Perimeter Medical Imaging AI (TSX-V:PINK; OTC:PYNKF) is poised to transform cancer surgery, as positive topline results from a pivotal study designed to support the company’s planned FDA premarket approval (PMA) submission to market its next-generation B-Series with ImgAssist AI 2.0 for use during breast-conserving surgeries (BCS) met its primary endpoint: a statistically significant reduction in patients with residual cancer during surgery.
“Perimeter’s B-Series OCT with ImgAssist AI 2.0 promises to give surgeons greater peace of mind, offering the most advanced intraoperative margin visualization available, providing real-time insights to help make critical decisions on margin status, potentially reducing re-excisions, minimizing patient trauma, and saving healthcare costs,” Adrian Mendes, CEO of Perimeter, says in an interview with BioTuesdays.
OCT is an optical imaging technique that produces ultra-high-resolution images of sub-surface tissue structures. Similar to ultrasound, OCT uses light instead of sound to create cross-sectional, 3D images—achieving ten times the resolution of ultrasound and X-ray, and 100 times that of MRI. A single beam of light rapidly scans the tissue specimen, with reflected light from an up to two-millimeter depth transformed into an OCT image.
Joining Mr. Mendes during the interview with BioTuesdays is Perimeter’s co-founder and CIO, Andrew Berkeley, and Sarah Butler, Ph.D., vice president, clinical and medical affairs.
“Intraoperative margin assessment is of critical concern for cancer surgeons, who face nearly one-in-five odds of repeat surgery due to positive margins,” Mr. Berkeley points out. “Meanwhile, under the current paradigm, patients must often endure up to a week of anxiety awaiting post-operative pathology results, only to potentially face the emotional and physical trauma of a second surgery due to cancer being left behind.”
Perimeter’s B-Series OCT with ImgAssist AI 2.0 system combines proprietary AI technology with wide-field OCT. ImgAssist is an AI-driven clinical decision support system designed to enhance productivity and decision-making during cancer surgery.
Mr. Berkeley says the portable OCT technology fits conveniently into the surgical suite, allowing it to be efficiently accessed by technical staff or surgeons during surgery. The system includes several components, including optical imaging system hardware, touchscreen software, and an intuitive user interface. To enhance image quality, the light vacuum of the specimen immobilizer helps to keep the surface of the excised tissue securely in place against the imaging glass. The single-use specimen immobilizer or container accommodates samples of up to ten centimeters in diameter and serves to keep them safely preserved for pathology.
“Clinicians have relied on high-resolution OCT imaging outside of the operating room for decades to inform their clinical decisions, and it is widely used in clinical settings ranging from ophthalmology and interventional cardiology to dermatology. To our knowledge however, Perimeter is the first company to uniquely adapt OCT technology, enabling the imaging of surgical specimens, allowing surgeons to see into the margins in real-time,” Mr. Berkeley says.
“It’s a powerful tool for visualizing surgical margins that are often microscopic, including suspicious tissue microstructures and features such as, abnormal ducts,” he adds.
Mr. Berkeley outlines that Perimeter layered its first-generation FDA 510(k)-cleared S-Series OCT system with the AI component, ImgAssist, to create the next-generation FDA breakthrough-device-designated investigational-use B-Series OCT. “While the S-Series, currently on the market, is highly effective, the B-Series offers even greater flexibility, driven by rapidly evolving AI that continuously learns, teaches, and improves. With data from S-Series users now, and soon B-Series users, the performance metrics can only get better and better, moving steadily in one direction—upward.”
“We have been training a dataset of several million proprietary images of both cancerous and healthy tissue captured with our OCT imaging technology,” Mr. Berkeley says. “Our world-class AI team has developed models tailored for real-time applications in healthcare settings and is using our vast image library to train these models to achieve an industry-leading results.”
The multi-center, pivotal clinical study, led by principal investigator Dr. Alastair Thompson at Baylor College of Medicine, evaluated the use of Perimeter B-Series OCT with ImgAssist AI 2.0 system during breast-conserving surgery. The study was conducted at some of the largest cancer centers in the U.S., including Mayo Clinic, MD Anderson Cancer Center, Moffitt Cancer Center, and Baylor College of Medicine.
Mr. Berkeley explains that the design of this pivotal trial was a within-subject study, rather than between-subject. Instead of the traditional randomized study, comparing outcomes between patients who used the device and those who didn’t, the trial was designed to assess the same patient before and after introducing the device, following standard-of-care treatment.
“To put the within-subject trial design into context,” Mr. Berkeley says, “imagine you had a blood pressure pill you wanted to test. You would first measure a patient’s blood pressure and record it, then give them the medication, followed by another blood pressure recording. This allows you to directly observe the pill’s effectiveness in that patient. That’s how we designed this trial. We removed ambiguity and focused on the single patient.”
