Abstract
BACKGROUND: Ovarian cancer is a highly malignant epithelial tumor. Recently, it has been reported the role of glycosyltransferases (GTs) in various cancers. However, the prognostic value of GTs-related genes in ovarian cancer remained largely unknown. METHODS: RNA-sequencing (RNA-seq) data and corresponding clinical characteristics of patients with ovarian cancer were extracted from the public database of the Cancer Genome Atlas (TCGA) and Genotype-Tissue Expression (GTEx). We constructed the least absolute shrinkage and selection operator (LASSO) Cox regression model to explore a multigene signature comprising GTs-related genes in the TCGA and GTEx cohort. Patients with ovarian cancer in International Cancer Genome Consortium (ICGC) database were applied for further validation. We also performed functional analysis on the differentially expressed genes (DEGs) of high-risk and low-risk groups in the TCGA cohort. Additionally, the immune status between the two risk groups was assessed, respectively. RESULTS: Our results showed that 64 GTs-related genes were differentially expressed between tumor tissues and normal tissues in the TCGA and GTEx cohort. A prognostic model of 15 candidate genes was constructed, which classified patients into high- and low-risk groups. Compared with low-risk patients, high-risk patients had an obvious worse overall survival (OS) (P < 0.0001 in the TCGA and GTEx cohort and P = 0.042 in the ICGC cohort). Multivariate Cox regression analysis revealed that the risk score was an independent factor for OS of ovarian cancer. Functional analysis indicated that these DEGs were also enriched in immune-related pathways, and the immune status was significantly different between the two risk groups in TCGA cohort. CONCLUSION: In conclusion, a novel GTs-related gene signature may be used for the prognosis of ovarian cancer. Targeting GTs-related gene can act as a therapeutic alternative for ovarian cancer.