Lunit Unveils New AI-Powered Tumor Microenvironment Insights in Lung Cancer at WCLC 2026

Three studies explore the relationship between tumor microenvironment, genomic alterations, and treatment response in non-small cell lung cancer using AI-based analysis

SEOUL, South Korea, Sept. 8, 2026 /PRNewswire/ -- Lunit (KRX:328130), a leading provider of AI for cancer diagnostics and precision oncology, today announced the presentation of three studies at the 2026 World Conference on Lung Cancer (WCLC 2026), to be held in Seoul, South Korea, from September 12 to 15.

Across the three studies, Lunit researchers and collaborators leveraged AI-based analysis to investigate the tumor microenvironment (TME) in non-small cell lung cancer (NSCLC), exploring its relationship with genomic characteristics and treatment response.

In the first study, researchers used Lunit SCOPE IO to analyze 494 hematoxylin and eosin (H&E)-stained whole-slide images from EGFR-mutant NSCLC cases, identifying distinct TME characteristics across mutation subtypes. Exon 19 deletion (Ex19del) tumors showed significantly lower intratumoral tumor-infiltrating lymphocyte (TIL) density, while L858R tumors showed relative enrichment of both TIL and macrophage infiltration, and exon 20 insertion (Ex20ins) tumors demonstrated higher endothelial cell density.

The findings suggest that differences in treatment outcomes across EGFR mutation subtypes may involve not only kinase kinetics but also subtype-specific characteristics of the TME.

In another study, conducted in collaboration with Paola Nisticò, M.D., of the Regina Elena National Cancer Institute in Rome, Italy researchers combined Lunit SCOPE IO with spatial transcriptomics and high-plex spatial proteomics to analyze 32 NSCLC patients treated with neoadjuvant chemo-immunotherapy. The study was conducted through the Lunit Research Program for SITC Members, which provides eligible Society for Immunotherapy of Cancer (SITC) members with research access to Lunit SCOPE IO for AI-powered tumor microenvironment analysis in cancer immunotherapy research. Tumors from patients who achieved pathological complete response (pCR) showed a highly inflamed and spatially organized microenvironment with prominent tertiary lymphoid structures (TLS), while non-pCR tumors were characterized by immune exclusion and activated fibroblast-rich stroma.

The findings highlight distinct spatial and immunological features associated with pathological response and support further investigation of spatially resolved biomarkers.

The third study evaluated an H&E-based AI model for predicting TP53 mutation status in lung adenocarcinoma. Validated in an independent cohort of 462 cases, the model achieved an area under the receiver operating characteristic curve (AUROC) of 0.759, with 82% sensitivity and 63% specificity. TME analysis further showed that TP53-mutant tumors were more frequently immune-inflamed, while TP53-wild-type tumors were characterized by immune exclusion and higher stromal fibroblast and endothelial cell densities.

The findings highlight the potential of H&E-based AI as a screening tool to enrich for patients with TP53 mutations who may be eligible for emerging clinical trials targeting specific TP53 variants.

"These studies demonstrate the expanding potential of AI-based analysis to generate deeper insights into lung cancer biology, from genomic characteristics and the tumor microenvironment to features associated with treatment response," said Brandon Suh, CEO of Lunit. "By broadening the range of insights that can be derived from routinely available pathology images, we aim to advance AI-powered biomarker discovery and patient stratification, ultimately contributing to more personalized treatment strategies for patients with cancer."

Lunit's presentations at WCLC 2026 include:

• [Poster #P3.156] AI-Powered Tumor Microenvironment Analysis Across EGFR-Mutation Subtypes in Non-Small Cell Lung Cancer, September 15, 9:30 AM KST, Exhibits and Posters, Hall C, 3F
• [Poster #P2.126] Spatial Multi-Omics and AI-Driven Digital Pathology to Identify Determinants of Pathological Response to Neoadjuvant Chemo-IO in NSCLC, September 14, 10:30 AM KST, Exhibits and Posters, Hall C, 3F
• [Poster #P2.134] Development of an H&E-Based Genotype Predictor and Tumor Microenvironment Analysis of TP53-Mutated Lung Adenocarcinoma, September 14, 10:30 AM KST, Exhibits and Posters, Hall C, 3F

###

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 from risk prediction and early detection to precision oncology. Our FDA-cleared Lunit INSIGHT Breast 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.

SOURCE Lunit