Bertis Presents Blood Test Model for Pancreatic Cancer Diagnosis with 92% Accuracy Based on 13 Biomarkers

-   Presenting an exploratory trial result for early diagnosis of PDAC using LC-MS technology effective for disease screening

-   Conducting a confirmatory trial for pancreatic cancer’s early diagnosis blood test ‘Panccheck,’ having secured blood samples from 513 individuals, including 220 pancreatic cancer patients

On September 24th, Bertis, a company specializing in the development of proteomics-based precision medicine technology (led by CEOs Dong-young Noh and Seung-man Han), announced that the findings on protein biomarkers for the diagnosis of Pancreatic Ductal Adenocarcinoma (PDAC) were published in the September issue of the prestigious international journal ‘Molecular & Cellular Proteomics.’

Pancreatic cancer is highly lethal due to its asymptomatic nature in the early stages, making early detection extremely challenging. Most patients are diagnosed at advanced stages, resulting in very low survival rates. According to data released by the Korea Central Cancer Registry in 2024, the 5-year cancer relative survival rate for pancreatic cancer from 2017 to 2021 is only 15.9%, making it the most difficult to treat among the top 10 cancers. In particular, PDAC, which accounts for more than 90% of pancreatic cancer, is considered one of the most lethal cancer types with a 5-year average survival rate of less than 10%.

Currently, the blood biomarker Carbohydrate Antigen 19-9 (CA19-9) is used for pancreatic cancer screening, but its usefulness is considered limited due to its low positive predictive value. This study focused on developing an effective diagnostic method for PDAC screening based on Liquid Chromatography-Mass Spectrometry (LC-MS) technology, which allows for the simultaneous and precise quantification of multiple protein biomarkers in blood and is effective for disease screening tests that require economical and rapid analysis of a large number of samples.

In this study, the Bertis research team selected 12 protein biomarkers that were differentially expressed in PDAC patients compared to a control group of healthy individuals and/or those with benign pancreatic diseases, using mass spectrometry and bioinformatics techniques. They then developed a Multi-Marker Panel consisting of 13 biomarkers combined with CA19-9. In the next step, they developed a diagnostic test model with improved performance and robustness using machine-learning technology and evaluated its diagnostic performance for PDAC screening. As a result, the model demonstrated an accuracy of 92% for all stages of PDAC and stage 1 to 2 PDAC, showing improved diagnostic performance in all cases compared to when CA19-9 was used alone.

Un-beom Kang, Head of the Bertis Biomarker Research Institute and co-leader of the research, stated, “The results of this study represent a significant achievement in providing a useful diagnostic model for the early detection of pancreatic cancer, a particularly hard-to-treat cancer, based on Bertis’ advanced technologies in proteomics, bioinformatics, and machine learning. We are committed to follow-up research and development to ensure that healthcare professionals and patients can utilize this technology in clinical settings, thereby facilitating early diagnosis of pancreatic cancer.”

Bertis is promoting the commercialization of its blood test, Panccheck®, to aid in the early diagnosis of pancreatic cancer based on this research achievement. Currently, a confirmatory trial is underway, with blood samples secured from 513 individuals, including 220 pancreatic cancer patients, in preparation for approval from the Ministry of Food and Drug Safety. Additionally, Bertis is developing an early diagnosis test for ovarian cancer, which has the highest mortality rate among gynecological cancers and lacks effective early detection methods. The marker panel development is complete, and a clinical trial for algorithm enhancement are planned.

Seung-man Han, CEO of Bertis, emphasized, “Bertis is working to introduce multi-marker diagnostic solutions with improved accuracy based on more biomarkers for major diseases by adding the latest machine learning technology to the proteomics technology we have developed over the past decade. Building on our capability to commercialize Mastocheck, the world’s first proteomics-based blood testing for early breast cancer diagnosis, we are committed to providing effective early diagnosis solutions for hard-to-treat diseases like pancreatic cancer in clinical practice.”

Meanwhile, Bertis combines proteomics and bioinformatics to develop innovative biomarkers for cancer and other major diseases, as well as provides diagnosis and analysis solutions. It succeeded in commercializing proteomics technology with Mastocheck®, the blood testing solution for early breast cancer diagnosis, and PASS (Pan-omics Analysis Service & Solution), an omics analysis solution. Bertis continues to conduct research on early diagnosis solutions for major diseases.