Announcement • May 24
Lunit Presents Ai-Powered Ihc Biomarker Quantification and Spatial Tumor Microenvironment Analysis At Asco 2026 Lunit announced that five studies featuring its AI biomarker platforms are being presented at the 2026 American Society of Clinical Oncology (ASCO) Annual Meeting, taking place May 29–June 2 in Chicago, IL. This year's presentations reflect Lunit's expanding biomarker research beyond conventional immune profiling toward integrated AI-powered HER2/IHC biomarker quantification and spatial tumor microenvironment (TME) analysis. Using AI to characterize HER2 expression, immune phenotypes, tertiary lymphoid structures (TLS), tumor-infiltrating lymphocytes (TILs), and endothelial cells, the studies highlight the potential of integrated biomarker analysis to improve precision patient stratification across multiple cancer types. The five presentations include studies across biliary tract cancer (BTC), non-small cell lung cancer (NSCLC), adenoid cystic carcinoma (ACC), microsatellite-stable (MSS), metastatic colorectal cancer and advanced gastric cancer (AGC). One of the featured studies, led by researchers at Yonsei University College of Medicine, evaluates a first-line quadruplet regimen consisting of trastuzumab, nivolumab, gemcitabine, and cisplatin in HER2-positive advanced BTC. Using AI-powered whole-slide image (WSI) analysis, researchers analyze HER2 expression and immune phenotypes within tumor tissue. Among 40 patients enrolled in the multi-center phase Ib/II trial, the combination of therapy demonstrates an objective response rate (ORR) of 55%, disease control rate (DCR) of 95%, and median progression-free survival (PFS) of 10.6 months. Patients with HER2 IHC 3+ tumor cell proportions =10%, as identified by AI analysis, achieve substantially higher response rates compared to those below the threshold (80% vs. 36.4%). The study demonstrates how integrated AI-powered HER2 and spatial tumor microenvironment analysis may help identify patients more likely to benefit from HER2-targeted combination therapies. Another study characterizes the tumor microenvironment landscape of HER2-overexpressing NSCLC using AI-powered spatial analysis. Across more than 2,000 NSCLC whole-slide images, HER2-overexpressing tumors demonstrate significantly reduced tumor-infiltrating lymphocyte density and lower proportions of inflamed immune phenotype compared to non-overexpressing tumors. The immune-cold phenotype becomes more pronounced in tumors with higher proportions of HER2 3+ cells. The study provides additional insight into the biological relationship between HER2 overexpression and tumor immune status, while highlighting the broader potential of combining AI-powered HER2 analysis with spatial tumor microenvironment characterization in biomarker research. Another featured study, conducted with Seoul National University Hospital, explores AI-powered spatial tumor microenvironment analysis in adenoid cystic carcinoma (ACC), a rare cancer with limited treatment options. Researchers identify a subgroup of ACC patients with high endothelial cell and tumor-infiltrating lymphocyte (TIL) densities who demonstrate significantly prolonged progression-free survival following axitinib treatment (19.6 months vs. 11.1 months). The findings suggest that AI-based spatial profiling may help distinguish responder populations that are difficult to stratify using conventional biomarkers alone. Researchers at Asan Medical Center investigate AI-powered tumor microenvironment analysis in MSS metastatic colorectal cancer, which is generally considered resistant to immune checkpoint inhibitors (ICIs). Researchers find that patients with larger tertiary lymphoid structure (TLS) regions identified by AI demonstrate improved progression-free survival and overall survival following immunotherapy. The findings highlight the potential of AI-powered spatial analysis to identify MSS colorectal cancer patients more likely to benefit from immunotherapy. New Risk • May 15
New minor risk - Shareholder dilution The company's shareholders have been diluted in the past year. Increase in shares outstanding: 28% This is considered a minor risk. Shareholder dilution occurs when there is an increase in the number of shares on issue that is not proportionally distributed between all shareholders. Often due to the company raising equity capital or some options being converted into stock. All else being equal, if there are more shares outstanding then each existing share will be entitled to a lower proportion of the company's total earnings, thus reducing earnings per share (EPS). While dilution might not always result in lower EPS (like if the company is using the capital to fund an EPS accretive acquisition) in a lot cases it does, along with lower dividends per share and less voting power at shareholder meetings. Currently, the following risks have been identified for the company: Major Risk Earnings are forecast to decline by an average of 38% per year for the foreseeable future. Minor Risks Currently unprofitable and not forecast to become profitable over next 2 years (₩121b net loss in 2 years). Shareholders have been diluted in the past year (28% increase in shares outstanding). Announcement • Apr 18
Lunit Presents Six AI Studies At AACR 2026 Highlighting Advances in Precision Oncology and Real-World Clinical Application Lunit's AACR 2026 presentations showed how AI-driven biomarkers can improve the efficiency of clinical workflows, uncover spatial features of the tumor microenvironment not captured by conventional methods, and enable integrated analysis to better support treatment decision-making. Lunit (KRX:328130), a provider of AI for cancer diagnostics and precision oncology, presented six studies at the American Association for Cancer Research (AACR) Annual Meeting 2026, taking place from April 17 to 22 in San Diego, California. The presentations highlighted Lunit's advancements in AI-driven biomarker development, tumor microenvironment (TME) analysis, and real-world clinical applicability. Several studies were conducted in collaboration with global partners, including Agilent Technologies. In a study conducted with Agilent Technologies and Ajou University Medical Center, researchers used Lunit SCOPE IO and uIHC to analyze over 25,000 non-small cell lung cancer (NSCLC) samples. The results showed that tumors with high c-MET expression exhibited a significant reduction in immune cell density within 30 µm of tumor cells (p<0.001), revealing a spatial immune exclusion pattern not captured by conventional analysis. These findings suggest a potential link between c-MET overexpression and immune evasion, supporting combination strategies involving MET-targeted therapy and immunotherapy. Researchers also present findings from an exploratory analysis of the phase II MOUNTAINEER trial, demonstrating that AI- quantified HER2 expression is strongly associated with treatment response in patients with HER2-positive metastatic colorectal cancer treated with tucatinib plus trastuzumab. The overall objective response rate (ORR) was 43.4%, increasing to as high as 80% in patients with higher HER2 expression, indicating a clearer dose-dependent relationship. Tumor-Infiltrating Lymphocyte (TIL) density independently predicted progression-free survival. Notably, patients with low stromal TIL levels showed no response (ORR 0%) and a significantly higher risk of disease progression. These findings highlight the increasing complexity of biomarker assessment, where both tumor characteristics and immune context need to be considered, underscoring the potential role of AI in supporting treatment decision-making. In addition to these representative studies, Lunit presented additional research abstracts at AACR 2026, further demonstrating the breadth of its AI-powered oncology research. These include studies on AI-based analysis of tumor-infiltrating lymphocyte in NSCLC in collaboration with Dr. David Rimm's lab at Yale University School of Medicine, AI-based target discovery for bi-specific antibodies, Biomarker research in CD47-targeted therapies. 