Objectives: To investigate the clinical value of anti-Sm antibodies in diagnosis and monitoring of systemic lupus erythematosus (SLE) and their ability to predict lupus flares compared with that of anti-dsDNA antibody and complement (C3) assays.
Methods: Autoantibodies against Smith antigen (Sm) and double-stranded DNA (dsDNA) in sera from SLE (n=232), myositis (n=26), systemic sclerosis (n=81), Sjögren's syndrome (n=88), and rheumatoid arthritis patients (n=165) and healthy donors (n=400) were determined by using enzyme-linked immunosorbent assays (both from Euroimmun). New thresholds for both autoantibodies were calculated by receiver operating characteristics (ROC) curve analysis. Cross-sectional, longitudinal and predictive analyses of anti-Sm and disease activity were also performed.
Results: Sensitivities of 25.9% for anti-Sm (cut-off: 3.6 relative units/ml) and 30.2% for anti-dsDNA (cut-off 157.4 international units/ml) were obtained at a specificity of 99%. 14.8% of anti-dsDNA-negative patients were positive for anti-Sm, and more than half (51.4%) of anti-dsDNA-positive patients were also positive for anti-Sm. Anti-Sm antibodies were associated with age (p=0.0174), the number of ACR criteria (p=0.0242), the ACR criteria renal (p=0.0350) and neurologic disorder (p=0.0239), the BILAG category constitutional symptoms (p=0.0227), fatigue (p=0.0311) and cross-sectional disease activity (r=0.2519, p=0.0224). Although no correlations with lupus activity were observed in the longitudinal and predictive analysis, a remarkable association was found between anti-Sm and proteinuria.
Conclusions: Anti-Sm antibodies are essential for diagnosis of SLE, especially in anti-dsDNA-negative patients. However, our data suggest that anti-Sm monitoring is only helpful in SLE patients with active lupus nephritis.