Proteomic methods have been widely used in disease marker discovery research. The aim of this study was to discover potential biomarkers for pancreatic cancer (PCa) using surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF-MS). Crude serum samples from 132 patients with PCa and 67 healthy controls (HCs) were analyzed in duplicate using SELDI. Support vector machine (SVM) analysis of the spectra was used to generate a predictive algorithm based on proteins that were maximally differentially expressed between patients with PCa and the HCs in the training cohort. This algorithm was tested using leave-one-out cross-validation in the test cohort. From the 4 significant peaks in the training cohort, a classifier for separating patients with PCa from HCs was developed. The classifier was challenged with all samples achieving 96.67% sensitivity and 100% specificity in the training cohort and 93.1% sensitivity and 78.57% specificity in the test cohort. Additionally, the classifier correctly classified 12/12 stage Ia and 13/16 stage IIa PCa cases. The combination of the SELDI panel and CA19-9 was superior to CA19-9 alone in distinguishing individuals with PCa from the healthy subject group. These results suggest that high-throughput proteomic profiling has the capacity to provide new biomarkers for the early detection and diagnosis of PCa.