#South Korea #Seoul #Lunit #HER2 #Biliary Cancer

Lunit SCOPE's Groundbreaking Study on HER2 Interpretation

In a significant advancement in cancer diagnostics, Lunit, a prominent South Korean firm specializing in artificial intelligence for oncology, has unveiled the results of a study concerning its Lunit SCOPE HER2 technology. This study, published in the esteemed journal Laboratory Investigation, underscores the high level of concordance between the AI-driven assessments of HER2 markers in advanced biliary tract cancer and the evaluations conducted by experienced pathologists.

Study Overview

The research specifically looked into the reliability of HER2 immunohistochemistry (IHC) interpretation, which plays a crucial role in determining the eligibility of patients for HER2-targeted therapies. Biliary tract cancer (BTC), characterized by its rarity and aggressive nature, poses challenges in biomarker evaluation, making precise assessment essential for effective treatment. The study involved a comprehensive analysis of 309 HER2-stained whole-slide images sourced from 291 patients diagnosed with advanced BTC between 2019 and 2022. Conducted at CHA Bundang Medical Center and CHA Ilsan Medical Center in South Korea alongside Lunit researchers, the methodology included detailed evaluations by three board-certified pathologists who assessed HER2 expression levels.

Results and Findings

Upon comparing the pathologists' interpretations with results generated by Lunit SCOPE HER2, it became evident that there exists a notable variability in HER2 evaluations. Specifically, the pathologists displayed complete agreement in only 62.1% of evaluations via light microscopy and 63.4% using digital pathology. In contrast, the AI tool achieved an impressive concordance rate of 83.5% with the established ground truths defined by pathologists. This discrepancy highlights the intricate challenges faced in HER2 interpretation, especially in instances where low or borderline expression is observed.

Implications for Precision Oncology

As precision oncology continues to transcend into the realms of less common and more complex cancers, the demand for accurate and repeatable assessments of biomarkers has only intensified. The findings from Lunit's study suggest that integrating AI-powered analysis within digital pathology workflows could provide a solution to standardizing HER2 interpretation, thereby enhancing the overall reliability of biomarker evaluation in biliary tract cancer. Brandon Suh, CEO of Lunit, highlighted, “This study shows that AI-powered pathology can closely align with expert pathologists in HER2 interpretation for biliary tract cancer, an area where variability has long posed challenges. We believe AI can play a meaningful role in supporting more standardized and reproducible decision-making.”

Future Directions

The full research titled "Pathologist–Artificial Intelligence Concordance in HER2 Interpretation for Advanced Biliary Tract Cancer Intraobserver, Interobserver, and Human–Artificial Intelligence Variability" is now accessible online in Laboratory Investigation. Lunit, founded in 2013, aims to utilize advanced AI technology to aid in cancer diagnosis and treatment, showcasing its innovative solutions across medical imaging and biomarker research. With over 10,000 installations worldwide, Lunit's mission to combat cancer is gaining traction globally, combining deep medical insights with cutting-edge technology to transform the landscape of cancer care.

For further insights into Lunit's initiatives and the potential applications of its AI technologies in cancer diagnostics, visit Lunit's official website.