ANCA-associated renal vasculitis

Abstract

A 62-year-old white woman presented to another hospital with a five-week history of profound malaise and small joint arthralgia. Initially she had noticed a mild sore throat that settled spontaneously. She had transient redness of her left conjunctiva. One week after the onset of symptoms, she had a single episode of macroscopic hematuria for which she received antibiotics from her family doctor. She denied dysuria or urinary frequency. For one week prior to admission, she had suffered nausea and vomiting without abdominal pain or bowel disturbance. She had lost one stone (14 pounds) in weight. She had been hypertensive five years previously, but she had had no other antecedent illnesses. There was no family history of renal disease, hypertension, or rheumatic complaints. She had four adult children, all of whom were well. Medication on admission comprised a beta blocker (atenolol, 100 mg daily) for hypertension. She did not smoke cigarettes and drank less than three units (24 g) of alcohol per week. She was a poultry farmer and lived in a rural setting. Clinical examination revealed that she was pale and dehydrated. She had no rash, no active synovitis, and no lymphadenopathy. Her temperature was 37.2°C; pulse, 70 beats/min; and blood pressure, 160/90 mm Hg lying and 145/80 mm Hg standing. Her jugular venous pressure was not elevated and she had no sacral or ankle edema. The heart sounds were normal with no added sounds. The peripheral pulses were present, equal, and symmetrical. No abnormalities were found in the respiratory, abdominal, or nervous systems. She had no retinopathy and ophthalmoscopy revealed normal fundi. Urinalysis disclosed 3+ blood and 1+ protein. Initial laboratory investigations demonstrated a serum creatinine of 513 μmol/L (5.8 mg/dL); hemoglobin, 9.2 g/dL; and normal white blood cell and platelet levels. Renal ultrasound revealed two normal-sized kidneys. She was rehydrated with normal saline, given a one-unit blood transfusion, and transferred to the Queen Elizabeth Hospital in Birmingham after 48 hours. At that point she was passing 1.0 to 1.5 liters urine/24 h and her blood pressure was 170/100 mm Hg. Further laboratory investigations demonstrated a serum creatinine of 317 μmol/L (3.6 mg/dL); BUN, 23 mmol/L; and serum albumin, 2.9 g/dL. The hemoglobin level was 8.8 g/dL with normal indices. The erythrocyte sedimentation rate was 70 mm/hr. C-reactive protein was 30 mg/L (normal <10 mg/L), and she had a positive antineutrophil cytoplasmic antibody (ANCA) test to a titer openface> 1:400 serum dilution by indirect immunofluorescence with a perinuclear pattern. Subsequent antigen-specific ELISA confirmed reactivity to myeloperoxidase (MPO), denoting MPO-ANCA Figure 1. The ANA, anti-DNA, and anti-GBM antibody tests were negative. Complement C3 and C4 levels were normal, and cryoglobulins were not detectable. Hepatitis B and C serologies were negative. After control of blood pressure, a renal biopsy was performed; the tissue contained eight glomeruli. Of these, one was globally sclerosed, three were normal, and four contained acute segmental lesions of various sizes with thrombosis, tuft disruption, and a few cells in Bowman's space Figure 2. Tubules were acutely damaged, with blood in a few. A patchy infiltrate of chronic inflammatory cells was present. Small arteries and arterioles appeared virtually normal. Immunoperoxidase study revealed no significant immunoprotein deposition within glomeruli.Figure 2Glomerulus in the patient's first renal biopsy showing an acute segmental lesion with thrombosis, disruption of capillary loops and adhesion of the tuft to Bowman's capsule. Periodic acid-methenamine silver (PA silver) ×250.View Large Image Figure ViewerDownload (PPT) The clinical history, renal biopsy findings, and blood serology were consistent with an acute vasculitic (pauci-immune) glomerulonephritis of the microscopic polyangiitis type. An assessment of vasculitic activity showed a Birmingham Vasculitis Activity Score (BVAS) of 13 and a Vasculitis Damage Index (VDI) score of 0. Therapy with prednisolone, 1 mg/kg/day, and oral cyclophosphamide, 2 mg/kg/day, was begun. Following 3 months of therapy, her serum creatinine was 95 μmol/L (1.1 mg/dL) with a creatinine clearance of 87 mL/min. ANCA were no longer detectable Figure 1. The BVAS score was 0, indicating complete remission, and the VDI was 2, indicating a mild to moderate degree of damage. The prednisolone dosage had been gradually reduced according to our local guidelines, and at 3 months she was receiving 15 mg/day. The cyclophosphamide dose had been maintained throughout, as there had been no episodes of leukopenia (white blood cell count <4.0 × 109/L). As she was in remission, cyclophosphamide was discontinued and azathioprine was commenced at 2 mg/kg/day. She remained well with no clinical evidence of vasculitic activity, and 18 months from diagnosis (15 months from the time of remission) her therapy was prednisolone, 