Identification of Prognosis-Related Genes and Construction of Multi-regulatory Networks in Pancreatic Cancer Microenvironment by Bioinformatics Analysis

Abstract

Background As one of the most lethal cancers, pancreatic cancer has been characterized by abundant supportive tumor-stromal cell microenvironment. Although the advent of tumor-targeted immune checkpoint blockers has brought light to patients with other cancers, its clinical efficacy in pancreatic cancer has been greatly limited due to the protective stroma. Thus, there is an urgent need to explore new targets and develop multi-regulatory networks to improve treatment and evaluate patient prognosis. Methods We followed a strategy based on mining the Cancer Genome Atlas (TCGA) database and ESTIMATE algorithm to obtain the immune scores and stromal scores. Differentially expressed genes (DEGs) associated with poor overall survival of pancreatic cancer were screened from a TCGA cohort. By comparing global gene expression with high vs. low immune scores and subsequent Kaplan-Meier analysis, DEGs that significantly correlate with poor overall survival of pancreatic cancer in TCGA cohort were extracted. After constructing the protein-protein interaction network using STRING and limiting the genes within the above DEGs, we utilized RAID 2.0, TRRUST v2 database and degree and betweenness analysis to obtain non-coding RNA (ncRNA)-pivotal nodes and TF-pivotal nodes. Finally, multi-regulatory networks have been constructed and pivotal drugs with potential benefit for pancreatic cancer patients were obtained by screening in the DrugBank. Results In this study, we obtained 246 DEGs that significantly correlate with poor overall survival of pancreatic cancer in the TCGA cohort. With the advent of 42 ncRNA-pivotal nodes and 7 TF-pivotal nodes, the multi-factor regulatory networks were constructed based on the above pivotal nodes. Prognosis-related genes and factors such as HCAR3, PPY, RFWD2, WSPAR and Amcinonide were screened and investigated. Conclusion The multi-regulatory networks constructed in this study are not only helpful to improve treatment and evaluate patient prognosis with pancreatic cancer, but also beneficial for implementing early diagnosis and personalized treatment. It is suggested that these factors may play an essential role in the progression of pancreatic cancer.