Background: Psoriasis is a prevalent, chronic, and recurrent inflammatory disorder, and new research reveals that ferroptosis plays a role in the pathogenic process.
Methods: Two microarray datasets (GSE6710 and GSE83582) about psoriasis were collected from the GEO database, and the ferroptosis-related genes were extracted from FerrDb. Enrichment scores associated with ferroptosis were obtained using gene set enrichment analysis (GSEA) and gene set variation analysis (GSVA) in two microarray datasets, and weighted gene co-expression network analysis (WGCNA) was used to identify ferroptosis-associated module genes in psoriasis, and took the intersection of the key modules. Next, the protein-protein interaction (PPI) and Gene Ontology (GO) pathway enrichment analyses, as well as the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment studies, were performed. The hub genes were identified using Cytoscape with CytoHubba, and a receiver operator characteristic (ROC) curve was utilized to discriminate psoriasis from controls.
Results: Ferroptosis-related hub genes in psoriasis (CXCL8, STAT3 and STAT1) were ultimately discovered, and they were able to discriminate psoriasis from controls. The AUC (area under the ROC curve) was above 0.75. To confirm that CXCL8, STAT1, and STAT3 could effectively differentiate psoriasis from controls, the diagnostic value was further validated in the GSE30999, GSE13355, and GSE78097 datasets. Three hub genes had AUC values that were also above 0.75 in three external datasets.
Conclusions: We discovered three hub genes (CXCL8, STAT1, and STAT3) that are linked to ferroptosis in psoriasis and can distinguish psoriasis from controls. As a result, CXCL8, STAT1, and STAT3 might be used as ferroptosis-associated biomarkers for disease diagnosis and treatment monitoring.