A Model for Predicting the Immune Microenvironment and Prognosis of Patients with Endometrial Cancer Based on Ferroptosis-Associated Genes

Abstract

Background: Endometrial cancer (EC) is one of the most prevalent cancers of the female reproductive tract. Ferroptosis, a novel pattern of programmed cell death, is shown in many studies to be a key factor in the development of EC in close association with the tumour and immunotherapy. The aim of this study was to build suitable risk prognostic models for EC patients assessing ferroptosis- and immune-related differentially expressed genes. Methods: Inthe Cancer Genome Atlas (TCGA) dataset, differentially expressed ferroptosis-related genes (DEFRGs) in EC were identified. Cox regression, least absolute shrinkage and selection operator (LASSO) analyses were used to identify genes independently associated with the risk of suffering EC. After calculating risk scores using selected genes, patients were categorized in two groups depending on the median of the risk score, high- or low-risk group. Immune infiltrating cells, the tumour microenvironment, the tumour mutational burden and clinical were compared between groups. Clinical prognostic predictive models were built with the combination of patient’s clinical information and risk scores. The area under receiver operating characteristic (ROC) curves served as a measure of the models’ accuracy;Kaplan–Meier analysis was conducted to assess 3-, 5- and 7-year survival rate in both groups;And the clinical prognosis of patients was assessed by constructing nomograms. Results: Twelve DEFRGs associated with risk the risk of suffering EC (TRIB3, SLC2A1, CDKN2A, RRM2, ASNS, TSC22D3, AURKA, ATP6V1G2, HIG1, GPX4, TLR4 and SAT1) were identified and used to predict prognosis and survival of EC patients. Both groups showed significant differences in immune infiltrating cells (dendritic cells and CD4+ T cells), somatic mutation status and immune checkpoints (p < 0.05). The 1-, 3-, and 5-year survival rates of EC patients were found to be 3-year area under the curve [AUC (Area under the curve)] = 0.791;5-year AUC = 0.795;And 7-year AUC = 0.783 by establishing a nomogram, respectively. Conclusions: We have created a ferroptosis-related risk model which is capable of effectively predicting the outcome of EC patients, including survival and response to immunotherapy. However, additional research is needed to clarify the underlying mechanisms of ferroptosis in EC.