An AI-imaging device will be the first in the US to be used to improve surgeons’ ability to detect difficult-to-see cancer during breast-conserving surgery and may reduce the need for reoperations.
The FDA has granted premarket approval to Claire (formerly known as Perimeter B-Series OCT with ImgAssist AI 2.0).
It is designed for intraoperative breast cancer margin assessment in patients with stage 0 to III invasive ductal carcinoma and/or ductal carcinoma in situ.
Claire can deliver 10 times higher resolution than an X-ray or ultrasound at a depth of 2mm.
The technology was developed and trained using Perimeter’s proprietary, expanding optical coherence tomography image library, which contains over 2 million breast-tissue images.
During surgery, Claire can assist in determining whether tissues should be removed before the procedure concludes.
Additionally, a predetermined change control plan was authorised as part of the premarket approval.
This will enable AI improvements without requiring further FDA authorisation or interaction. As more surgeries are performed, Perimeter’s database grows, providing more data to enhance the product and improve outcomes for future patients.
Alastair Thompson is a surgeon and professor, section chief of breast surgery, and Olga Keith Wiess Chair of Surgery at Baylor College of Medicine. He is also the leader of the Breast Cancer Program at the Dan L Duncan Comprehensive Cancer Centre and the primary principal investigator of the pivotal trial that supported Claire's premarket approval application.
He said: ‘Despite progress in breast cancer treatment, intraoperative margin assessment remains challenging, often leading to the removal of healthy tissue, reoperations, and anxiety for patients awaiting pathology results. Claire has the potential to become a new standard in breast surgical care, helping to reduce re-excisions while improving patient outcomes.’
Thompson shared data on Claire at the 26th Annual Meeting of the American Society of Breast Surgeons. Out of 206 patients, 35 had residual disease after standard lumpectomy alone.
Claire identified residual disease in 40% of cases and cleared 20% of additional patients, meeting the primary goal. The average tissue removed was 74cm³, with Claire accounting for 2.8cm³, and Claire's overall accuracy was 88.1%.
A total of 26 patients showed a clinical benefit rate from Claire’s residual disease detection after SOC was completed. Notably, six patients with Claire-aided shaves had pathology-confirmed disease that SOC missed, and prior histopathology at the margin.
Thompson noted that the trial results show Claire can change the paradigm by helping surgeons identify areas of interest, improve intraoperative decisions, and lower re-excision rates due to residual disease after lumpectomy.
