SimBioSys 的 TumorSight Risk 獲 FDA 510(k) 許可，用於估算乳癌復發風險

SimBioSys announced on October 15, 2026 that the U.S. Food and Drug Administration has granted 510(k) clearance to TumorSight Risk, an artificial intelligence software device that analyzes breast MRI scans alongside pathology and clinical data to estimate a patient’s risk of breast cancer recurrence at five and ten years.

In its announcement, the Chicago-based company described TumorSight Risk as the world’s first AI-powered, MRI-derived prognostic tool for assessing breast cancer recurrence risk. The software is intended for use as a prognostic indicator of recurrence-free survival at five and ten years in adult women with hormone receptor-positive, HER2-negative Stage I–IIIA breast cancer, with results to be considered alongside other relevant clinicopathologic factors when assessing risk and planning care. The company states that this HR-positive, HER2-negative subtype represents approximately 75 percent of all breast cancers in the U.S.

How the Software Works

The cloud-based platform applies proprietary spatial AI modeling to tumor characteristics, surrounding tissue context and clinical factors, producing a proprietary score from zero to 100 and a categorical high-, intermediate- or low-risk classification. It uses data already collected as part of routine care, requires no additional procedure, tissue shipment or specialized hardware, and is designed to make results available within standard clinical workflows. The product is built on the company’s previously cleared TumorSight Viz foundation.

According to the FDA’s 510(k) clearance letter and summary, the device operates in sequence with TumorSight Viz. A patient’s pre-treatment dynamic contrast-enhanced MRI, acquired according to American College of Radiology guidelines, is first processed by TumorSight Viz, which generates a multi-tissue segmentation map. That map passes automatically into TumorSight Risk, which computes intermediate imaging values and combines them with age, race, clinical tumor stage, clinical nodal stage and tumor histologic grade in a regression-based machine learning model. Hormone receptor and HER2 status are used by the software only to confirm that a case meets the indications for use and are not inputs to the algorithm itself. The output is a report containing the score, the risk category, and observed five- and ten-year recurrence risks from the clinical validation dataset. A score above 35 is classified High, a score above 20 through 35 is Intermediate, and a score of 20 or below is Low. The device is cleared for use with GE 1.5T, GE 3T, Siemens 1.5T and Siemens 3T MRI scanners.

The FDA Record

The clearance letter is dated September 25, 2026, and the premarket notification was dated and received January 5, 2026. The FDA determined the device substantially equivalent to a legally marketed predicate, permitting it to be marketed subject to the general controls provisions of the Federal Food, Drug, and Cosmetic Act. TumorSight Risk is a Class II prescription device under the FDA regulation covering software algorithm devices that analyze digital images for cancer prognosis, product code SHW. The summary names ArteraAI Breast, an Artera Inc. device that analyzes scanned histopathology whole-slide images to estimate five- and ten-year risks of distant metastasis, as the predicate, and concludes substantial equivalence on the basis of the same intended use, similar indications and shared technological characteristics. The TumorSight Risk algorithm was trained and tuned on 1,133 samples from six geographically distributed U.S. sites and was locked after training, before analytical precision testing and clinical validation.

Validation Evidence

The clearance is supported by a blinded, prospectively designed retrospective, multi-institutional study of 2,129 patients from Mayo Clinic, Emory University, Stanford Medicine and Mercy Medical Cedar Rapids; the company states the results were presented at the 2026 American Society of Clinical Oncology Annual Meeting. SimBioSys reported that patients classified as high risk had a 3.3-fold greater risk of recurrence than those classified as low risk, and that each 10-point increase in the score was associated with a 22 percent increase in relative recurrence risk. Estimated five-year recurrence risk ranged from 2.2 percent for patients with the lowest scores (zero to 10) to 13.6 percent for those with the highest scores (58 to 100), while ten-year estimates ranged from 6.2 percent to 34.2 percent across the same score groups. The score remained significantly associated with recurrence-free survival after adjustment for standard prognostic information including pathologic tumor and nodal stage, grade, age, race and treatment (p=0.002).

The FDA summary records that the validation population was diagnosed between January 1, 2010 and December 31, 2024 across four U.S. sites. It reports a five-year negative predictive value of 97.0 percent and a ten-year value of 91.4 percent, alongside positive predictive values of 9.8 percent at five years and 26.4 percent at ten years. Estimated recurrence risk by category was 3.0 percent at five years and 8.6 percent at ten years for low-risk patients, 5.7 and 13.4 percent for intermediate-risk patients, and 9.8 and 26.4 percent for high-risk patients. A precision study ran 120 patient images, split evenly across the four cleared scanner and field-strength combinations, through the system three times each with 3 percent added simulated noise; 119 images processed successfully, and the summary reports 99.4 percent agreement across replicate runs. A separate human factors study with 30 participating physicians and clinical support staff completed all critical tasks, according to the summary.

Statements From SimBioSys and Mayo Clinic

“Achieving FDA clearance of TumorSight Risk marks a defining moment for SimBioSys and for precision breast cancer care,” said Stacey Stevens, President and CEO of SimBioSys. “For the first time, clinicians can use the rich biological and spatial information already captured in a patient’s breast MRI—together with standard pathology and clinical factors—to gain personalized insight into recurrence risk.”

Stevens said the product expands the TumorSight platform from surgical visualization into prognostic intelligence. Matthew Goetz, M.D., Enterprise Deputy Director of Translational Research and Director of the Mayo Breast Cancer SPORE at Mayo Clinic Rochester, said the multi-institutional validation data show that MRI-derived assessments may provide prognostic information beyond standard clinicopathologic factors. The company positions the tool as a complement to existing tissue-based genomic assays, offering an imaging-native assessment available earlier in the care pathway. The announcement notes that Mayo Clinic has a financial interest with SimBioSys.