'Power of AI' Lunit Presents 9 Research Findings at ESMO

Medical AI company Lunit announced on the 16th that it will present nine research results demonstrating the predictive validity of cancer treatment effects using the AI biomarker 'Lunit Scope' at the '2023 European Society for Medical Oncology (ESMO 2023)' held in Madrid, Spain from the 20th to the 24th. Established in 1975, ESMO is a global academic organization with over 30,000 members, including cancer researchers, oncologists, and representatives from global pharmaceutical and biotech companies from about 170 countries worldwide, serving as a key organization for cancer research and treatment improvement.

Lunit's 'Lunit Scope IO'
Photo by Lunit

At this conference, Lunit announced that it applied AI biomarkers to various cancer types such as advanced biliary tract cancer (BTC), head and neck squamous cell carcinoma (HNSCC), and non-small cell lung cancer (NSCLC), enhancing the prediction of treatment responses and reaffirming clinical value.

One of the major studies revealed by Lunit analyzed cancer tissue images of 337 patients with advanced biliary tract cancer using AI to predict responses to anti-PD-1 therapy. Using Lunit Scope, the distribution of tumor-infiltrating lymphocytes (TILs), which are immune cells, was classified into three types: immune active, immune excluded, and immune deficient. The immune active group showed better outcomes with an overall survival (OS) of 12.5 months, progression-free survival (PFS) of 5.0 months, and objective response rate (ORR) of 27.5%, compared to the immune deficient group’s 5.1 months, 2.0 months, and 7.7%, respectively.

In the study on advanced and metastatic head and neck squamous cell carcinoma, it was confirmed that the density of TILs in the tumor microenvironment is highly correlated with the treatment response to immune checkpoint inhibitors (ICI). Measuring TIL density using Lunit Scope showed that patients with high TIL density had an ORR of 21.6% and a median PFS (mPFS) of 3.2 months, which was improved compared to 5.7% and 1.6 months in patients with lower density. This suggests that TIL measurement using Lunit Scope could be a predictive tool for immunotherapy response in head and neck squamous cell carcinoma patients.

Lunit will also present results from three clinical trials where Lunit Scope was applied as an exploratory biomarker. In a clinical study conducted jointly with the Mayo Clinic, one of the top cancer research hospitals in the U.S., it was revealed that AI-measured TIL density within tumor epithelial tissue plays a significant role in distinguishing between mismatch repair-deficient (dMMR) and normal tissues, which is crucial for treatment decisions in colorectal cancer patients. Clinical studies led by researchers from Italy and France will also present analyses related to metastatic colorectal cancer using Lunit Scope. Another study showed that measuring human epidermal growth factor receptor 2 (HER2) levels in advanced biliary tract cancer using Lunit Scope yielded results highly consistent with those measured by pathologists.

Finally, research results showed that applying Lunit Scope to select non-small cell lung cancer patients without mutations (All-WT) achieved a probability of 95.2%, which is below the typical error margin of 5%, potentially reducing unnecessary additional tests. Additionally, it was reported that patients tested with Lunit Scope had a threefold higher likelihood of detecting the 'MET exon deletion mutation,' a lung cancer oncogene, compared to the general patient group.

Seobum Seok, CEO of Lunit, said, “Lunit Scope is a biomarker platform that can significantly improve the effectiveness of immunotherapy across various cancer types, aiming to enhance patient survival rates and quality of life. Through this conference, we intend to further solidify our leading position in AI-driven cancer research and strive to provide practical benefits to more patients.”

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