Abstract
Introduction: Keloid formation is a prevalent dermatological condition characterized by abnormal dermal connective tissue proliferation. Despite ongoing research, the underlying mechanisms of keloid formation remain insufficiently understood. The aim of this research is to identify and verify molecular biomarkers associated with keloid and to explore potential therapeutic targets. Methods: Transcriptomic data from keloid tissue specimens and normal skin controls were retrieved from the Gene Expression Omnibus (GEO) database. We performed differential expression and functional enrichment analyses after batch effect correction. We performed differential gene analysis, weighted Gene Co-expression Network Analysis (WGCNA), and protein-protein interaction (PPI) analyses to verify hub genes, explore their functions, and evaluate their connection to keloid formation, therapeutic potential, and immune-related characteristics. Key genes were validated through experimental assays. Results: 679 differentially expressed genes (DEGs) were identified. Through WGCNA and Venn diagram analysis, 41 DEGs most closely associated with keloid were identified. These 41 overlapping DEGs were confirmed to be markedly involved in metabolic pathways, nucleotide excision repair, and amino acid biosynthesis by functional enrichment analysis. PPI analysis identified CDK7 and DDB2 as hub genes, each demonstrating strong diagnostic performance in ROC curve analysis (AUC = 0.80), with comparable results in validation datasets (AUC = 0.86). Basic experiments confirmed higher expression of CDK7 and DDB2 in keloid tissue compared to normal skin. Conclusion: Our findings demonstrate that CDK7 and DDB2 are promising biomarkers for diagnostic and potential therapeutic targets in keloid, providing novel insights into its pathogenesis and offering promising druggable targets.