Comparability of patients with ANCA-associated vasculitis enrolled in clinical trials or in observational cohorts.

OBJECTIVES: To analyse the differences between patients with granulomatosis with polyangiitis (GPA) or microscopic polyangiitis (MPA) entered into randomised clinical trials (RCTs) and those followed in large observational cohorts. METHODS: The main characteristics and outcomes of patients with generalised and/or severe GPA or MPA with a five-factor score ≥ 1 enrolled in the French Vasculitis Study Group (FVSG) or the US-Canadian-based Vasculitis Clinical Research Consortium cohorts were compared to those enrolled in one of 2 FVSG clinical RCTs (WEG91, WEGENT) or 3 European Vasculitis Society clinical trials (CYCLOPS, CYCAZAREM, IMPROVE). RESULTS: 657 patients (65.3% with GPA) in RCTs were compared to 437 in cohorts (90.6% with GPA). RCT patients were older at diagnosis than the cohort patients (56.6 ± 13.9 vs. 46.8 ± 17.3 years), had higher Birmingham vasculitis activity score (19.5 ± 9.1 vs. 16.9 ± 7.4), and more frequent kidney disease (84.0% vs. 54.9%) but fewer ear, nose, and throat symptoms (56.8% vs. 72.2%). At 56 months post-diagnosis, mortality and relapse rates, adjusted for age and renal function, were higher for patients with GPA in RCTs vs. cohorts (10.7% vs. 2.5% [p=0.001] and 22.5% vs. 15.6% [p=0.03], respectively) but similar for patients with MPA (6.2% vs. 6.6% [p=0.92] and 16.6% vs. 10.1% [p=0.39], respectively). CONCLUSIONS: Patients with GPA or MPA in RCTs and those in observational cohorts show important differences that should be remembered when interpreting results based on these study populations.

Pathogenesis of anti-neutrophil cytoplasmic autoantibody (ANCA)-associated vasculitis.

Anti-neutrophil cytoplasmic autoantibody (ANCA)-associated vasculitis is an autoimmune disease in which the contributions of genetic, epigenetic and environmental factors to aetiology and pathogenesis are being unravelled. The ANCA immunoglobulin G targeting proteinase 3 and myeloperoxidase affects several neutrophil functions, usually to augment or dysregulate these, promoting a proinflammatory phenotype whereby neutrophils have enhanced capabilities of causing collateral damage to endothelial and other cells. In addition, B cells are intimately involved in pathogenesis as anti-B cell therapies are highly effective, but the manner of this involvement still needs to be delineated. Similarly, the T cell compartment is disturbed in ANCA vasculitis and numerous alterations in T cell subsets have been described, but recognition of a novel CD8(+) T cell transcription signature which can predict likelihood of relapse in ANCA vasculitis indicates that more needs to be learnt about the influence of T cells in the disease process. Finally, the role of the alternative complement pathway and the potential therapeutic value of its neutralization is under active investigation after compelling studies in murine models have demonstrated that C5 and factor-B knock-out mice are protected.

ANCA-stimulated neutrophils release BLyS and promote B cell survival: a clinically relevant cellular process.

OBJECTIVES: To determine a role for antineutrophil cytoplasmic antibody (ANCA)-activated neutrophils in promoting B cell survival through the release of B lymphocyte stimulator (BLyS). METHODS: Neutrophil BLyS expression was measured by flow cytometry. Concentrations of BLyS in cell supernatants and donor serum samples were measured by ELISA. Cell survival assays were carried out using an L3055 cell line and viability measured by flow cytometry. RESULTS: Tumour necrosis factor α and formyl-Met-Leu-Phe (fMLP) treatment of non-primed neutrophils and treatment of primed neutrophils with anti-PR3 ANCA IgG resulted in a significant increase in surface expression of BLyS within 30 min which returned to basal levels by 2 h. Supernatants from ANCA-stimulated neutrophils were shown to contain increased levels of BLyS and to promote the survival of the centroblast cell line L3055. Serum BLyS concentrations are increased in patients with active ANCA-associated systemic vasculitis and these levels are increased further following 1-3 months of treatment with rituximab. CONCLUSIONS: ANCA specifically causes the release of BLyS from activated neutrophils which can support B cell survival in vitro. The presence of serum BLyS in active disease and its increase following B cell depletion suggest it is an important factor in disease pathogenesis and may facilitate disease relapse.

ANCA-associated vasculitis is linked to carriage of the Z allele of α1 antitrypsin and its polymers.

