On February 16, 2024 Lunit (KRX:328130.KQ), a leading provider of AI-powered solutions for cancer diagnostics and therapeutics, reported a groundbreaking study published in the Journal for ImmunoTherapy of Cancer (JITC) (Press release, Lunit, FEB 16, 2024, View Source;published-in-the-jitc-302064098.html [SID1234640191]). The study demonstrates the ability of Lunit SCOPE IO, to enable quantitative immune phenotyping from H&E stained slides as a broadly accessible biomarker for immunotherapy.
Schedule your 30 min Free 1stOncology Demo!
Discover why more than 1,500 members use 1stOncology™ to excel in:
Early/Late Stage Pipeline Development - Target Scouting - Clinical Biomarkers - Indication Selection & Expansion - BD&L Contacts - Conference Reports - Combinatorial Drug Settings - Companion Diagnostics - Drug Repositioning - First-in-class Analysis - Competitive Analysis - Deals & Licensing
Schedule Your 30 min Free Demo!
The study, conducted on a real-world multicenter cohort of 1,806 Immune Checkpoint Inhibitor (ICI)-treated patients across 27 tumor types, showcases a correlation between the Inflamed Immune Phenotype and positive ICI treatment responses. There is an unmet need for improved immunotherapy biomarkers, and this study highlights the importance of Lunit SCOPE IO’s ability to quantify immune phenotype (IP) as Inflamed, Excluded, or Desert, purely from H&E whole slide images (WSIs).
Utilizing advanced machine learning (ML) models, Lunit SCOPE IO segments tissue into cancer area (CA) and cancer stroma (CS) within WSIs. The model also detects Tumor-Infiltrating Lymphocytes (TILs) using a cell detection model trained on over 17,000 WSIs spanning multiple solid tumor types.
Based on TIL density, the model classifies the tumor into one of three immune phenotypes: Inflamed (IIP; high TIL density within CA), Immune Excluded (IEP; TILs within CS but excluded from CA), and Immune Desert (IDP; low TIL density within both CA and CS).
In an independent real-world dataset of ICI-treated patients, Lunit SCOPE IO demonstrated predictive power for clinical outcomes, including objective response rates (ORR), progression-free survival (PFS), and overall survival (OS). In the study, IIP patients showed significantly favorable clinical outcomes post-ICI treatment. More favorable ORRs (26.3% vs 15.8%), prolonged PFS (5.3 vs. 3.1 months) and OS (25.3 vs. 13.6 months) were observed in IIP patients, irrespective of ICI regimen or programmed death-ligand 1 (PD-L1) expression. The dataset reflected global diversity, with data coming from Stanford University, Samsung Medical Center, Seoul National University Bundang Hospital, Chonnam National University Hospital, and Northwestern Memorial Hospital, and more.
This study paves the way for more precise patient selection with a time-efficient and labor-efficient analysis at scale in immunotherapy. Lunit plans to further validate and deploy Lunit SCOPE IO, ultimately enabling more personalized and effective immunotherapy strategies, especially under the current limitations of traditional biomarkers.
"This study marks a major step towards better biomarkers for immunotherapy driven by AI, analyzing the tumor microenvironment to determine immune phenotype quantitatively and predict patient responses to ICI therapy," said Brandon Suh CEO of Lunit. "Our commitment to advancing cancer care through innovation has never been clearer. By providing a robust tool for personalized treatment strategies, Lunit SCOPE IO promises improved outcomes and could redefine the standard of care for patients in several cancer types where predictive biomarkers are lacking."
Published in the JITC, the official journal of the Society for Immunotherapy of Cancer (SITC) (Free SITC Whitepaper), the study also contributes to SITC (Free SITC Whitepaper)’s mission of enhancing cancer patient outcomes by advancing the science and application of cancer immunotherapy. SITC (Free SITC Whitepaper) is the world’s leading member-driven organization that includes over 4,650 members from 35 medical specialties across 63 countries worldwide.