Identification of novel candidate biomarkers and efficacious small molecule drugs for pancreatic adenocarcinoma based on comprehensive bioinformatics technology

Abstract

Background: Pancreatic adenocarcinoma (PAAD) is considered as one of most serious solid tumors with high mortality as well as incidence. We performed this present study to screen out novel biomarkers and potential efficacious small drugs in PAAD using integrated bioinformatics analyses. Methods: Expression profiles derived from TCGA and GTEx datasets were integrated by following bioinformatics methods: DEGs analysis, WGCNA, KEGG and GO enrichment analyses. Kaplan-Meier method was conducted to assess the prognostic performance of genes, and the predictive value of identified genes was determined by the ROC analysis. CMap analysis was utilized to identify the related small molecule drugs in PAAD. Results: A total of 4777 DEGs containing 2536 up-regulated and 2241 down-regulated genes were identified. WGCNA identified six modules that were related with clinical characteristics, and functional enrichment analyses showed that all genes of blue module were enriched in EMT, PI3K/Akt signaling pathway and other important biological processes and pathways. Moreover, three genes ( ITGB1 , ITGB5 , and OSMR ) of blue module were determined as key genes with higher expression levels and significant prognostic value. ROC analysis revealed that these three genes had excellent diagnostic efficiency for PAAD. These three key genes were highly regulated in PAAD tissues compared with normal controls. A panel of small molecule drugs including thapsigargin, viomycin, adiphenine, and CP-690334-01 were screened out to treat PAAD. Conclusion: In conclusion, our findings identified three key genes implicated in PAAD and screened out several potential small drugs to treat it, which may shed novel insights into the development of PAAD and its treatment.