Specific autoantibody profiles and disease subgroups correlate with circulating micro-RNA in systemic sclerosis

Abstract

OBJECTIVE: To evaluate the expression profiles of cell-free plasma miRNAs in SSc and to characterize their correlation with disease subgroups (lcSSc and dcSSc) and with autoantibody profiles. METHODS: Using quantitative RT-PCR, the abundance of 45 mature miRNAs in plasma was determined in 95 patients (lcSSc = 63; dcSSc = 32), representing the following autoantibody subgroups: ACA, anti-DNA topoisomerase I, anti-RNA polymerase III and anti-U1-ribonucleoprotein. MiRNA data were correlated with clinical and paraclinical data. Multiple regression was used to model membership of the lcSSc, dcSSc and autoantibody subgroups, based on miRNA expression profiles. RESULTS: Thirty-six miRNAs were measurable in all samples. Four (miRNA-223, -181b, -342-3p and -184) were differently expressed in lcSSc and dcSSc (false discovery rate < 0.05). Ten miRNAs exhibited statistically significantly different levels in one or more autoantibody groups, and five (miRNA-409, -184, -92a, -29a and -101) remained significant after correction for multiple comparisons. Multiple regression models accurately predicted ACA and anti-DNA topoisomerase I antibody-positive patients (area under the curve (AUC) = 0.97 and 0.93, respectively) as well as membership of the dcSSc and lcSSc groups (AUC = 0.88). CONCLUSION: Circulating miRNA profiles differ between lcSSc and dcSSc patients and between patients with different autoantibodies. This is the first time autoantibody profiles, disease phenotypes and plasma miRNA profiles have been shown to correlate in an autoimmune disease. The data support a pathobiological role of miRNAs because specific miRNAs associate with autoantibody profiles of known diagnostic and prognostic value.