BACKGROUND: Small studies have linked α1 antitrypsin (α1AT) deficiency to patients with antineutrophil cytoplasmic antibodies (ANCA)-associated vasculitis (AAV). OBJECTIVE: To test the validity and the mechanism of this association between α1AT and AAV. METHODS: The distribution of α1AT deficiency alleles Z and S was compared between 856 White Europeans with AAV and 1505 geographic and ethnically matched healthy controls. Genotyping was performed by allelic discrimination assay. RESULTS: were compared between cases and controls using χ(2) tests. The serum and renal biopsies for α1AT polymers were compared using the polymer-specific 2C1 antibody. The role of α1AT polymers in promoting inflammation was investigated by examining their ability to prime neutrophils for ANCA activation as assessed by CD62L shedding, superoxide production and myeloperoxidase degranulation. Results The Z but not the S allele was over-represented in the patients compared with controls (HR=2.25, 95% CI 1.60 to 3.19). Higher concentrations of polymers of α1AT were detected in serum from patients carrying the Z allele than in those not carrying the Z allele (median (IQR) 1.40 (0.91-3.32) mg/dl vs 0.17 (0.06-0.28) mg/dl, p<0.001); polymers of α1AT were also seen in the renal biopsy of a patient with vasculitic glomerulonephritis. Polymers of α1AT primed neutrophils with CD62L shedding and increased superoxide production following ANCA activation. Carriage of the Z allele was not associated with disease severity, survival or relapse. CONCLUSIONS: The Z but not the S deficiency allele is associated with AAV. Polymers of α1AT are present in the serum and glomeruli of at least some patients with the Z allele, which may promote inflammation through priming of neutrophils.

Endothelial cells augment T cell interleukin 2 production by a contact-dependent mechanism involving CD2/LFA-3 interaction.

We have demonstrated that endothelial cells (EC) augment IL-2 production by PHA-stimulated PBMC or purified CD4+ T cells and that the increase is apparent both in the amount of soluble IL-2 secreted and in the level of specific mRNA detectable by Northern blot hybridization. The ability of EC to affect levels of IL-2 cannot be reproduced by soluble factors, including the cytokines IL-1, IL-6, IFN-gamma, or TNF, conditioned medium from resting EC or IL-1, IFN-gamma- or TNF-treated EC, or from resting PBMC + EC cultures. Separation of the EC and PBMC by a Transwell membrane demonstrated that cell contact was required for augmentation of IL-2 synthesis and that this effect was unlikely to be mediated by a short-lived soluble signal. The cell-cell interaction required the ligand pair CD2/LFA-3, since augmentation could be inhibited by antibodies to these structures. Antibodies to ICAM-1, LFA-1, CD4, and MHC class II were without effect. A contact-dependent pathway involving CD2/LFA-3 interactions also may be used by EC to augment IL-2 production from T cells stimulated more specifically through the TCR/CD3 complex with antibody OKT3. This pathway provides a proliferative advantage to T cells stimulated with OKT3 in the presence of EC and may also be involved in the proliferative response of resting T cells to allogeneic class II MHC-expressing EC. We propose that EC augmentation of T cell IL-2 synthesis may be critical in the ability of EC to elicit primary T cell antigen responses and may have consequences for the development of localized cell-mediated immune reactions.

The role of the endothelium in systemic small vessel vasculitis.

The endothelium is the back-drop against which the effects set in train by interactions between Anti-Neutrophil Cytoplasm Antibodies (ANCA) and neutrophils are played out. This review considers the mechanisms of the endothelial cell injury that may result but also questions the impact of endothelial heterogeneity and endothelial cell activation in facilitation of vasculitic lesions, as well as the potential roles for endothelial-dependent anti-inflammatory mechanisms in controlling inflammation.

Differential glycosylation of polyclonal IgG, IgG-Fc and IgG-Fab isolated from the sera of patients with ANCA-associated systemic vasculitis.