“The within-subject model required fewer participants and provided more accurate performance data. While the study did have a single-arm design, we added a small 2:1 control ratio to minimize bias. In these cases, surgeons randomly opened an envelope to determine device use. However, we did not include the control data in the final analysis, ensuring objective performance evaluation,” he adds.
According to Dr. Butler, the pivotal study results indicated that B-Series OCT with ImgAssist AI 2.0 system enabled surgeons to more effectively address residual cancer at the surgical margin during surgery as compared to current standard methods. “The pivotal trial easily met its primary endpoint, achieving a statistically significant (p-value = 0.0050) reduction in patients with residual cancer during surgery.”
“At one point during the trial, we decided to raise the bar on the study by adding a more clinically meaningful increase in our performance goal, which we achieved. So, we didn’t just hit superiority at the primary endpoint; we hit ‘super-superiority’, meaning the confidence in the treatment is strong, and the effect is more than just a small improvement—it is meaningful and impactful for patients,” Dr. Butler contends.
Perimeter’s pivotal study results demonstrated that the B-Series OCT with ImgAssist AI 2.0 system has the potential to aid surgeons in achieving clear surgical margins during surgery, potentially lowering the need for reoperation.
“To achieve super-superiority is quite encouraging and indicates where we can ultimately take this technology in the future—not just in breast cancer but in other cancer types such as head and neck, prostate, colorectal, lung and bronchus, ovarian, kidney and renal, and liver cancers,” Mr. Berkeley asserts.
Approximately 200,000 breast lumpectomies are performed in the U.S. annually, with 21% resulting in re-excisions at an average cost of $17,000 per surgery. Together, re-excisions for the top eight cancers, including breast, head and neck, prostate, colorectal, lung, ovarian, kidney, and liver, burden the U.S. healthcare system with $1.2 billion per year.
Mr. Mendes notes that Perimeter’s OCT technology and AI are highly differentiated, as they are well-protected with multiple issued and pending patents. Additionally, the company has a solid library containing more than two million breast cancer images, along with trade secrets around how it builds its software and algorithms and manufactures its device.
“We are very pleased with these positive topline pivotal trial results, which are encouraging for this patient population undergoing BCS. Intraoperative margin assessment using OCT combined with Perimeter’s AI technology may help lower re-excision rates by identifying regions of interest and guiding real-time decisions by surgeons on margin status in the OR,” Mr. Mendes says.
“This data is just the primary endpoint to support our FDA PMA submission. There will much more information and data available from this trial over the coming months. We look forward to submitting our FDA PMA for B-Series OCT with ImgAssist AI 2.0 in early 2025, with the goal of getting the B Series to market as soon as possible,” he adds.
Based on the positive topline pivotal trial results, Mr. Mendes expresses confidence that their technology is well-positioned to make a real difference in the treatment of cancer patients. “Should the FDA approve the PMA, B-Series OCT with ImgAssist AI 2.0 has the potential for rapid acceptance and adoption by cancer surgeons throughout the U.S.,” he concludes.
• • • • •
Editor’s Note:
This information is intended for an investor audience and not for use by healthcare professionals. The information contained herein should not be used to guide clinical practice.
S-Series OCT
The S-Series OCT is indicated for use as an imaging tool in the evaluation of excised human tissue microstructure, by providing two-dimensional, cross-sectional, real-time depth visualization, with image review manipulation software for identifying and annotating regions of interest.
The S-Series OCT has 510(k) clearance under a general indication and has not been evaluated by FDA specifically for use in breast tissue, breast cancer, other types of cancer, margin evaluation, and reducing re-excision rates. The safety and effectiveness of these uses has not been established.
For full information on unapproved/off-label uses, visit: perimetermed.com/disclosures or contact [email protected].
B-Series OCT
The Perimeter B-Series OCT System is an adjunctive three-dimensional imaging tool which provides volumetric cross-sectional, real-time depth visualization, coupled with an artificial intelligence computer aided detection algorithm which identifies and marks focal areas suspicious for breast cancer and is used concurrently with physician interpretation of the images. Selene is intended for use in conjunction with other standard methods for evaluation of the margins of an excised lumpectomy specimen during surgical procedures in patients with a biopsy-confirmed diagnosis of breast cancer.
Perimeter B-Series OCT is not available for sale in the United States. CAUTION – Investigational device. Limited by U.S. law to investigational use.
• • • • •
To connect with Perimeter Medical Imaging AI or any other companies featured on BioTuesdays, send us an email at [email protected].
