The early diagnosis of ovarian cancer (OC) is critical to improve the prognosis and prevent recurrence of patients. Nevertheless, there is still a lack of factors which can accurately predict it. In this study, we focused on the interaction of immune infiltration and ferroptosis and selected the ESTIMATE algorithm and 15 ferroptosis-related genes (FRGs) to construct a novel E-FRG scoring model for predicting overall survival of OC patients. The gene expression and corresponding clinical characteristics were obtained from the TCGA dataset (n = 375), GSE18520 (n = 53), and GSE32062 (n = 260). A total of 15 FRGs derived from FerrDb with the immune score and stromal score were identified in the prognostic model by using least absolute shrinkage and selection operator (LASSO)-penalized COX regression analysis. The Kaplan-Meier survival analysis and time-dependent ROC curves performed a powerful prognostic ability of the E-FRG model via multi-validation. Gene Set Enrichment Analysis and Gene Set Variation Analysis elucidate multiple potential pathways between the high and low E-FRG score group. Finally, the proteins of different genes in the model were verified in drug-resistant and non-drug-resistant tumor tissues. The results of this research provide new prospects in the role of immune infiltration and ferroptosis as a helpful tool to predict the outcome of OC patients.
Keywords: ferroptosis; ovarian cancer; prognostic; the cancer genome atlas; tumor infiltrating immune cells.
Copyright © 2021 Li, Xiong, Wen and Zhang.