News | Digital Pathology | September 22, 2023

Lunit SCOPE IO Enhances Breast Cancer Analysis and Treatment by Accurately Assessing Tumor-Infiltrating Lymphocytes

Lunit SCOPE IO transforms pathologist discordance into concordance and predicts treatment response for breast cancer, enhancing the landscape of personal oncology care

Lunit AI-powered TIL analyzer, Lunit SCOPE IO

September 22, 2023 — Lunit, a leading provider of AI-powered solutions for cancer diagnostics and therapeutics, today announced a pioneering study result, recently published in npj Breast Cancer (Nature Partner Journal Breast Cancer). The study presents significant advancements in breast cancer treatment planning utilizing Lunit's AI-powered tumor-infiltrating lymphocytes (TIL) analyzer, Lunit SCOPE IO.

Breast cancer treatment outcomes are heavily influenced by the evaluation of TIL within the tumor microenvironment. However, the challenge of variability in the unaided TIL assessment has hindered TIL's use as a biomarker. Lunit has tackled this challenge with Lunit SCOPE IO, by enabling TIL assessment and immune phenotyping pan-cancer for effective diagnosis and treatment planning.

In this study, in a comprehensive evaluation of 402 whole slide images of breast cancer, three pathologists initially assessed stromal TIL (sTIL) scores. Among 402 images, 210 cases (52.2%) showed sTIL score differences of less than 10 percentage points between pathologists. The standalone performance of Lunit SCOPE IO was notable, demonstrating a concordance correlation coefficient of 0.755 compared to the pathologists' scores.

For the 226 slides (56.2%) that exhibited a 10-percentage point or greater variance between pathologists and the deep learning-based TIL analyzer, revised assessments were made with the assistance of Lunit SCOPE IO. This process reduced the number of discordant cases to just 116 (28.9%), with a statistically significant difference. The Lunit SCOPE IO-assisted revision also significantly improved the concordance correlation coefficient of the sTIL score among every two pathologists.

The study's implications extend to treatment decisions for breast cancer patients. In cases where triple-negative and HER2-positive breast cancer patients underwent neoadjuvant chemotherapy, the Lunit SCOPE IO-assisted revised assessment revealed significantly higher sTIL scores in responders. Additionally, the AI-assisted revised assessment unveiled a correlation between sTIL-high tumors and positive chemotherapeutic responses.

This groundbreaking research highlights the potential of AI analysis as a reliable reference for sTIL scoring in breast cancer assessment. By enhancing inter-pathologist concordance in sTIL interpretation and predicting neoadjuvant chemotherapy response, Lunit SCOPE IO offers a path towards improved breast cancer treatment pathways.

“Breast cancer treatment decisions have been challenged by the inconsistent assessment of tumor-infiltrating lymphocytes. Lunit SCOPE IO addresses this issue by not only providing consistent and accurate assessments of TIL, but also predicting treatment outcomes," said Brandon Suh, CEO of Lunit. “This thoughtful use of Lunit AI has the potential to improve how we analyze and treat breast cancer. By paving the way for more effective and personalized treatment strategies, we are bringing a new era to patients and adding a powerful tool for oncologists.”

For more information: www.lunit.com