Post-translational modifications (PTMs) of proteins produced in vivo may be tissue, developmentally and/or disease specific. PTMs impact on the stability and function of proteins and offer a challenge to the commercial production of protein biotherapeutics. We have previously reported a marked deficit in galactosylation of oligosaccharides released from polyclonal IgG isolated from sera of patients with the anti-neutrophil cytoplasmic antibodies (ANCA) associated vasculitides; Wegener's granulomatosis (WG) and microscopic polyangiitis (MPA). Whilst normal polyclonal IgG molecules are glycosylated within the IgG-Fc region, approximately 20% of molecules also bear oligosaccharides attached to the variable regions of the light or heavy chain IgG-Fab. It is of interest, therefore to compare profiles of oligosaccharides released from the IgG-Fc and IgG-Fab of normal IgG with that isolated from the sera of patients with WG or MPA. This study shows that whilst the oligosaccharides released from ANCA IgG-Fc are hypogalactosylated those released from IgG-Fab are galactosylated and sialylated. These results show that hypogalactosylation of IgG-Fc is not due to a defect in the glycosylation or processing machinery. It rather suggests a subtle change in IgG-Fc conformation that influences the addition of galactose. Remarkably, this influence is exerted on all plasma cells. Interestingly, a licensed monoclonal antibody therapeutic, produced in Sp2/0 cells, is also shown to be hypogalactosylated in its IgG-Fc but fully galactosylated in its IgG-Fab.

IgG antiendothelial cell autoantibodies from scleroderma patients induce leukocyte adhesion to human vascular endothelial cells in vitro. Induction of adhesion molecule expression and involvement of endothelium-derived cytokines.

IgG autoantibodies that bind human endothelial cells (AECA) were detected by ELISA in 30 of 42 samples of sera from patients with scleroderma. Pretreatment of human umbilical vein endothelial cells with AECA-positive scleroderma sera, or IgG purified from these sera, led to a dose- and time-dependent increase in the ability of the cells to bind human U937 monocytic cells. Threshold-active IgG concentrations were 1-10 micrograms/ml; effects were significant after 3 h and maximal after 6-12 h. IgG from AECA-negative sera or normal sera were without effect. Increased adhesion of U937 cells was accompanied by increased expression of endothelial intercellular adhesion molecule-1, vascular cell adhesion molecule-1, and E-selectin. Transfer of endothelial cell-conditioned media after pretreatment with AECA and immunodepletion of IgG demonstrated the presence of transferable activity that mimicked the effects of AECA. Treatment with neutralizing anticytokine antibodies indicated that IL-1, generated by the endothelial cells in response to AECA, was involved in the upregulation of adhesion molecules and U937 cell adhesion. We conclude that AECA can play a pathogenic role in scleroderma by activating endothelial cells, in part due to autocrine or paracrine actions of IL-1.

Endothelium and regulatory cell interactions.

Endothelial cells are injury targets in vasculitis and other diseases. However, their abilities to regulate their own fate are becoming increasingly recognized and may influence their susceptibility to injury in different vascular beds.

Optimizing the electrical stimulation of retinal ganglion cells.

Epiretinal prostheses aim to restore visual perception in the blind through electrical stimulation of surviving retinal ganglion cells (RGCs). While the effects of several waveform parameters (e.g., phase duration) on stimulation efficacy have been described, their relative influence remains unclear. Further, morphological differences between RGC classes represent a key source of variability that has not been accounted for in previous studies. Here we investigate the effect of electrical stimulus waveform parameters on activation of an anatomically homogenous RGC population and describe a technique for identifying optimal stimulus parameters to minimize the required stimulus charge. Responses of rat A2-type RGCs to a broad array of biphasic stimulation parameters, delivered via an epiretinal stimulating electrode (200 × 200 µ m) were recorded using whole-cell current clamp techniques. The data demonstrate that for rectangular charge-balanced stimuli, phase duration and polarity have the largest effect on threshold current amplitude-cells were most responsive to cathodic-first pulses of short phase duration. Waveform asymmetry and increases in interphase interval further reduced thresholds. Using optimal waveform parameters, we observed a drop in stimulus efficacy with increasing stimulation frequency. This was more pronounced for large cells. Our results demonstrate that careful choice of electrical waveform parameters can significantly improve the efficacy of electrical stimulation and the efficacy of implantable neurostimulators for the retina.

Human vascular endothelial cells do not induce anergy in allogeneic CD4+ T cells unless costimulation is prevented.

