Transcriptional profiling of peripheral blood mononuclear cells in pancreatic cancer patients identifies novel genes with potential diagnostic utility

Abstract

Background: It is well known that many malignancies, including pancreatic cancer (PC), possess the ability to evade the immune system by indirectly downregulating the mononuclear cell machinery necessary to launch an effective immune response. This knowledge, in conjunction with the fact that the trancriptome of peripheral blood mononuclear cells has been shown to be altered in the context of many diseases, including renal cell carcinoma, lead us to study if any such alteration in gene expression exists in PC as it may have diagnostic utility.

Methods and findings: PBMC samples from 26 PC patients and 33 matched healthy controls were analyzed by whole genome cDNA microarray. Three hundred eighty-three genes were found to be significantly different between PC and healthy controls, with 65 having at least a 1.5 fold change in expression. Pathway analysis revealed that many of these genes fell into pathways responsible for hematopoietic differentiation, cytokine signaling, and natural killer (NK) cell and CD8+ T-cell cytotoxic response. Unsupervised hierarchical clustering analysis identified an eight-gene predictor set, consisting of SSBP2, Ube2b-rs1, CA5B, F5, TBC1D8, ANXA3, ARG1, and ADAMTS20, that could distinguish PC patients from healthy controls with an accuracy of 79% in a blinded subset of samples from treatment naïve patients, giving a sensitivity of 83% and a specificity of 75%.

Conclusions: In summary, we report the first in-depth comparison of global gene expression profiles of PBMCs between PC patients and healthy controls. We have also identified a gene predictor set that can potentially be developed further for use in diagnostic algorithms in PC. Future directions of this research should include analysis of PBMC expression profiles in patients with chronic pancreatitis as well as increasing the number of early-stage patients to assess the utility of PBMCs in the early diagnosis of PC.

Figure 1. Global gene expression analysis in PBMCs from PC patients and healthy controls.

Hierarchical cluster analysis of global gene expression profile by cDNA whole genome microarray comparing healthy control and PC samples using all genes found to be statistically differentially expressed between the two groups (FDR<0.10, n = 383 genes). In no instance were samples pooled. Red indicates genes whose expression is elevated relative to the universal human reference (used to normalize all arrays) and green indicates genes whose expression is decreased relative to the universal human reference.

Figure 2. Dendrogram of sample relatedness.

A dendrogram of sample relatedness from the cluster analysis shown in Figure 1 using the statistically significant differentially expressed genes. Samples clustered into main groups, aligning well with classification of PC or HC. PC PBMC samples are indicated by grey bars while healthy PBMC samples are denoted by yellow bars.

Figure 3. Potential effect of the differential genetic expression of PBMCs.

All genes shown were found to be down-regulated greater than 1.5 fold. The respective amount of differential expression per gene as well as the stated function can be found in Table 3. The differential expression of these genes indicate that there is a global decrease in cell number, activation, and effectiveness of the adaptive immune system in patients with PDAC that may have a significant effect on both their morbidity and mortality. Dashed lines indicate the association of cells while solid lines indicate the differentiation or proliferation of a particular cell type. Numbers represent individual points of interaction between the genes and the immune differentiation and response pathway: 1, Presentation of antigen to Th0 cells; 2, Differentiation of Th0 cells down the Th1 or Th2 pathway; 3, Immune cell proliferation; 4, Stimulation of cytotoxic T-cell activity by Th1 cells; 5, Stimulation of humoral immunity by Th2 cells; 6, Recognition and response to target cells by cytotoxic T lymphocytes (CTL); 7, Differentiation of naïve B-cells; 8, Lysis of target cells by CTLs. Letters represent individual cell populations: a, Cytotoxic T-lymphocytes (CTL); b, B-cells. Decrease in genes associated with points a and b may represent a decrease in their respective associated cell's population.