Subtype Classification based on Ferroptosis-related Genes and Signature for Predicting Outcomes in Cervical Cancer

Abstract BACKGROUND: Cervical cancer (CC) mainly relies on tumor stage to determine patient prognosis and guide treatment, but the prognosis of patients with the same stage still varies greatly. Ferroptosis, a novel iron-dependent programmed cell death, has been reported in a variety of tumors, but its impact on CC prognosis is currently uncertain. Herein, the express situation and prognostic value of ferroptosis-related genes (FRGs) in CC are explored by collecting public database and constructing a corresponding prognostic signature. METHODS: Molecular data and corresponding clinicopathological data from the Cancer Genome Atlas-Cervical squamous cell carcinoma and endocervical adenocarcinoma (TCGA-CESC) cohort and the corresponding 10 Normal tissue samples of cervical canal from the Genotype-Tissue Expression (GTEx) database were collected. Applying univariate logistic regression analysis to identify prognostic FRGs. Subsequently, genes were further screened using differentially expressed genes (DEGs) and a prognostic model was constructed using the least absolute shrinkage and selection operator (LASSO)-COX stepwise regression. Finally, Validation of the risk model is achieved by using the corresponding data in the Gene Expression Omnibus (GEO) database of CC patients and clinical specimens from CC patients were collected for Quantitative reverse transcription PCR (qRT-PCR) validation. RESULTS: Stepwise regression analysis identified five FRG features used to predict outcomes in patients with CC, and further divided patients into two subgroups. KM survival analysis showed that the prognosis of the two subgroups of patients was significantly different, and the Receiver operating characteristic (ROC) curve analysis verified the good specificity and accuracy of the signature. The model was externally validated with the GEO44001 cohort, and the results show that the model has good prognostic power. Finally, detection of clinical specimens by qRT-PCR demonstrated that five FRGs were significantly highly expressed in tumor samples than in normal samples. CONCLUSION: A risk signature based on five FRGs validated to have excellent prognostic ability for CC patients. Our signature predicting outcomes in CC patients can contribute to targeted and personalized therapy for CC patients.