Pancreatic ductal adenocarcinoma (PDAC) carries a poor prognosis largely due to lack of efficient diagnostic means. We applied mass spectrometry-based high-coverage plasma proteome analysis accompanying with machine learning to develop a 5-protein diagnostic model: SNCA, GCLC, LBP, ALAD, and SORD. For differentiating PDAC from healthy controls (HCs), this model reached an area under the curve (AUC) of 0.973 with 100% sensitivity and 85% specificity in the discovery cohort, with nested cross-validation confirming robust performance (AUC = 0.958). Further validation centered on SNCA achieved an AUC of 0.835 in an independent validation cohort. SNCA also showed good diagnostic performance in PDAC patients with low CA19-9 level (AUC = 0.868), underscoring its potential value for this subgroup. Overall, these findings indicate SNCA as a promising candidate plasma diagnostic marker for PDAC.
Yu et al. (Mon,) studied this question.