Human vascular endothelial cells expressing MHC class II molecules have previously been shown to stimulate the proliferation of allogeneic CD4+ human T lymphocytes. Here we show that allogeneic CD4+ T cells from individual A (TA) respond to class II+ endothelial cells from individual B (EB) by inducing interleukin (IL)-2 mRNA, detectable by semiquantitative polymerase chain reaction, within 12 hr. Responding T cells (TA) that are harvested after 12 hr, rested for 3 days, and then re-exposed to the same class II+ EB stimulators can again respond by proliferation that is equivalent in degree to that observed with third-party class II+ endothelial cells (EC) as stimulators and a little greater than that observed in the primary responses. Incorporation of antibodies to LFA-3, an endothelial costimulatory molecule for T cells, or to both IL-2 and IL-2 receptor (R) during the first-round stimulation prevented the subsequent second-round proliferation of TA to class II+ EB but not to class II+ EC. This nonresponsiveness induced by anti-LFA-3 or anti-IL-2/IL-2R could be overcome by the incorporation of cyclosporine during the first-round stimulation or by incorporation of IL-2 during the second-round stimulation. These observations suggest that class II+ endothelial cells within allografts will not induce anergy in host CD4+ T cells unless costimulation is blocked or the ability of CD4+ T cells to respond by proliferation is prevented; even then the response may be modified by prevailing cyclosporine or IL-2 levels.

An interferon-gamma ELISPOT and immunohistochemical investigation of cytotoxic T lymphocyte-mediated tumour immunity in patients with paraneoplastic cerebellar degeneration and anti-Yo antibodies.

Eight patients with paraneoplastic cerebellar degeneration (PCD) and anti-Yo antibodies were investigated to determine whether there is any association between cytotoxic T lymphocyte (CTL) responses reactive with two previously defined Yo-derived, HLA-A2.1 restricted epitopes (cdr2-1 and cdr2-2) and the presence of tumour-infiltrating CD8+ CTLs. cdr2-1 and cdr2-2-specific CTL responses could not be detected in 5 HLA-A2.1(+) patients in an ex vivo interferon-gamma ELISPOT assay and only 2/9 tumour sections contained CD8(+) intratumoural lymphocytes suggesting a very limited role for CTL-mediated tumour immunity in this patient group, all of whom had evidence of widespread malignancy at the time of diagnosis and/or death.

Human CD4+ T cells proliferate to HLA-DR+ allogeneic vascular endothelium. Identification of accessory interactions.

Serially passaged human endothelial cell (EC) cultures will stimulate highly purified peripheral blood CD4+ T cells to proliferate if and only if the EC cultures are pretreated with IFN-gamma to induce de novo expression of MHC class II molecules, principally HLA-DR. HLA-DR-expressing EC alone appear sufficient to stimulate purified CD4+ T cell proliferation without the involvement of other leukocyte populations, as indicated by the following observations: (1) we find no contaminating leukocytes in our EC cultures by FACS analysis or fluorescence microscopy; specifically, there are no detectable CD45 or HLA-DR expressing cells; (2) neither the EC cultures nor the purified CD4+ T cells contain HLA-DR expressing cells detectable by polymerase chain reaction (PCR) of reverse-transcribed mRNA; (3) the stimulatory capacity of the EC cultures is maintained through serial subculture and through low-density replating, indicating that the stimulatory cell type must proliferate in culture as well as EC; and (4) in contrast to MLRs, the response to EC cultures is not inhibited by pretreatment of the stimulator cells and/or responding T cells with the monocyte toxin L-leucine-O-methyl ester. We have used mAb to investigate the role of various EC and T cell surface molecules in the T cell response. mAb to HLA-DR and CD4 inhibit proliferative responses of CD4+ T cells to EC cultures, as would be expected if T cells recognize and proliferate to IFN-gamma-induced allogeneic class II MHC molecules; whereas, also as expected, mAb to class I MHC molecules were without effect. Proliferation is also inhibited by mAbs to T cell CD2 and LFA-1 beta chain (CD18) and by mAbs to LFA-3 (CD58) and CD44, which are expressed by T cells and EC. mAb to ICAM-1 (CD54, a ligand for LFA-1) provides inconsistent inhibition, and mAb to ICAM-2, used with or without anti-ICAM-1, is not inhibitory. Because both of these mAb block adhesion of LFA-1 expressing T cells to EC, our data suggest that additional ligands for LFA-1 must be important for allogeneic proliferation. mAb to VLA-4 alpha or beta chains (CD49d, CD29) enhance proliferation, presumably through direct costimulation of the T cells by these antibodies. However, a mAb to VCAM-1, an EC ligand for VLA-4, is partially inhibitory.(ABSTRACT TRUNCATED AT 400 WORDS)

Characterization of human IgG-binding xenoantigens expressed by porcine aortic endothelial cells.

