Investigation of the prevalence and clinical associations of antibodies to human fibronectin in systemic lupus erythematosus.

objectivesTo assess the prevalence of antibodies to human fibronectin (anti-Fn) in sera of patients with certain connective tissue diseases and to determine their association with disease activity and the pattern of organ involvement in patients with systemic lupus erythematosus (SLE).

methodsA capture enzyme linked immunosorbent assay (ELISA) was developed to quantify anti-Fn antibodies in serum samples from 65 patients with well characterised SLE, 50 with rheumatoid arthritis (RA), 15 with Behçet's disease (BD), 15 with systemic vasculitis and 36 healthy subjects. An anti-Fn antibody titre greater than mean + 3SD of the healthy control log values after back transformation to the normal scale was considered positive. Disease activity in SLE patients was scored using the British Isles Lupus Assessment Group (BILAG) Index. Erythrocyte sedimentation rate (ESR), concentrations of anti-dsDNA antibody, soluble interleukin-2 receptors (sIL-2R), C3, C4, C3 degradation products (C3dg) and immunoglobulin, and antinuclear antibody (ANA) titres were measured in blood samples from SLE patients; neopterin concentration was measured in corresponding urine samples.

resultsAnti-Fn antibodies were found in 22 of 65 SLE patients (33.8%), seven of 50 with RA (14%), one of 15 with BD (6.6%) and none of the 15 subjects with vasculitis. Thirty SLE patients had active disease and 35 had inactive disease; their median anti-Fn concentrations were 117 u/ml (range 47-450) and 68 u/ml (range 17-334), respectively (p = 0.0001). The presence of anti-Fn did not correlate with immunoglobulin concentrations or ANA titres in these sera. No significant difference was found between SLE patients with disease activity in one major organ system compared with multiple organ involvement, as defined by BILAG (p = 0.19). However, patients with musculoskeletal manifestations had consistently greater anti-Fn concentrations compared with patients with other clinical manifestations. There were significant correlations between amounts of anti-Fn in SLE sera and ESR (rs = 0.25, p = 0.045), sIL-2R (rs = 0.28, p = 0.024) and urine neopterin (rs = 0.3, p = 0.016) but not with serum anti-dsDNA antibody titres, plasma C3, C3dg or C4. However multiple regression analysis showed a low significant correlation only with sIL-2R and BILAG score (p = 0.047 and 0.042, respectively).

conclusionAnti-Fn antibodies were detected in 34% of SLE patients and in small proportions of RA and BD patients. An association between serum anti-Fn and disease activity in SLE has been identified and most SLE patients with musculoskeletal involvement had increased anti-Fn antibody concentrations.