Lunit Presents AI-Based Prediction of Immunotherapy Response in Rare Cancers at AACR in the US

Medical artificial intelligence (AI) company Lunit announced on April 21 that it will present seven of its latest research achievements utilizing its AI biomarker platform, Lunit Scope, at the 2025 American Association for Cancer Research (AACR 2025) to be held in Chicago, USA, from April 25 to 30 (local time).

The seven research presentations will include previously announced studies by Lunit, such as the validation of the immunotherapy drug Tecentriq in collaboration with global pharmaceutical company Genentech, as well as an AI-based EGFR mutation prediction study conducted with AstraZeneca.

In addition, one of the main studies Lunit will present at this year’s conference focuses on predicting the efficacy of neoadjuvant immunochemotherapy in patients with the rare cancer salivary gland carcinoma (SGC). The study integrated spatial transcriptomics analysis using Xenium by 10X Genomics, AI analysis based on Lunit Scope IO, and sequencing analysis, and was conducted on 14 patients with salivary gland carcinoma.

The research team analyzed over 910,000 cells within salivary gland tumors and found that, in the group of patients who experienced recurrence, genes associated with immune evasion and cancer cell metastasis were highly expressed. At the same time, the expression of CXCL9?a gene that induces tumor-infiltrating lymphocyte (TIL) density and immune cell infiltration?was found to be low.

By combining spatial transcriptomics analysis with Lunit Scope IO, this study revealed the characteristics of the tumor microenvironment (TME) and is expected to provide important scientific evidence for establishing immunotherapy strategies for the rare cancer salivary gland carcinoma.

Additionally, Lunit will present AI-based analysis results predicting the risk of small cell lung cancer (SCLC) transformation in non-small cell lung cancer (NSCLC) patients with EGFR mutations. The research team analyzed tissue slides from 106 patients using AI, classifying tumor heterogeneity at the cellular level and identifying patient groups with morphological features similar to small cell lung cancer.

The analysis showed that patients in the group with SCLC-like characteristics had a significantly smaller average nuclear size compared to those not in the SCLC-like group. These patients also had a notably shorter progression-free survival (PFS) after TKI therapy, and a higher rate of transformation to small cell lung cancer upon further testing.

This study is the first in the world to demonstrate that AI can identify patients at risk of SCLC transformation, suggesting that it can make a significant contribution to establishing more precise treatment strategies and predicting prognosis.

Seo Bumseok, CEO of Lunit, stated, "The studies we are presenting at AACR 2025 demonstrate that Lunit Scope can provide meaningful insights for treatment even in rare and poor-prognosis cancers. Going forward, Lunit will continue to set new standards for personalized cancer treatment through AI-based analysis."

Meanwhile, Lunit is participating in AACR?one of the world's top three cancer conferences alongside the American Society of Clinical Oncology (ASCO) and the European Society for Medical Oncology (ESMO)?for the seventh consecutive year, consistently demonstrating the clinical efficacy and expanding research applications of Lunit Scope.