Xenoreactive antibodies (XAb) play a major role in the rejection of xenografts. In this study, human IgG XAb that bind to xenoantigens expressed by porcine aortic endothelial cells (PAEC) were characterized, together with their corresponding xenoantigens. Using an ELISA with both fixed and unfixed confluent monolayers of PAEC, XAb of both IgG and IgM classes in pooled and individual normal human serum were identified. The binding of these IgG XAb to the endothelium is mediated by F(ab')2 and the only detectable subclasses that bind to the endothelium are IgG1 and IgG2. On the basis of direct binding experiments, inhibition and antibody adsorption studies, and enzymatic digestions, it is shown that only a minor component of the XAb binding is directed against galactose in an alpha 1,3 linkage with galactose on PAEC surfaces. There is some cross-reactivity with antigens expressed on porcine lymphocytes, but not porcine red blood cells. Histological examination of sections of porcine aortae, snap-frozen and stained using immunoperoxidase techniques, confirmed interaction with the vascular endothelium. Labeling of the PAEC with 125I, followed by cell lysis and immunoprecipitation under reducing conditions, showed binding of IgG XAb to several components on the endothelial cell surface, the most prominent of which have apparent molecular masses of 75 kDa, 110 kDa, 180 kDa, and 210 kDa. The 110-kDa component and the 180-kDa component were sensitive to digestion with endoglycosidase F, which suggests the participation of N-linked carbohydrate structures. These studies demonstrate that human IgG XAb recognize multiple determinants expressed by PAEC, a minor population of which contain alpha 1,3-linked galactose residues. Cross-reactive determinants are expressed on porcine lymphocytes but not porcine red blood cells.

Anti-glomerular basement membrane antibody-mediated nephritis complicating transplantation in a patient with Alport's syndrome.

Loss of an allograft caused by anti-GBM antibody-mediated nephritis is a rare complication of renal transplantation in Alport's syndrome. We describe a patient in whom this occurred. He belongs to the subgroup of patients with hereditary nephritis and deafness with an abnormal Goodpasture antigen, and he developed a high level of circulating anti-GBM antibodies within 20 days of transplantation of a kidney with a presumably normal Goodpasture antigen. The antibody titer fell, only to rise again when he developed evidence of acute infection with CMV. Coincident with this second rise in antibody titer he developed an anti-GBM antibody-mediated crescentic nephritis with resultant loss of graft function and transplant nephrectomy. This case provides support for the hypothesis that the abnormality in the basement membrane in some patients with Alport's syndrome involves the Goodpasture antigen, and raises the possibility that viral infection may have triggered autoantibody production.

The potential roles of vascular endothelium in immune reactions.

Cell-mediated immune reactions are initiated and regulated by antigen specific CD4+ helper T cells. However, T cells cannot function independently. In order for a CD4+ T cell to recognize antigen, it must be presented in association with a class II major histocompatibility complex molecule by another cell type and, in order to lead to functional T-cell activation, the antigen presenting cell must also provide costimulatory signals. Once activated, CD4+ T cells function in vivo by secreting cytokines that elicit an inflammatory infiltrate of other cell types that serves to eliminate the source of foreign antigen. In vivo, the development of inflammation requires vascular responses as well as contributions of blood-derived leukocytes. Although several cell types in vitro can present antigen, provide costimulation, and perform actions that contribute to inflammation, vascular endothelial cells may be uniquely important immune accessory cells because they are anatomically uniquely positioned to function in vivo during cell-mediated immune reactions. In this report, we shall review recent data from our laboratories which further characterize the immune accessory functions of endothelial cells.

International Consensus Statement on Testing and Reporting of Antineutrophil Cytoplasmic Antibodies (ANCA)

Antineutrophil cytoplasmic antibody (ANCA) tests are used to diagnose and monitor inflammatory activity in the primary systemic small vessel vasculitides. ANCA is best demonstrated in these diseases by using a combination of indirect immunofluorescence (IIF) of normal peripheral blood neutrophils and enzyme-linked immunosorbent assays (ELISAs) that detect ANCA specific for proteinase 3 (PR3) or myeloperoxidase (MPO). For ANCA testing in "new" patients, IIF must be performed on all serum samples. Serum samples containing ANCA, any other cytoplasmic fluorescence, or an antinuclear antibody (ANA) that results in homogeneous or peripheral nuclear fluorescence then should be tested in ELISAs for PR3-ANCA and MPO-ANCA. Optimally, ELISAs for PR3-ANCA and MPO-ANCA should be performed on all serum samples. Inclusion of the most recent positive sample in the IIF or ELISA may help demonstrate a change in antibody level. Reports should use recommended terms. Any report of positive neutrophil fluorescence issued before the ELISA results are available should indicate that positive fluorescence alone is not specific for the diagnosis of Wegener granulomatosis or microscopic polyangiitis and that decisions about treatment should not be based solely on the ANCA results.

