Construction and validation of a novel Ferroptosis-related gene signature predictive model in rectal Cancer

• DOI: 10.1186/s12864-022-08996-6
• Publication Date: 2022-11-23T01:27:19Z
• Authors: Wei-Kun Shi; Yu-Xin Liu; Xiao-Yuan Qiu; Jing-Ya Zhou; Jiao-Lin Zhou; Guo-Le Lin
• Keywords: Univariate; Lasso; Gene signature; Hierarchical clustering

Rectal cancer (RC) is one of the most common malignant tumors. Ferroptosis an iron-dependent form cell death, which plays important role in various cancers. However, correlation between ferroptosis-related genes (FRGs) and prognosis RC remains unclear.Gene expression data from The Cancer Genome Atlas Rectum adenocarcinoma (TCGA-READ) GSE87211 were downloaded. Clustering functional enrichment evaluated. A FRGs risk score was established based on univariate Cox analysis Least absolute shrinkage selection operator (LASSO) analysis. K-M ROC conducted to determine prognostic values. qRT-PCR performed validate levels mRNA expression. Multivariate used build a prediction model score.Based FRGs, patients grouped into two clusters. In differentially expressed clusters, immune-related pathways dominated. novel signature with 14 related overall survival (OS) established. identified TP63, ISCU, PLIN4, MAP3K5, OXSR, FANCD2 ATM overexpressed tissue; HSPB1, MAPK1, ABCC1, PANX1, MAPK9 ATG7 underexpressed; TUBE1 had no difference. high-risk group significantly lower OS than low-risk (P < 0.001), curve confirmed signature's predictive capacity. demonstrated that age independent factors.A can be predict RC, as well guide individual treatment.