Abstract
BACKGROUND: Rheumatoid arthritis (RA) is a systemic autoimmune disorder marked by chronic synovial inflammation and progressive joint damage. Increasing evidence points to non-coding RNAs, particularly circular RNAs (circRNAs), as crucial regulators of immune and inflammatory responses in RA. However, their functional roles and clinical relevance remain incompletely understood. METHODS: We conducted an integrative analysis combining bioinformatics and experimental validation to investigate the expression profiles of circRNAs and their host genes in RA. Transcriptomic datasets (GSE124373, GSE169082, GSE189338) were analyzed to identify differentially expressed mRNAs, circRNAs, and miRNAs in PBMCs of RA patients. Functional enrichment, protein-protein interaction (PPI) network construction, and competing endogenous RNA (ceRNA) regulatory analyses were performed. Subsequently, qPCR validation was carried out in clinical samples from 25 RA patients and 25 healthy controls. RESULTS: Analysis revealed 1366 differentially expressed mRNAs, 47 circRNAs, and 223 miRNAs. Notably, hsa_circ_0092125 and its host gene G6PD were significantly upregulated in RA samples. The ceRNA network indicated their involvement in immune-inflammatory pathways. qPCR validation confirmed elevated expression of hsa_circ_0092125 (fold change = 4.35, P = 0.0114) and G6PD (fold change = 2.23, P = 0.0048). ROC (Receiver Operating Characteristic) curve analysis demonstrated moderate diagnostic value, particularly for G6PD (area under the curve (AUC) = 0.7824). CONCLUSION: Our integrative bioinformatics and experimental approach identify hsa_circ_0092125 and G6PD as potential biomarkers for RA. These findings enhance our understanding of the molecular mechanisms underlying RA pathogenesis and suggest new avenues for biomarker development and targeted therapies.