TITLE:
Identification of Prognosis-Related Genes and Key Target Genes for Pancreatic Cancer: A Bioinformatics Analysis
AUTHORS:
Zhonghua Shang, Nicaise Patient Woulaidjei Ntomo, Achi Ntiak, Ernestina Apeku
KEYWORDS:
Pancreatic Cancer, Target Genes, Protein-Protein Network, Bioinformatics
JOURNAL NAME:
Journal of Biosciences and Medicines,
Vol.12 No.6,
June
21,
2024
ABSTRACT: Objective: The mortality and morbidity rates associated with pancreatic cancer (PaCa) are extremely high. Various studies have demonstrated that pancreatic cancer will be the fourth cancer-related death by 2030, raising more concern for scholars to find effective methods to prevent and treat in order to improve the pancreatic cancer outcome. Using bioinformatic analysis, this study aims to pinpoint key genes that could impact PaCa patients’ prognosis and could be used as therapeutic targets. Methods: The TCGA and GEO datasets were integratively analyzed to identify prognosis-related differentially expressed genes. Next, the STRING database was used to develop PPI networks, and the MCODE and CytoNCA Cytoscape in Cytoscape were used to screen for critical genes. Through CytoNCA, three kinds of topology analysis were considered (degree, betweenness, and eigenvector). Essential genes were confirmed as potential target treatment through Go function and pathways enrichment analysis, a developed predictive risk model based on multivariate analysis, and the establishment of nomograms using the clinical information. Results: Overall, the GSE183795 and TCGA datasets associated 1311 and 2244 genes with pancreatic cancer prognosis, respectively. We identified 132 genes that were present in both datasets. The PPI network analysis using, the centrality analysis approach with the CytoNCA plug-in, showed that CDK2, PLK1, CCNB1, and TOP2A ranked in the top 5% across all three metrics. The independent analysis of a risk model revealed that the four key genes had a Hazard Ratio (HR) > 1. The monogram showed the predictive risk model and individual patient survival predictions were accurate. The results indicate that the effect of the selected vital genes was significant and that they could be used as biomarkers to predict a patient’s outcome and as possible target therapy in patients with pancreatic cancer. GO function and pathway analysis demonstrated that crucial genes might affect the P53 signaling pathway and FoxO signaling pathway, through which Meiotic nuclear division and cell cycle may have a significant function in essential genes affecting the outcome of patients who have pancreatic cancer. Conclusions: This study suggests that CDK2, CCNB1, PLK1 and TOP2A are four key genes that have a significant influence on PaCa migration and proliferation. CDK2, CCNB1, PLK1, and TOP2A can be used as potential PaCa prognostic biomarkers and therapeutic targets. However, experimental validation is necessary to confirm these predictions. Our study comes into contributions to the development of personalized target therapy for pancreatic cancer patients.