Leukocyte-endothelial interactions in antineutrophil cytoplasmic antibody-associated systemic vasculitis.

The etiology of ANCA-associated vasculitis is unknown. Currently, it is believed that disease may be triggered by infection with the release of proinflammatory cytokines in genetically susceptible individuals. Priming of PMNs and endothelial cells by these cytokines allows ANCAs to activate PMNs, with damage localized to the endothelium, resulting in early lesions. Damage and activation of endothelial cells produces proinflammatory chemokines and cytokines with influxes of monocytes and T cells that intensify endothelial damage. In the kidney, these changes eventually lead to crescent formation. Antigen-specific memory T cells persist after disease remission with the potential of reactivation and disease relapse. Although our understanding of the pathophysiologic mechanisms of ANCA-associated vasculitis is far greater now than when ANCAs were first identified in 1982, more experimental work in combination with clinical observations is required to further elucidate these mechanisms.

Human IgG xenoreactive antibodies mediate damage to porcine endothelial cells in vitro by both humoral and cellular mechanisms.

A major barrier to the transplantation of a porcine organ into a human recipient is the hyperacute rejection response which has been shown to be mediated by xenoreactive antibodies (XAb) and complement proteins. Here, evidence is presented that normal human sera contain IgG XAb that are able to mediate increased adhesion of polymorphonuclear leucocytes (PMN) to porcine aortic endothelial cells (PAEC) in vitro, to initiate damage to endothelial cells via antibody-dependent cellular cytotoxicity mechanisms (ADCC) along with peripheral blood mononuclear cells (PBMC) and to activate the classical complement cascade. PMN exhibit a 2.8-fold increase in adhesion to PAEC from 19.9 +/- 4% to 56 +/- 1.7% at an IgG concentration of 16 mg/ml. This increase is independent of treatment of the PMN with interferon-gamma. PAEC are lysed in the presence of complement: 42.8 +/- 0.7% lysis occurs with a 1/8 dilution of human serum as a source of immunoglobulin of all classes, while 39.3 +/- 3% lysis occurs with purified IgG at 13 mg/ml in the presence of baby rabbit complement. PAEC are also lysed by PBMC in the presence of human IgG XAb, a maximum of 52.0 +/- 5% being observed at effector-to-target (E:T) ratios of 30:1. PBMC bearing Fc gamma RIII receptors for the Fc portion of the IgG molecule mediate the endothelial cell damage since the anti-Fc gamma RIII monoclonal antibody, 3G8, can inhibit lysis by up to 77 +/- 5%. We conclude that human IgG are able to damage porcine endothelial cells using cellular and humoral mechanisms and that IgG XAb can efficiently activate the classical complement cascade in this model system.

Focal segmental necrotizing glomerulonephritis in rheumatoid arthritis.

We report ten patients with rheumatoid arthritis (RA) who developed a focal segmental necrotizing glomerulonephritis (FSNGN) and extracapillary proliferation typical of vasculitic glomerulonephritis. Five patients also had extrarenal vasculitis. Renal presentation was with renal impairment (n = 9) (median creatinine 726 mumol/l, range 230-1592 mumol/l), microscopic haematuria (n = 8) and proteinuria (n = 10). Nine patients were seropositive for rheumatoid factor and nine had bone erosions. Serum from four of five patients tested by indirect immunofluorescence was positive for antineutrophil cytoplasmic antibody (ANCA) with perinuclear staining. Only three patients had penicillamine or gold therapy. Treatment was with prednisolone and cyclophosphamide (six patients, two of whom were also plasma-exchanged), prednisolone and azathioprine (two patients) and prednisolone alone (two patients). There was a marked improvement in renal function in eight patients. Two patients with dialysis-dependent renal failure recovered renal function, although in one patient this was transient and she required further dialysis 4 months later. Two other patients progressed to dialysis at 3 months and 1 year respectively. Four patients died, one remains dialysis-dependent, and four continue to have good renal function at 5 year follow-up (median creatinine 148.5 mumol/l, range 120-193 mumol/l). One patient was lost to follow-up at 5 years. FSNGN should be considered in all patients with RA and renal impairment, proteinuria and/or microscopic haematuria. This diagnosis appears to be more likely in patients with clinical extrarenal vasculitis, bone erosions or who are seropositive. In these cases, an urgent renal biopsy is indicated.