Serologic markers of lupus nephritis in patients: use of a tissue-based ELISA and evidence for immunopathogenic heterogeneity.

In order to assess the ability of various serologic assays to correlate with lupus nephritis, we analysed sera obtained from 60 patients with systemic lupus erythematosus (SLE). Patients were categorized as having active nephritis (group 1), active lupus without nephritis (group 2), inactive lupus with prior nephritis (group 3), or inactive lupus without prior nephritis (group 4). Three parameters were assessed including anti-dsDNA antibodies (Farr assay), immune complexes (C1q binding), and anti-C1q antibodies (salt-stable C1q binding). Additionally, glomerular binding activity (GBA) was measured using a new solid-phase immunoassay that detects immune elements by their ability to bind glomerular tissue. We found that patients with nephritis (group 1) exhibited higher mean values for each assay than patients in each of the other three groups (P = 0.001, 0.009, 0.14, and 0.23 in the GBA, C1q, anti-dsDNA, and anti-C1q assays, respectively). The only assay which distinguished patients with nephritis (group 1) from patients having active disease without nephritis (group 2) was the GBA (mean 0.48 +/- 0.09 versus 0.15 +/- 0.04, P < 0.05). In terms of utility, all tests were specific for diagnosing nephritis among patients with lupus; however, only the GBA was reasonably sensitive. The information provided by the anti-dsDNA and C1q assays were not correlated with one another, nor additive to the GBA. Patients with false negative GBA tended to have received more intensive immunosuppression. The qualitative characteristics of GBA varied among patients with nephritis. These data suggest the pathogenesis of lupus nephritis is complex, and may be mediated by an array of immune elements. Moreover, the data indicate the potential utility for a broad tissue-based approach to detection of pathogenic immune elements over other, specific immunologic markers.