A novel melanoma prognostic model based on the ferroptosis-related long non-coding RNA

Yamin Rao; Jinchao Zhu; Haiyan Zheng; Wei Dong; Qingyuan Lin

doi:10.3389/fonc.2022.929960

A novel melanoma prognostic model based on the ferroptosis-related long non-coding RNA

Front Oncol. 2022 Oct 12:12:929960. doi: 10.3389/fonc.2022.929960. eCollection 2022.

Authors

Yamin Rao¹, Jinchao Zhu¹, Haiyan Zheng¹, Wei Dong², Qingyuan Lin¹

Affiliations

¹ Department of Pathology, Shanghai Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China.
² Department of Pathology, Eastern Hepatobilliary Surgery Hospital, The Second Military Medical University, Shanghai, China.

Abstract

Ferroptosis is an iron-dependent programmed cell death related to the biological process of many kinds of tumors. Long noncoding RNAs (LncRNA) have been found to play essential roles in the tumor, and their functions in the ferroptosis of tumor cells have been partially discovered. However, there is no summary of ferroptosis-related LncRNA and its functions in melanoma. In the present study, we aim to explore the expression profile of ferroptosis-related LncRNA genes and their value in melanoma prognosis by bioinformatics analysis. The expression of ferroptosis-related gene (FRG) from melanoma clinical data was extracted based on the Cancer Genome Atlas (TCGA) database. By screening the RNA expression data of 472 cases of melanoma and 810 cases of normal skin, eighteen ferroptosis-related differential genes were found to be related to the overall survival rate. Furthermore, 384 ferroptosis-related LncRNAs were discovered through constructing the mRNA-LncRNA co-expression network, and ten of them were found with prognostic significance in melanoma by multivariate Cox analysis. Risk assessment showed that the high expression of LncRNA00520 is associated with poor prognosis, while the increased expression of the other LncRNA is beneficial to the prognosis of patients with melanoma. From univariate and multivariate Cox regression analysis, there were ten ferroptosis-related LncRNA risk models towards to be significant independent prognostic factors for patients with melanoma and valuable predictive factors for overall survival (OS)(P<0.05). The ROC curve further suggested that the risk score has relatively reliable predictive ability (AUC=0.718). The protein level of ferroptosis-related genes was verified by the HPA database and IHC test, leading to the discovery that the expressions of ALOX5, PEBP1, ACSL4, and ZEB1 proteins up-regulated in tumor tissues, and existed differences between tumor tissues and normal tissues. In conclusion, we identified ten ferroptosis-related LncRNA and constructed a prognosis model base.

Keywords: bioinformatics analysis; ferroptosis; lncRNA; melanoma; prognosis model.