Exploration of a Predictive Model for Keloid and Potential Therapeutic Drugs Based on Immune Infiltration and Cuproptosis-Related Genes

The aim of this study was to investigate the correlation between cuproptosis-related genes and immunoinfiltration in keloid, develop a predictive model for keloid occurrence, and explore potential therapeutic drugs. The microarray datasets (GSE7890 and GSE145725) were obtained from Gene Expression Omnibus database to identify the differentially expressed genes (DEGs) between keloid and nonkeloid samples. Key genes were identified through immunoinfiltration analysis and DEGs and then analyzed for Gene Ontology and Kyoto Encyclopedia of Genes and Genomes, followed by the identification of protein-protein interaction networks, transcription factors, and miRNAs associated with key genes. Additionally, a logistic regression analysis was performed to develop a predictive model for keloid occurrence, and potential candidate drugs for keloid treatment were identified. Three key genes (FDX1, PDHB, and DBT) were identified, showing involvement in acetyl-CoA biosynthesis, mitochondrial matrix, oxidoreductase activity, and the tricarboxylic acid cycle. Immune infiltration analysis suggested the involvement of B cells, Th1 cells, dendritic cells, T helper cells, antigen-presenting cell coinhibition, and T cell coinhibition in keloid. These genes were used to develop a logistic regression-based nomogram for predicting keloid occurrence with an area under the curve of 0.859 and good calibration. We identified 32 potential drug molecules and extracted the top 10 compounds based on their P-values, showing promise in targeting key genes and potentially effective against keloid. Our study identified some genes in keloid pathogenesis and potential therapeutic drugs. The predictive model enhances early diagnosis and management. Further research is needed to validate and explore clinical implications.