Identification of New Potential Targets for Pancreatic Ductal Adenocarcinoma by Integrated Bioinformatic Analysis

Abstract

Background/Aim: Pancreatic cancer is the fourth leading cause of cancer-related death in both men and women worldwide. The 5-year relative survival rate for pancreatic ductal adenocarcinoma (PDAC) is 10%, which is the lowest among all cancers. This study aimed to find more effective targets to improve the diagnosis, prognostic prediction, and treatment of PDAC. Materials and Methods: Three datasets were selected from the GEO database. Correlation analysis was used to screen the datasets and samples. Differentially expressed genes were identified using GEO2R. Metascape was used to perform pathway and process enrichment analysis. Survival analysis using the GEPIA2 and Kaplan-Meier plotter databases was conducted to filter hub genes. Principal component analysis and LASSO regression analyses were used to further filter the key genes. Gene expression in PDAC and normal tissues and in different pathological stages was analyzed using the GEPIA2 database. Thereafter, gene expression was detected in three PDAC and HPDE cell lines using real-time polymerase chain reaction. Results: LPAR5, CYP2C18, SERPINH1, ACSL5, and HCAR3 exhibited higher transcription levels in PDAC tissues compared to matched normal tissues, whereas the PNLIP expression was lower. LPAR5, CYP2C18, SERPINH1 and ACSL5 were markedly upregulated in stage IV PDAC. LPAR5, CYP2C18, SERPINH1 and ACSL5 were upregulated in PDAC cell lines. Further verification suggested that the expression levels of these four genes were closely related to histological type, pathologic stage, therapeutic effects and prognosis of pancreatic cancer. Conclusion: LPAR5, CYP2C18, SERPINH1 and ACSL5 may serve as potential diagnostic, prognostic, and therapeutic targets for PDAC.