Lunit Presents Research Findings on 'c-MET Mutation' in NSCLC at AACR 2026

Medical artificial intelligence (AI) company Lunit announced on April 20 that it presented six research abstracts utilizing its AI biomarker platform, Lunit SCOPE, at the 2026 American Association for Cancer Research (AACR 2026), held in San Diego, United States, from April 17 to 22, 2026.

The main study unveiled by Lunit at this conference focuses on the association between c-MET (hepatocyte growth factor receptor) expression and the tumor microenvironment in non-small cell lung cancer (NSCLC). This research was conducted in collaboration with global diagnostics and analytics company Agilent Technologies and Ajou University School of Medicine, among others.

c-MET is a protein that promotes cancer growth and metastasis. Recently, antibody-drug conjugates (ADCs) targeting c-MET have gained attention following regulatory approval. However, the relationship between c-MET expression, the tumor microenvironment, and response to immunotherapy has not yet been fully elucidated.

The research team analyzed 25,674 NSCLC tissue samples using Lunit SCOPE IO and Lunit SCOPE uIHC to measure the distribution of immune cells around cancer cells according to c-MET expression levels.

The analysis revealed that tumors with high c-MET expression—particularly those with greater expression on the cell membrane relative to the cytoplasm—had a significantly lower density of tumor-infiltrating lymphocytes (TILs) in their vicinity. This suggests a link between high c-MET expression and immune evasion. The researchers explained that improving the immune-evasive microenvironment through MET-targeted therapy, followed by combination treatment with immune checkpoint inhibitors, could enhance therapeutic efficacy.

Lunit also released results from a Phase 2 clinical trial targeting HER2-positive metastatic colorectal cancer. The research team used AI to analyze tumor samples from 30 patients who received a combination therapy of Tukysa (ingredient: tucatinib) and Herceptin (ingredient: trastuzumab), measuring the proportion of HER2-high-expressing tumor cells and the density of surrounding immune cells, and compared these metrics with actual treatment outcomes.

The objective response rate (ORR) among all patients was 43.4%. However, response rates increased stepwise with higher proportions of HER2-high-expressing cells. The group with a HER2-high-expressing cell proportion of 50% or more had an 83% lower risk of disease progression compared to the group with less than 50%.

Conversely, in the patient group where the density of stromal tumor-infiltrating lymphocytes (sTILs) was in the lowest 25%, the results differed. Even among patients with high HER2 expression, there were no responders (ORR 0.0%), and the risk of disease progression was 4.4 times higher. The research team concluded that, for accurate prediction of treatment response, data on immune cells surrounding the tumor should be considered along with HER2 expression levels.

Beomseok Seo, CEO of Lunit, stated, "Through this research, we have verified the practical applicability of AI biomarkers in greater detail," adding, "We will accelerate commercialization so that Lunit SCOPE becomes an essential tool in global clinical settings by continuing to expand collaborations with international medical institutions and companies."

This content was produced with the assistance of AI translation services.