Lunit SCOPE Demonstrates High Concordance with Pathologists in HER2 Interpretation for Advanced Biliary Tract Cancer

Study published in Laboratory Investigation, the official journal of the United States and Canadian Academy of Pathology

SEOUL, South Korea, Dec. 23, 2025 /PRNewswire/ -- Lunit (KRX: 328130) , a leading provider of AI for cancer diagnostics and precision oncology, today announced that a new study evaluating its AI-powered pathology solution, Lunit SCOPE HER2, has been published in Laboratory Investigation, the official journal of the United States and Canadian Academy of Pathology (USCAP).

The study examined the consistency of HER2 immunohistochemistry (IHC) interpretation in advanced biliary tract cancer (BTC) by comparing assessments from pathologists using light microscopy and digital pathology with results generated by Lunit SCOPE HER2. BTC is a rare and aggressive cancer in which accurate biomarker evaluation is critical for identifying patients who may benefit from HER2-targeted therapies.

Conducted by investigators at CHA Bundang Medical Center and CHA Ilsan Medical Center in South Korea, in collaboration with Lunit researchers, the retrospective analysis included 309 HER2-stained whole-slide images from 291 advanced BTC patients treated between 2019 and 2022. Three board-certified pathologists independently evaluated HER2 expression, and their assessments were compared against AI-based scoring performed using Lunit SCOPE HER2.

According to the study, complete agreement among pathologists was observed in 62.1% of light microscopy assessments and 63.4% of digital pathology assessments, reflecting the inherent variability in HER2 interpretation. Against this background, Lunit SCOPE HER2 achieved an overall concordance rate of 83.5% with pathologist-defined ground truth, indicating close alignment with expert-reviewed interpretations.  The findings underscore the well-recognized challenges of HER2 interpretation in biliary tract cancer, particularly in cases with low or borderline expression, where variability among pathologists was more pronounced. Within this context, the study suggests that AI-based analysis may help support more consistent interpretation when integrated into digital pathology workflows.

"Accurate and consistent biomarker assessment is critical as precision oncology continues to expand into more complex and less common cancers," said Brandon Suh, CEO of Lunit. "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 been a longstanding challenge. We believe AI can play a meaningful role in supporting more standardized and reproducible decision-making as new targeted therapies emerge."

The full study, titled "Pathologist–Artificial Intelligence Concordance in HER2 Interpretation for Advanced Biliary Tract Cancer: Intraobserver, Interobserver, and Human–Artificial Intelligence Variability," is available online in Laboratory Investigation.

About Lunit

Founded in 2013, Lunit (KRX: 328130) is a global leader on a mission to conquer cancer through AI. Our clinically validated solutions span medical imaging, breast health, and biomarker analysis—empowering earlier detection, smarter treatment decisions, and more precise outcomes across the cancer care continuum.

Lunit offers a comprehensive suite spanning risk prediction and early detection to precision oncology. Our FDA-cleared Lunit INSIGHT suite and breast health solutions support cancer screening in thousands of medical institutions worldwide, while the Lunit SCOPE platform is used in research partnership with global pharma and laboratory leaders for biomarker research, and companion diagnostic development.

Trusted by over 10,000 sites in more than 65 countries, Lunit combines deep medical expertise with continuously evolving datasets to deliver measurable impact—for patients, clinicians, and researchers alike. Headquartered in Seoul with global offices, Lunit is driving the worldwide fight against cancer. Learn more at lunit.io/en .

SOURCE Lunit