Abstract
Uterine corpus endometrial carcinoma (UCEC) is the most common type of gynecologic malignancy worldwide. Despite advances in the treatments of UCEC, its incidence and mortality rates are still increasing. N6-methyladenosine (m6A) is the most common form of RNA modification and has attracted increasing interest in cancer pathogenesis and progression. Thus, we aimed to identify the landscape of m6A regulators and build a prognostic gene signature in UCEC. In this study, we first analyzed copy number variations (CNVs), single nucleotide variations (SNVs) and gene expression profiles as well as matched clinical information of UCEC patients from The Cancer Genome Atlas (TCGA) database. Next, we determined that CNVs in m6A regulatory genes had a significant negative impact on patient survival. The mRNA expression levels of a total of 16 m6A regulators were significantly correlated with different CNV patterns. Using univariate Cox regression analysis, IGF2BP1, KIAA1429, IGF2BP3, YTHDF3, and IGF2BP2 were found to be closely associated with UCEC patient survival outcomes. Based on the least absolute shrinkage and selection operator (LASSO) and multivariate Cox regression models, we built a 3-gene (IGF2BP3, KIAA1429 and IGF2BP1) signature of m6A regulators with prognostic value in UCEC that could effectively predict patient prognosis (log-rank test p-value < 0.0001). In addition, risk scores were significantly different between patients stratified by tumor stage, SNV, and CNV. Multivariate Cox regression analysis suggested that risk score might be an independent prognostic indicator for the overall survival of patients with UCEC (p-value < 0.05). Gene enrichment analysis indicated that high IGF2BP1 gene expression is associated with cytoplasmic stress granules. KIAA1429 gene expression is associated with cellular nucleic acid metabolism. The expression of the IGF2BP3 gene is associated with RNA binding processes. In conclusion, we determined that genetic alterations in m6A regulatory genes could be effective and reliable biomarkers for UCEC prognosis prediction.