Diagnostic efficacy of long non-coding RNAs in multiple sclerosis: a systematic review and meta-analysis.

Abstract

Currently, an increasing body of research suggests that blood-based long non-coding RNAs (lncRNAs) could serve as biomarkers for diagnosing multiple sclerosis (MS). This meta-analysis evaluates the diagnostic capabilities of selected lncRNAs in distinguishing individuals with MS from healthy controls and in differentiating between the relapsing and remitting phases of the disease. We conducted comprehensive searches across seven databases in both Chinese and English to identify relevant studies, applying stringent inclusion and exclusion criteria. The quality of the selected references was rigorously assessed using the QUADAS-2 tool. The analysis involved calculating summarized sensitivity (SSEN), specificity (SSPE), positive likelihood ratio (SPLR), negative likelihood ratio (SNLR), and diagnostic odds ratio (DOR) with 95% confidence intervals (CIs). Accuracy was assessed using summary receiver operating characteristic (SROC) curves. Thirteen high-quality studies were selected for inclusion in the meta-analysis. Our meta-analysis assessed the combined diagnostic performance of lncRNAs in distinguishing MS patients from healthy controls. We found a SSEN of 0.81 (95% CI: 0.74-0.87), SSPE of 0.84 (95% CI: 0.78-0.89), SPLR of 5.14 (95% CI: 3.63-7.28), SNLR of 0.22 (95% CI: 0.16-0.31), and DOR of 23.17 (95% CI: 14.07-38.17), with an AUC of 0.90 (95% CI: 0.87-0.92). For differentiating between relapsing and remitting MS, the results showed a SSEN of 0.79 (95% CI: 0.71-0.85), SSPE of 0.76 (95% CI: 0.64-0.85), SPLR of 3.34 (95% CI: 2.09-5.33), SNLR of 0.28 (95% CI: 0.19-0.40), and DOR of 12.09 (95% CI: 5.70-25.68), with an AUC of 0.84 (95% CI: 0.81-0.87). This analysis underscores the significant role of lncRNAs as biomarkers in MS diagnosis and differentiation between its relapsing and remitting forms.