Identification of Novel Biomarkers for Pancreatic Cancer Using Integrated Transcriptomics With Functional Pathways Analysis

Pancreatic cancer is a leading cause of cancer death largely, but the genetic alterations associated with the initiation and progression of this disease are still not well understood. To comprehensively investigate potential utility of miRNAs and protein encoding transcripts (mRNAs) in pancreatic cancer as biomarkers of the disease, we exhaustively mined genomic data from two publically available datasets: the National Center for Biotechnology Information gene expression omnibus (GEO) and the Oncomine databases. We identified 26 miRNAs that were differentially expressed in pancreatic cancer tissue from five microarray datasets. Using data from five pancreatic cancer studies, we found 260 deregulated mRNAs associated with pancreatic cancer. Among these 260 deregulated transcripts, there were six transcripts encode proteins (COL1A2, CEACAM5, LAMA3, CP, ENO2, and FN1) secreted in blood, which may be developed as blood-based biomarkers of pancreatic cancer. Further, we found that 13 abnormally expressing transcripts targeted by deregulated miRNAs were involved in the epithelial adherens junction signaling, Wnt/β-catenin signaling, TR/RXR activation, hepatic fibrosis/hepatic stellate cell activation, and actin cytoskeleton canonical signaling pathways. Integrated bioinformatics analyses of multiple independent transcriptomic datasets revealed tumor associated deregulated miRNAs and protein encoding mRNAs involved in critical signaling pathways, which warrant further investigations to be validated as sensitive and specific biomarkers of pancreatic cancer detection and prognosis.