M C GALLICCHIO

Anti-neutrophil cytoplasm antibodies (ANCA) occur occasionally in rheumatoid arthritis (RA), but their incidence and clinical significance have been unclear. In this study we have investigated 58 patients with RA. In 22 patients the disease was inactive and the remaining 36 with active disease were further subdivided into those without clinical evidence of vasculitis (26), those with cutaneous vasculitis (8) and those with systemic vasculitis (2). ANCA were demonstrated by indirect immunofluorescence in 10 of the 58 patients (17%). While both perinuclear (pANCA) and cytoplasmic (cANCA) staining were detected, pANCA were more common (70%). Neutrophil-specific anti-nuclear antibodies (ANNA) were demonstrated in a further eight sera (14%) and ANA were detected on Hep-2 cells in 30 of the 58 sera (52%). ELISAs for the detection of anti-myeloperoxidase and anti-elastase antibodies were then established. Five sera with pANCA and five that contained ANNA were negative for both anti-myeloperoxidase and anti-elastase antibodies, suggesting other as yet unidentified cytoplasmic antigens as the target molecules. However, anti-myeloperoxidase or anti-elastase antibodies were found in four sera that had homogeneous or speckled ANA on both Hep-2 cells and neutrophils. One serum contained both antibodies. The presence of ANCA detected by indirect immunofluorescence or of anti-myeloperoxidase or anti-elastase antibodies in these patients with RA was not associated with disease activity nor with the demonstration of cutaneous vasculitis or renal disease (P NS). A possible association with systemic vasculitis remains to be confirmed. There is an incomplete correlation between indirect immunofluorescence patterns and antibody specificity in ELISA systems.

Autoantibodies that produce a perinuclear pattern on indirect immunofluorescent examination of ethanol-fixed neutrophils (pANCA) are found in about half of all cases of microscopic polyarteritis. These antibodies are often directed against myeloperoxidase or elastase and we have developed sensitive reproducible ELISAs for their detection and study. Seven sera from 19 patients with microscopic polyarteritis or segmental necrotizing glomerulonephritis contained anti-myeloperoxidase or anti-elastase antibodies or both. In contrast, only one of 18 sera from patients with Wegener's granulomatosis, where the pattern of immunofluorescence is predominantly cytoplasmic, had anti-myeloperoxidase antibodies and no anti-elastase antibodies were detected. Using sera from patients with microscopic polyarteritis, both anti-myeloperoxidase and anti-elastase antibodies were demonstrated to be of high affinity. There was no immunoglobulin class, subclass or light chain restriction noted. Anti-myeloperoxidase and anti-elastase antibodies were also found occasionally in anti-glomerular basement membrane disease, mixed connective tissue disease, systemic lupus erythematosus, post-streptococcal glomerulonephritis and atypical pneumonia. In further studies these antibodies were not associated with other lung infections, although anti-elastase antibodies were noted in one of 14 sera positive for ASOT that were tested. Anti-myeloperoxidase antibodies were found more frequently than anti-elastase antibodies and these antibodies were occasionally present together. In addition some sera with pANCA had neither anti-myeloperoxidase nor anti-elastase antibodies. The target molecules in these cases remain unclear.