Study Reveals AI's Role In Boosting Breast Cancer Diagnosis Accuracy

In a significant stride towards enhancing breast cancer diagnosis and treatment, Lunit, a frontrunner in AI-powered cancer diagnostics and therapeutics, has unveiled findings from a new study. This study, conducted in collaboration with Prof. So-Woon Kim of Kyung Hee University College of Medicine and Prof. Minsun Jung of Yonsei University College of Medicine, showcases the efficacy of Lunit SCOPE HER2 and Lunit SCOPE ER/PR. Published in the Breast Cancer Research journal, the research underscores the potential of these AI analyzers in improving the accuracy and consistency of pathologists in classifying breast cancer molecular subtypes.

At the heart of breast cancer treatment and prognosis lies the precise determination of human epidermal growth factor receptor 2 (HER2), estrogen receptor (ER), and progesterone receptor (PR) expression levels. Traditional immunohistochemistry (IHC) analysis, however, often falls short in accuracy due to variability in interpretation among pathologists, particularly with intermediate expressions. Lunit SCOPE HER2 and Lunit SCOPE ER/PR, developed from thousands of IHC-stained breast cancer images, aim to bridge this gap.

The study involved an external validation cohort of 201 breast cancer cases analyzed using these AI tools. The results were telling: agreement on HER2 status improved from 49.3% to 74.1%, on ER status from 93.0% to 96.5%, and on PR status from 84.6% to 91.5%. This enhancement in concordance was especially notable in intermediate weakly positive cases for HER2 2+ and HER2 1+, where agreement surged from 46.2% to 68.4% and from 26.5% to 70.7%, respectively.

Consequently, the overall agreement among pathologists on the classification of breast cancer molecular subtypes leaped from 58.2% to 78.6%. "This study not only validates the performance of Lunit SCOPE HER2 and Lunit SCOPE ER/PR but also highlights the potential application of AI biomarkers in predicting treatment responses, especially for HER2-targeted therapies," remarked Brandon Suh, CEO of Lunit.

Lunit's mission is clear: to transform the landscape of personalized cancer diagnostics and therapeutics through precision and informed decision-making facilitated by AI technology. Founded in 2013, Lunit has been at the forefront of integrating artificial intelligence into healthcare to ensure accurate diagnosis and optimal treatment for each cancer patient.

The company's flagship Lunit INSIGHT suite has received FDA clearance and the CE Mark, marking its clinical use in over 3,000 hospitals and medical institutions across more than 40 countries globally. With its headquarters in Seoul, South Korea, Lunit continues to expand its reach with offices and representatives worldwide.

Lunit SCOPE represents a suite of AI-powered software designed for digital pathology and AI biomarker development. It aims at optimizing workflow and providing more accurate predictive clinical data for clinicians and researchers alike. The platform offers various AI-powered tissue analysis products that streamline digital pathology workflow, diagnostics, and enhance drug development processes.

Through its comprehensive approach towards leveraging AI in medical imaging analytics and biomarker development, Lunit is setting new standards in personalized cancer care, ensuring that each patient receives the most informed and effective treatment strategy possible.