Diagnosis of seronegative rheumatoid arthritis (SNRA) is difficult due to the lack of diagnostic markers. The study aims to construct a novel diagnostic model based on long noncoding RNAs (lncRNAs) expression and laboratory indicators to provide a new idea for diagnostic methods of SNRA. Differentially expressed lncRNAs in peripheral blood cells of RA patients were screened through eukaryotic long noncoding RNA sequencing and validated by quantitative real-time PCR. Meanwhile, the correlation between lncRNAs expression and laboratory indicators was analyzed. The diagnostic value was evaluated by receiver operating characteristic curve analysis. Finally, combined with laboratory indicators, a diagnostic model for SNRA was constructed based on logistic regression and visualized by nomogram. Expression of ADGRE5, FAM157A, PTPN6 and PTPRE in peripheral blood was significantly increased in RA than healthy donors. Meanwhile, we analyzed the relationship between lncRNAs and erythrocyte sedimentation rate, C-reactive protein and CD4 + T cell-related cytokines and transcription factors. Results showed that FAM157A and PTPN6 were positively related to RORγt, and negatively related to GATA3. Moreover, PTPRE has potential discrimination ability between SNRA and healthy donor (AUC = 0.6709). Finally, we constructed a diagnostic model based on PTPRE, neutrophil count and red blood cell distribution width (RDW). The AUC of the model was 0.939 and well-fitted calibration curves. Decision curve analysis indicated the model had better predict performance in SNRA diagnosis. Our study constructed a novel diagnostic model based on PTPRE, neutrophil count and RDW which may serve as a potential tool for the diagnosis of SNRA.
Keywords: Diagnostic model; Long noncoding RNA; Rheumatoid arthritis; Seronegative RA.
© 2024. The Author(s).