Lunit's 'Lunit Scope PD-L1' Clinical Utility Evaluation Results Published in Latest Issue of EJC

[Asia Economy Reporter Chunhee Lee] Medical artificial intelligence (AI) company Lunit announced on the 20th that the research results evaluating the clinical utility of its AI pathology analysis solution 'Lunit Scope PD-L1' have been published in the latest issue of the international oncology journal 'European Journal of Cancer (EJC)'.

PD-L1 is a protein found on the surface of cancer cells, and the degree of its expression is known to affect the treatment response to immune checkpoint inhibitors. Therefore, oncologists use the PD-L1 expression level, interpreted by pathologists, as a biomarker to predict the effectiveness of immune checkpoint inhibitors. However, there have been issues such as discrepancies in opinions during the interpretation process and the significant time required for accurate readings.

Lunit explained that these problems can be solved through its AI-based pathology analysis solution, Lunit Scope PD-L1. By applying AI trained on a vast amount of tissue slides and cancer cells, it enables more accurate and objective analysis.

In this study, the PD-L1 expression levels of 479 non-small cell lung cancer (NSCLC) patients were interpreted by three pathologists both without AI assistance and with AI assistance. As a result, the concordance rate of PD-L1 interpretation among the three specialists was 81.4% without AI, whereas it significantly increased to 90.2% when using Lunit Scope PD-L1. In actual treatment response prediction for NSCLC patients receiving immune checkpoint inhibitors, AI-assisted prediction was found to be more accurate.

Lunit also verified differences in treatment response based on the PD-L1 positive expression score (TPS) cutoff of less than 1%, which is a criterion for predicting immune checkpoint inhibitor treatment response. Pathologists participating in the study reclassified some patients initially categorized as having no treatment effect (PD-L1 TPS less than 1%) into candidates for immune checkpoint inhibitor treatment (PD-L1 TPS 1% or higher) after using Lunit Scope PD-L1, compared to diagnoses made without AI assistance.

As a result, the difference in progression-free survival (PFS) between the two groups?PD-L1 TPS less than 1% and 1% or higher?became larger, and a significant difference between the groups was also observed in overall survival (OS) when using Lunit Scope PD-L1. This indicates that Lunit Scope PD-L1 can quantify PD-L1 expression and identify additional candidates who may respond to immune checkpoint inhibitor treatment.

Seobeom Seok, CEO of Lunit, said, "Through the publication of this EJC paper, we have demonstrated that Lunit Scope PD-L1 can positively influence specialists' decision-making in the actual interpretation process," adding, "The results confirm that Lunit Scope PD-L1 not only assists specialists in making more accurate interpretations but also contributes to improving treatment prediction for cancer patients." Currently, Lunit Scope PD-L1 has received European medical device certification (CE-IVDD) last month, and Lunit aims to launch it in Europe within the second half of this year.