Medical AI company Lunit will present three research abstracts using its AI biomarker platform Lunit SCOPE IO at the 2026 World Conference on Lung Cancer (WCLC 2026), which will be held at COEX in Seoul from Sept. 12-15, the company said Sept. 8.
It will be the first time in 19 years that WCLC has been held in South Korea since 2007. At the conference, which brings together lung cancer researchers and medical professionals from around the world, Lunit plans to showcase research findings on AI-powered precision medicine for lung cancer.
First, researchers analyzed 494 whole-slide images of non-small cell lung cancer (NSCLC) tumors with subtypes of EGFR (epidermal growth factor receptor) mutations, the most common genetic alterations found in NSCLC, using Lunit SCOPE IO.
The analysis found that the exon 19 deletion subtype had a high density of tumor-infiltrating lymphocytes, while the L858R subtype showed high macrophage infiltration and the exon 20 insertion subtype had the highest density of endothelial cells that make up blood vessels.
This indicates that the tumor microenvironment differs depending on the subtype, even among EGFR-mutated lung cancers. The tumor microenvironment is the environment surrounding cancer cells, consisting of immune cells, blood vessels, fibroblasts and other components. It affects not only cancer growth and metastasis but also responses to treatments such as immunotherapy.
The researchers said the study could provide clues for understanding differences in treatment outcomes by subtype while serving as a basis for designing more precise treatment strategies.
Next, Lunit worked with a research team led by Dr. Paola Nistico of the Regina Elena National Cancer Institute in Italy to analyze tissue samples from 32 patients with NSCLC who received neoadjuvant chemotherapy and immunotherapy using Lunit SCOPE IO and explore their association with pathologic complete response (pCR).
The analysis found that patients who achieved pCR showed active immune-cell infiltration and prominent tertiary lymphoid structures (TLS), which regulate immune responses within the tumor microenvironment. In contrast, patients who did not achieve pCR showed an immune-excluded pattern in which immune cells failed to sufficiently penetrate the tumor, along with prominent tumor cell proliferation.
The study is expected to provide clues for developing biomarkers that could help determine whether individual patients need postoperative adjuvant therapy by identifying tumor microenvironment characteristics associated with pCR through AI.
Finally, Lunit analyzed tissue samples from 462 patients with lung adenocarcinoma to predict whether TP53, a gene that plays a role in preventing the proliferation of damaged cells, was mutated.
When researchers applied Lunit SCOPE IO to analyze differences in the tumor microenvironment according to TP53 mutation status, cancers with mutations showed a higher proportion of immune activation, while cancers without mutations showed an immune-excluded pattern and significantly higher densities of fibroblasts and endothelial cells in the stroma.
Lunit CEO Seo Beom-seok said, “At this year’s World Conference on Lung Cancer, we confirmed that AI can help more easily identify clues linking genetic mutations, the tumor microenvironment and treatment response. We will continue using Lunit SCOPE to help develop more precise treatment strategies for cancer patients.”
