Further, if circulating tumor cells from the pancreas were the cause of the differential expression profile seen in this study, it would be expected that genes normally expressed in AZ 960 pancreatic cells but not peripheral blood cells would be identified by microarray. While this was the case for USH1C, Nutlin-3 CRISP3, and USP30, all genes that are expressed at low to moderate levels in the pancreas but not expressed normally in the peripheral blood, PLEKHA1, a gene that is normally highly expressed in the pancreas but only expressed at very low levels in the peripheral blood, was shown to be down regulated in our samples, adding to the evidence that the differential expression we report truly is from PBMCs. In conclusion, we have shown that a differential gene expression profile exists in PBMCs of patients with PC. Further, an 8 gene classifier set has been established which provides, in a blinded subset of our samples, an improved sensitivity over CA19-9 with a similar specificity. Significantly, there was no decrease in sensitivity when employing samples from patients prior to any form of chemotherapy or surgery. Comparison with other studies points toward this differential expression profile as being specific to PBMCs and particularly to PC. Additionally, the differential gene expression seems to represent a systemic compromise of both cellular and humoral immunity, although it does not point toward one particular underlying mechanism. Based on these results, future research is needed to establish the various mechanisms behind PC-induced differential PBMC genetic expression and how much effect this differential expression actually has on the body��s immunologic capabilities. Further, the 8 gene classifier set must be tested in an expanded set of both healthy controls and PC patients as well as in a set of non-PC patients with benign/malignant disease to clarify its sensitivity and specificity. PC sample selection for such a study should be biased toward early stage patients to elicit the diagnostic capabilities of PBMC differential expression in the patient population in which it has the greatest likelihood of having a positive impact on patient outcome. Due to the difficulty of attaining ample specimens from early stage patients, preliminary study of early stage PC diagnosis through PBMC expression analysis may be first carried out in a spontaneous PC murine model which recapitulates the preneoplastic and early neoplastic processes seen in human PC as a proof of concept. Upon conclusion of such a murine study, resources could then be expended in the recruitment and testing of enough early stage human PC subjects for ample analysis to be conducted. Once PBMC expression has been diagnostically validated in an expanded sample set, the gene set could also be used to characterize potential prognostic abilities of the PBMC differential expression in PC.