Lunit Brings AI to ASCO 2025 with 12 Precision Oncology Studies | Non-generative ai examples | Generative ai mckinsey | Generative ai in banking pdf | Turtles AI

Lunit, a South Korean company specializing in AI solutions for cancer diagnostics and treatment, will present 12 studies at ASCO 2025, highlighting the application of AI in precision medicine.

Key Points:

• Lunit will present 12 studies at ASCO 2025, focused on the use of AI in oncology.
• One study showed that AI could identify patients with biliary tract cancer who could benefit from anti-HER2 therapies.
• Another study showed that AI was effective in assessing PD-L1 expression in lung cancer.
• AI was also used to predict CLDN18.2 expression in gastric cancer, potentially guiding treatment decisions.

At ASCO 2025, Lunit will present 12 studies highlighting the application of AI in precision medicine oncology. One study, in collaboration with the National Cancer Center Hospital East in Japan, used Lunit’s AI analyzer to assess HER2 expression in patients with biliary tract cancer. Results showed strong agreement between AI and pathologist-assessed scores, suggesting that AI can identify patients who may benefit from HER2-targeted therapies, even among those traditionally classified as having low HER2 intensity.

Another study assessed the agreement between pathologist-assessed and AI-assessed PD-L1 expression in patients with lung cancer, using Lunit SCOPE PD-L1. The study included 847 patients with non-small cell lung cancer and 102 with small cell lung cancer. The overall agreement between the AI ​​model and three expert pathologists was 70%, with higher values ​​in key subgroups. AI identified additional patients with higher PD-L1 expression, suggesting the potential for AI to improve immunotherapy candidate selection.

A third study introduced an AI model to predict CLDN18.2 expression in gastric cancer, a therapeutic target for zolbetuximab. Using H&E slides, the model achieved AUROCs greater than 0.751, suggesting the potential for efficient pre-screening of CLDN18.2-positive patients. Analysis of immune phenotypes showed that patients with an “inflammatory” phenotype performed better when treated with immune checkpoint inhibitors plus chemotherapy.

In addition to these studies, Lunit will present 9 additional abstracts covering a wide range of topics, including AI-based subcellular profiling to assess the druggability of 74 membrane proteins in 34 cancer types and deep learning analysis of endothelial cells to understand how the tumor vascular environment influences response to immunotherapy.

Founded in 2013, Lunit is a leader in AI for cancer diagnostics and therapy. Its FDA-approved Lunit INSIGHT suite supports cancer screening at over 4,800 medical institutions in more than 55 countries. Lunit’s clinical studies have been published in top-tier journals and presented at major conferences such as ASCO and RSNA.

These studies highlight the growing role of AI in improving cancer diagnosis and treatment, opening up new opportunities for precision medicine.