NEW YORK – Ibex Medical Analytics on Tuesday said it has signed an agreement with AstraZeneca and Daiichi Sankyo to develop an artificial intelligence tool for assessment of HER2 immunohistochemistry scoring in breast cancer.
Such scoring is usually carried out visually by pathologists using a microscope, a subjective assessment that can lead to inconsistent results. Ibex's Galen Breast HER2 system detects tumor regions and quantifies HER2 expression into four categories — IHC 0, 1+, 2+, and 3+ — according to the 2018 guidelines from the American Society of Clinical Oncology and College of American Pathologists.
In the collaboration, AstraZeneca and Daiichi Sankyo will work with Ibex to develop and clinically validate the product and conduct studies supporting adoption of the technology. The partners didn't disclose the financial details of the deal. "This collaboration will allow pathologists to utilize our technology to optimize breast cancer diagnosis and ultimately improve the identification of patients eligible for HER2-directed therapy," Ibex CEO Joseph Mossel said in a statement.
In one study looking to validate the Galen Breast HER2 system's ability to score 453 breast tumors, Ibex found that the underlying algorithm provided accurate automated assessments of HER2 status with an accuracy of 82.1 percent agreement with pathologists' assessment. Those results were presented at the 2022 San Antonio Breast Cancer Symposium.
The collaboration to advance this AI tool to improve HER2 IHC scoring comes at a time when more breast cancer patients with a range of HER2 tumor expression can receive HER2-targeted therapy. Historically, only patients with high HER2 expression could receive such treatment. However, US regulators approved AstraZeneca and Daiichi Sankyo's Enhertu (trastuzumab deruxtecan) last year for patients with unresectable or metastatic breast cancer who have received chemotherapy and have low HER2-expressing tumors with IHC 1+ scores or are IHC 2+ and in situ hybridization negative.
The Tel Aviv, Israel-based firm is also working to expand the Galen Breast system to include quantification of other breast cancer biomarkers, such as estrogen receptor, progesterone receptor, and Ki-67, on IHC slides.