Identification of Novel Diagnostic Biomarkers and Classification Patterns for Osteoarthritis by Analyzing a Specific Set of Genes Related to Inflammation

Abstract

Osteoarthritis (OA) is a prevalent joint disease globally. TNFA is recognized as a crucial inflammatory cytokine that plays a significant role in the pathophysiological mechanisms that occur during the progression of OA. However, the TNFA_SIGNALING_VIA_NFKB (TSVN)-related genes (TRGs) during the progression of OA remain unclear. By conducting a combinatory analysis of OA transcriptome data from three datasets, various differentially expressed TRGs were identified. The logistic regression model was used to mine hub TRGs for OA, and a nomogram prediction model was subsequently constructed using these TRGs. To identify new molecular subgroups, we performed consensus clustering. We then conducted functional analyses, including GO, KEGG, GSVA, and GSEA, to elucidate the underlying mechanisms. To determine the immune microenvironment, we applied xCell. The logistic regression analysis identified three hub TRGs (BHLHE40, BTG2, and CCNL1) as potential biomarkers for OA. Based on these TRGs, we constructed an OA predictive model. This model has demonstrated promising results in enhancing the accuracy of OA diagnosis, as evident from the ROC analysis (AUC merged dataset = 0.937, AUC validating dataset = 0.924). We identified two molecular subtypes, C1 and C2, and found that the C1 subtype showed activation of immune- and inflammation-related pathways. The involvement of TSVN in the development and progression of OA has been established. We identified several hub genes, such as BHLHE40, BTG2, and CCNL1, that may have a significant association with the progression of OA. Furthermore, our logistic regression model based on these genes has shown promising results in accurately diagnosing OA patients.

Keywords: diagnosis.; immune microenvironment; inflammation; logistic regression; osteoarthritis.