Before breast cancer surgery, patients are often waiting on answers. An AI tool that reads a routine breast MRI to estimate whether cancer will come back just got a green light from the FDA. The breast cancer recurrence risk software, called TumorSight Risk, could give patients a number before they ever reach the operating room.
Chicago-based SimBioSys announced the 510(k) clearance on Oct. 15. The company calls it the first AI-powered, MRI-derived tool of its kind. That “world’s first” label is the company’s own claim, and we couldn’t verify it independently.
How this breast cancer recurrence risk tool works
The tool is cloud-based and built for adult women with HR+/HER2-negative breast cancer at stages I through IIIA. That subtype makes up about 75% of U.S. breast cancers, according to the company. It is not cleared for triple-negative or HER2-positive disease, AuntMinnie reported.
It starts with the pre-treatment MRI a patient already gets. The software then combines measurements from that scan with age, race, tumor stage, nodal stage and grade. The result is a 0-to-100 breast cancer recurrence score. It also places each patient in a low, intermediate or high risk category, with estimates for recurrence over five and ten years.
Per the FDA summary as reported by Unite.AI, a score above 35 counts as high risk. The tool is also cleared for GE and Siemens MRI scanners.
What the study showed about cancer relapse risk
The company’s validation study covered 2,129 patients from Mayo Clinic, Emory University, Stanford Medicine and Mercy Medical Cedar Rapids. It was presented at ASCO 2026, and its cancer relapse risk numbers drew the most attention.
Five-year recurrence-free survival was 95.5% for the low-risk group. For the high-risk group, it was 89.6%.
High-risk patients faced more than three times the recurrence risk of low-risk patients, according to ABC News and trade coverage. In addition, every 10-point jump in score meant a 22% higher relative risk.
Why it could matter to patients
Here’s the thing: breast cancer treatment is a string of hard decisions, and each one can feel heavy. Chemotherapy and hormone therapy both carry real side effects. A better read on breast cancer return risk could help patients and doctors weigh those trade-offs together. It is the same logic behind research on whether common vaccines protect the heart: better information up front, and a clearer sense of what remains unproven.
SimBioSys says results take under two hours and need no extra scans. By comparison, molecular tests can take days or weeks. “That insight can be available before surgery,” CEO Stacey Stevens said, “giving women and their doctors more information, sooner.”
The fine print on any recurrence risk assessment
A clearance is not a cure. Physician Matthew Kurian, whose LinkedIn commentary Oncodaily republished, put it plainly: “FDA clearance permits marketing for its intended use, not proof of improved outcomes.” That caution applies here, much as it did when the FDA cleared a Fragile X trial without approving the drug.
The validation was retrospective. The tool is also prognostic, which means it estimates risk. It does not predict whether chemotherapy will help.
No one has yet shown how this recurrence risk assessment stacks up against established genomic tests like Oncotype DX or MammaPrint.
A few details are also unsettled. One outlet put the FDA decision in late September, while the announcement came Oct. 15. Cutoffs for “high risk” also differ between sources. We left both out until the FDA record clears them up.
The rollout to U.S. hospitals and cancer centers has begun. Pricing and participating sites weren’t available. The question now is whether a score on a screen changes what patients are told to do, or just what they know.
