Abstract
BACKGROUND: Gastric cancer is a leading cause of cancer deaths globally due to its often late diagnosis and poor survival rates. There's an urgent need for reliable non-invasive biomarkers for early detection. Claudin-9 (CLDN9), a protein implicated in epithelial-mesenchymal transition (EMT), has shown elevated expression in various cancers. This study investigates CLDN9's potential as a diagnostic marker for GC, with particular focus on mitochondrial pathway involvement. METHODS: The analysis of CLDN9 expression in gastric cancer was conducted and validated through immunohistochemistry using data from The Cancer Genome Atlas (TCGA) database. The identification of microRNAs regulating CLDN9 utilized machine learning techniques, such as LASSO regression and random forest algorithms. The diagnostic potential of hsa-miR-4496, a primary regulatory miRNA, was evaluated in plasma and saliva samples, with diagnostic accuracy assessed using ROC curve analysis. RESULTS: CLDN9 is significantly overexpressed in GC tissues and is associated with advanced stages and reduced survival rates. Immunohistochemical analysis confirmed the increased expression of CLDN9 protein in tumor tissues. Machine learning algorithms identified hsa-miR-4496 as the primary regulatory factor of CLDN9, with miRNA-regulated mRNA pathway analysis emphasizing that miRNA could exert its effects through the regulation of mitochondrial pathways. Pathway enrichment analysis highlighted mitochondrial processes as key regulatory pathways. Diagnostic evaluation of CLDN9 in plasma and saliva showed an AUC of 0.823, indicating strong diagnostic potential. CONCLUSIONS: The results underscore the potential of CLDN9 and hsa-miR-4496 as promising non-invasive biomarkers for gastric cancer, with mitochondrial pathways being integral to their regulatory mechanisms. These biomarkers present potential for incorporation into clinical protocols, thereby facilitating early intervention and personalized treatment strategies for gastric cancer (GC).