Background Pancreatic ductal adenocarcinoma (PDAC) is the most frequently diagnosed form of pancreatic cancer worldwide. It is associated with poor survival rates (~ 5%) mainly due to the disease being usually diagnosed at late stages. Few gene expression studies have been conducted on samples from PDAC patients, however their sample size was limited and their final outcome inconclusive. We aimed to identify general PDAC disease biomarkers that may improve earlier diagnosis and patient stratification for improved mortality outcomes. Methods Publicly available gene expression data from 10 studies with tumour tissue (448 samples) blood samples (128 samples) from PDAC patients prior to treatment were analysed. Validation of markers was performed using Cancer Genome Atlas (TCGA) PDAC expression data. Tissue samples had AJCC (American Joint Committee for Cancer) staging information available. Differential gene expression analysis was carried out to compare tumour and normal samples (stage-specific tissue samples vs. normal tissue samples and PDAC blood samples vs. normal blood samples). Active subnetwork search and miRNA enrichment analysis were used to identify enriched gene networks and miRNA interactions. Results We identified 820 consistently deregulated (either up- or down-regulated) genes between tissue samples of all stages and blood samples. The prognostic potential of these markers was validated in TCGA data in predicting PDAC outcome (dead/alive status), in the form of custom risk scores (up-regulated genes score: p = 0.004 and down-regulated genes score: p = 0.03). Active subnetwork analysis revealed enriched ribosome, proteasome, adherens junction and cell cycle pathways in tumors across all stages and blood samples. Stage-specific enriched miRNAs were also identified (miR-21, miR-29, miR-124, miR-30, for stages 1–4 respectively). Conclusions We identified PDAC markers deregulated across all stages and different sample sets. Extensive gene expression deregulation was found in all clinical stages with significant overlap. Additionally, miRNA contribution to PDAC pathology may be important and probably mediated by distinct miRNAs in each stage of PDAC. We therefore present a list of markers and miRNAs that could potentially act as a diagnostic tool for early detection of PDAC onset to be evaluated in other clinical and epidemiologic studies.

Pancreatic ductal adenocarcinoma (PDAC) is the most frequently diagnosed form of pancreatic cancer worldwide. PDAC is associated with poor survival rate mainly due to the disease being usually diagnosed at late stages. Publicly available gene expression data from 10 studies with tumour tissue (448 samples) and/or blood samples (128 samples) from PDAC patients were pooled together and analysed for the identification of stage-specific (American Joint Committee for Cancer, AJCC staging) and global diagnostic markers. Validation of markers was performed using Cancer Genome Atlas (TCGA) PDAC expression data. Differential gene expression analysis was carried out to compare tumour and normal samples (stage-specific tissue samples vs. normal tissue samples and stage-agnostic blood samples vs. normal blood samples). Active subnetwork search and miRNA enrichment analysis were used to identify enriched gene networks and miRNA interactions. We identified 820 consistently deregulated genes in tissue samples of all stages and blood samples. Active subnetwork analysis revealed enriched ribosome, proteasome, adherens junction and cell cycle pathways across all stages and blood samples suggesting biological plausibility. Stage-specific enriched miRNAs with diagnostic potential were also identified (miR-21, miR-29, miR-124, miR-30, for stages 1-4 respectively). Extensive gene expression deregulation was found in all tumor stages with significant overlap. Additionally, miRNA contribution to PDAC pathology may be important and probably mediated by distinct miRNAs in each stage of PDAC. We therefore present a list of markers and miRNAs that could potentially act as a diagnostic tool for early detection of PDAC onset.