High-performance Collective Biomarker from Liquid Biopsy for Diagnosis of Pancreatic Cancer Based on Mass Spectrometry and Machine Learning

This study developed a highly accurate diagnostic system for pancreatic ductal adenocarcinoma (PDAC) using a collective biomarker approach based on liquid chromatography/electrospray ionization mass spectrometry (LC/ESI-MS) and machine learning. The researchers analyzed serum samples from Japanese patients, focusing on primary metabolites (PM) and phospholipids (PL). They found that integrating both PM and PL databases significantly improved diagnostic accuracy compared to using either database alone or single metabolites. A refined algorithm incorporating 36 statistically significant metabolites achieved a diagnostic accuracy of 97.4%. The study also explored the potential of this system to monitor the effects of neoadjuvant chemotherapy (NAC), observing distinct metabolic patterns in NAC-treated patients. This method offers a promising, non-invasive approach for early PDAC detection and for evaluating treatment efficacy, addressing the critical need for improved diagnostic and monitoring tools in pancreatic cancer management. Research Background and Significance Pancreatic ductal adenocarcinoma (PDAC) remains one of the most lethal malignancies worldwide, largely due to its asymptomatic early stages and the lack of reliable, non-invasive diagnostic methods. Early detection is crucial for improving patient prognosis, yet conventional biomarkers such as CA19-9 lack sufficient sensitivity and specificity. Liquid biopsy, which enables minimally invasive assessment of circulating biomolecules, has emerged as a promising avenue for cancer diagnostics. In this context, metabolomics combined with advanced analytical techniques like liquid chromatography coupled with mass spectrometry (LC/MS) provides a powerful platform to uncover disease-specific metabolic signatures. This s