
15 Oct 2026
FDA clears SimBioSys' AI tool that predicts breast cancer recurrence from an MRI
SimBioSys said on Thursday, Oct. 15, 2026, that the FDA granted 510(k) clearance to TumorSight Risk, which it calls the world's first AI-powered, MRI-derived prognostic tool for breast cancer recurrence risk, producing individualized five- and ten-year estimates from routine breast MRI, pathology and clinical data.
Most women with this breast cancer already get an MRI. TumorSight Risk tries to squeeze a recurrence forecast out of that scan before surgery, instead of waiting on a tissue test that costs money and time. The numbers are retrospective and Mayo has a financial interest, so the real test is how oncologists use it in clinic. But an FDA-cleared prognosis from an image you already took is the kind of AI health story that matters more than any chatbot.
SimBioSys, a Chicago company, announced FDA 510(k) clearance of TumorSight Risk on Oct. 15, 2026. A 510(k) is the Food and Drug Administration path that lets a device be sold when the maker shows it is close enough to a device already allowed on the market. It is a device clearance. It is not a drug approval. Those lines are SimBioSys'.
Intended use, as the company states it: a prognostic indicator of recurrence-free survival at five and ten years in adult women with HR-positive, HER2-negative Stage I–IIIA breast cancer, to be considered alongside other clinicopathologic factors. HR-positive, HER2-negative means the tumor grows in response to hormones and does not have extra HER2, a growth protein. That subtype is roughly 75% of US breast cancers, the company says. Prognostic, here, means it estimates the chance the cancer comes back. It does not, on this filing, replace genomic tissue tests or on its own tell a doctor whether to give chemotherapy.
The software analyzes routinely collected breast MRI, pathology, and clinical data. Results can be available before surgery. SimBioSys says there is no extra procedure, tissue shipment, or specialized hardware, and that the product is cloud-based. Those lines are SimBioSys'.
The output is a proprietary 0–100 score plus a high, intermediate, or low risk category. The company says it is built on the FDA-cleared TumorSight Viz platform, using spatial AI modeling of tumor characteristics and surrounding tissue. Spatial, here, means the software looks at where things sit in the scan, not only a single lab number. Those lines are SimBioSys'.
Validation, as the release describes it: a blinded, prospectively designed retrospective multi-institutional study of 2,129 patients from Mayo Clinic, Emory University, Stanford Medicine, and Mercy Medical Cedar Rapids, presented at ASCO 2026. Blinded means the people scoring did not know the later outcomes while they scored. Retrospective means the scans and records already existed. Those lines are SimBioSys'.
Results, as the company reports them: high-risk patients had 3.3 times the recurrence risk of low-risk patients. Each 10-point score increase was associated with a 22% increase in relative recurrence risk. Estimated 5-year recurrence risk ranged from 2.2% (scores 0–10) to 13.6% (scores 58–100). Ten-year risk ranged from 6.2% to 34.2%. The score stayed significant after adjusting for stage, grade, age, race, and treatment (p=0.002). Those figures are SimBioSys'. The study is retrospective.
The company frames TumorSight Risk as a complement to genomic tissue assays, which can add cost, logistics, and turnaround time, and notes that premenopausal and node-positive women have historically had fewer validated options. Complement means it sits beside those tests. The release does not say it replaces them. Pricing and reimbursement are not disclosed.
Stacey Stevens, president and CEO of SimBioSys, called the clearance a "defining moment." Matthew Goetz, M.D., of Mayo Clinic, said MRI-derived assessments "may provide prognostic information beyond standard clinicopathologic factors." The release discloses that Mayo Clinic has a financial interest with SimBioSys. Those lines are on the company release.
The picture is SimBioSys' TumorSight 3D model built from a breast MRI, showing a tumor with slow-growing and fast-growing regions and surrounding blood vessels. It is the company's graphic. The frame does not print a calendar date.
In plain terms, SimBioSys said on Thursday that the FDA cleared TumorSight Risk, software that turns a routine breast MRI plus pathology and clinical data into five- and ten-year recurrence scores for the most common early breast cancer, often before surgery. The validation numbers are from a 2,129-patient retrospective study. Mayo Clinic, one of the study sites, has a financial interest in the company. The tool is cleared as a prognostic aid beside other clinical factors, not as a stand-in for genomic tests or a chemo decision on its own.
RELATED
- Everlywell puts Clairity’s FDA-authorized AI breast-risk score online nationwide
- AstraZeneca pays insitro up to $1.5 billion to find AI drug targets for liver and heart disease
- FDA clears Tempus ECG-MR to flag undiagnosed mitral regurgitation on a standard ECG
- FDA advisers back GRAIL’s Galleri cancer blood test on safety; split on efficacy
- AZmed wins 6th FDA clearance, expanding AZchest on chest X-rays