A Four-miRNA-Based Diagnostic Signature for Rheumatoid Arthritis

BACKGROUND: As a chronic inflammatory disease, rheumatoid arthritis (RA) usually leads to cartilage and bone damage, even disability. Earlier detection and diagnosis are crucial to improve the therapeutic efficacy, and the aim of our study is to identify a potential diagnostic signature for RA. METHODS: We downloaded the GSE124373 dataset from the Gene Expression Omnibus (GEO) database. And differential expression analysis of miRNAs was conducted using the limma package of R language. The potential targeted mRNAs of differentially expressed miRNAs were predicted using the MiRTarBase database. The clusterProfiler package in R language was used to conduct functional enrichment analysis (GO term and KEGG pathway). Then, based on the key miRNAs screened by stepwise regression analysis, the logistic regression model was built and it was evaluated using a 5-fold cross-validation method. RESULTS: A total of 19 differentially expressed miRNAs in the blood sample of RA patients compared with that of healthy subjects were identified. Nine optimal miRNAs were screened by using stepwise regression analysis, and four key miRNAs hsa-miR-142-5p, hsa-miR-1184, hsa-miR-1246, and hsa-miR-99b-5p were further optimized. Finally, a logistic regression model was built based on the four key miRNAs, which could reliably separate RA patients from healthy subjects. CONCLUSION: Our study established a logistic regression diagnostic model based on four crucial miRNAs, which could separate the sample type reliably.