Systematic Construction and Validation of a Novel Ferroptosis-Related Gene Model for Predicting Prognosis in Cervical Cancer.

Methods: Datasets containing RNA sequencing and corresponding clinical data of cervical cancer patients were obtained from searching publicly accessible databases. The "NMF" R package was conducted to calculate the matrix of the screened prognosis gene expression. Ferroptosis-related differential genes in cervical cancer were detected using the "limma" R function and WGCNA. The least absolute shrinkage and selection operator (LASSO) algorithm and Cox regression analysis were conducted to develop a novel prognostic signature. The prediction model was verified by the nomogram integrating clinical characteristics; the GSE44001 dataset was used as an external verification. Then, the immune status and tumor mutation load were explored. Finally, immunohistochemistry as well as quantitative polymerase chain reaction (RT-qPCR) was utilized to ascertain the expression of FRGs. Results: Two molecular subgroups (cluster 1 and cluster 2) with different FRG expression patterns were recognized. A ferroptosis-related model based on 4 genes (VEGFA, CA9, DERL3, and RNF130) was developed through TCGA database to identify the unfavorable prognosis cases. Patients in cluster 1 showed significantly decreased overall survival in contrast with those in cluster 2 ( Conclusion: A novel and comprehensive ferroptosis-related gene model was proposed for cervical cancer which was capable of distinguishing the patients independently with high risk for poor survival, and targeting ferroptosis may represent a promising approach for the treatment of CC.