Autoantibodies to the Sm antigens are specifically found in 5 to 30% of patients with systemic lupus erythematosus (SLE) depending on the detection system and the patient group. Several immunoassays designed for research and diagnostic laboratory use have been developed. The autoantigens employed in these tests include purified native proteins, recombinant polypeptides, and synthetic peptides. In the present study, we compared the clinical accuracy of anti-Sm autoantibody assays from commercial suppliers including different conventional enzyme-linked immunosorbent assay (ELISA) systems based on purified Sm antigens, an addressable laser bead assay and a newly developed anti-Sm peptide assay. Although the clinical sensitivity of all assays under investigation was comparable, relatively poor correlations and significant differences in specificity were found with a patient cohort of 150 patients. The sensitivity and specificity were 10 and 94%, respectively, for the anti-Sm ELISA from Euroimmun, 10 and 90%, respectively, for the QuantaLite Sm (INOVA), 12 and 88%, respectively, for the Sm assay in the Varelisa ReCombi ANA profile (Pharmacia Diagnostics), 10 and 94%, respectively, for the QuantaPlex Sm (INOVA), and 12 and 100%, respectively, for the new SmD3 peptide-based ELISA (Varelisa Sm Antibodies). The majority of positive test results within the control groups were found in patients with mixed connective tissue disease. Based on the results, we conclude that the detection of anti-Sm antibodies strongly depends both on the nature of the antigen and on the detection system. Finally, we conclude that the recently identified SmD peptide containing a symmetrical dimethylarginine at position 112 of D3 represents a promising tool for the detection of a highly specific subpopulation of anti-Sm antibodies.