ph.D., both of Washington University School of Medicine. They co-founded Prognosia Inc., a startup launched in 2024 with support from WashU’s Office of Technology Management and BioGenerator Ventures, which provided funding and business guidance through its Entrepreneur-in-Residence, David Smoller, Ph.D. The software was trained on tens of thousands of mammograms from patients at Siteman Cancer Center, including those who later developed breast cancer. This allowed the AI to detect subtle, early signs of disease that may be invisible even to experienced radiologists. The FDA’s Breakthrough Device designation accelerates the review process, enabling faster patient access to promising technologies that address serious or life-threatening conditions. The developers emphasize that Prognosia Breast is intended to support, not replace, radiologists’ interpretations. It complements existing screening protocols, including breast density assessments. With over 75% of women aged 50 to 74 having had a mammogram in the past two years, the technology could be integrated into routine care with minimal disruption. Early detection is critical—about 34% of U.S. breast cancer cases are diagnosed at later stages, when treatment is more difficult and survival rates decline. The team plans to launch a clinical trial at Siteman Cancer Center, combining Prognosia Breast’s risk scores with standard screening. Women identified as high-risk will be referred to specialists for personalized guidance on prevention and screening options. Looking ahead, researchers aim to enhance the system by analyzing multiple mammograms over time, which could further improve predictive accuracy. The long-term goal is to make the technology available globally, improving breast cancer risk assessment for women everywhere.