Development and validation of a gene signature for pancreatic cancer: based on inflammatory response-related genes

Abstract

Pancreatic cancer (PC) is one of the most common malignant tumors in the world with a poor prognosis. There were limited studies investigating the genetic signatures associated with inflammatory responses, tumor microenvironment (TME), and tumor drug sensitivity prediction. In the Cancer Genome Atlas (TCGA) dataset, we constructed an inflammatory response-related genes prognostic signature for PC, and predictive ability of the model was assessed via the International Cancer Genome Consortium (ICGC) database. Then, we explored the differences of TME, immune checkpoint genes and drug resistance genes, and the cancer cell sensitivity to chemotherapy drugs between different risk score group. Based on the TCGA and ICGC databases, we constructed and validated a prognostic model, which consisted of 5 genes (including AHR, F3, GNA15, IL18, and INHBA). Moreover, the prognostic model was independent prognostic factors affecting overall survival (OS). The low-risk score group had better OS, and lower stromal score, compared with patients in the high-risk score group. The difference of antigen-presenting cells, T cell regulation, and drug resistance genes between different risk score groups was found. In addition, the immune checkpoint genes were positively correlation to risk score. The expression levels of AHR, GNA15, IL18, and INHBA were related to the sensitivity of anti-tumor chemotherapy drugs. Gene set enrichment analysis (GSEA) showed significant pathway such as calcium signaling pathway and p53 signaling pathway. We successfully constructed a 5-inflammatory response-related gene signature to predict survival, TME, and cancer cell sensitivity to chemotherapy drugs in PC patients. Furthermore, substantiation was warranted to verify the role of these genes in tumorigenesis.

Keywords: Drug resistance genes; Immune checkpoint inhibitors; Inflammatory response–related genes; Pancreatic cancer; Prognostic model; Tumor microenvironment.