5 mg/day, and azathioprine, 100 mg/day. In addition, she was receiving three agents for control of hypertension, calcium and vitamin D tablets (later changed to Didrone® PMO), and an H2 receptor antagonist. At 17 months (∼500 days), MPO-ANCA again was detectable in the serum Figure 1. Therapy was not altered because there was no clinical evidence of vasculitis, urinalysis showed 1+ protein, renal function was stable, and the C-reactive protein level was <10 mg/L. She remained under review in the clinic. She was well until 44 months following her acute illness, when she presented to the renal vasculitis outpatient clinic with a 4-week history of fatigue, upper respiratory tract symptoms of coryza and a sore throat, and new episcleritis. The BVAS score had risen to 8. Urinalysis revealed 3+ blood and 1+ protein. The serum creatinine had risen to 125 μmol/L (1.4 mg/dL). The MPO-ANCA titers had risen further (Figure 1; ∼1320 days), and the C-reactive protein level was 37 mg/L. A nasal swab did not reveal carriage of Staphylococcus aureus, and although 6 of 6 nasal swabs taken during the first 6 months of illness had been positive for this organism, later swabs had been negative. A relapse of microscopic polyangiitis was suspected. She was admitted to the hospital and a second renal biopsy was performed. The specimen contained 15 glomeruli, one globally sclerosed, one with a small old area of capsular adhesion, 5 normal, and 8 with segmental lesions of various sizes showing thrombosis, tuft disruption, and a few cells in Bowman's space Figure 3. The tubules appeared acutely damaged, with blood in a few and some patchy atrophy. Interstitial tissues were edematous and contained a patchy inflammatory infiltrate. Small arteries and arterioles had chronic intimal thickening. One small artery contained a patch of fibrinoid necrosis Figure 4. Immunoperoxidase staining showed no significant deposition of immunoproteins in the glomeruli. Features were those of reactivation of severe renal vasculitis, both acute vasculitic glomerulonephritis and an acute arteritis. Daily oral cyclophosphamide and prednisolone were begun as previously. Within 10 weeks, the BVAS score was 0, urinalysis was negative for blood and protein, serum creatinine had returned to 96 μmol/L (1.1 mg/dL), C-reactive protein was <10 mg/L, and ANCA were no longer detectable in the serum.Figure 4Small artery in the patient's second renal biopsy showing localized destruction of the wall with acute inflammation. PA silver × 250.View Large Image Figure ViewerDownload (PPT) DR. CAROLINE O. S. SAVAGE (Professor of Nephrology, MRC Centre for Immune Regulation, and Division of Medical Sciences, The Medical School, The University of Birmingham, United Kingdom): The patient today is typical of many with ANCA-associated renal vasculitis of the microscopic polyangiitis type. The presenting signs of systemic disease were subtle, but the marked fatigue, arthralgia, and episcleritis attest to the systemic nature of microscopic polyangiitis. In fact, the presentation of this patient is almost identical to that of Patient 1 presented by Dr. Ronald Falk in the last Nephrology Forum to consider vasculitis, 11 years ago1Falk R.J. ANCA-associated renal disease.Kidney Int. 1990; 38: 998-1010Abstract Full Text PDF PubMed Scopus (100) Google Scholar. That patient had cANCA with reactivity towards proteinase 3 by ELISA, denoting PR3-ANCA. At the time there was uncertainty as to the spectrum of ANCA-associated vasculitides, and three types were proposed. Patient 1 was given the diagnostic label “polyarteritis nodosa”; the other two suggested types of ANCA-associated vasculitis were Wegener's granulomatosis and “idiopathic” crescentic glomerulonephritis. The question of specificity of cANCA for Wegener's granulomatosis was raised, given that Patient 1 did not fulfill the diagnostic criteria developed by Godman and Churg in the 1950s2Godman G.C. Churg J. Wegener's granulomatosis: pathology and review of the literature.Arch Pathol. 1954; 58: 533-551PubMed Google Scholar. In the intervening 11 years since Falk's Forum, diagnostic criteria have been put forward by the Chapel Hill International Consensus Conference3Jennette J.C. Falk R.J. Andrassy K. et al.Nomenclature of systemic vasculitides: The proposal of an International Consensus Conference.Arthritis Rheum. 1994; 37: 187-192Crossref PubMed Scopus (3362) Google Scholar, large multinational studies have been undertaken to examine the sensitivity and specificity of ANCA for vasculitis, and three ANCA-associated vasculitides have become widely recognized, namely Wegener's granulomatosis, microscopic polyangiitis, and Churg-Strauss syndrome. Nowadays Patient 1 would be considered typical of microscopic polyangiitis. A variety of descriptive labels have been applied to this vasculitis limited to the kidney, including idiopathic, crescentic, and pauci-immune, and the disorder is now recognized for its propensity to progress to systemic microscopic polyangiitis or Wegener's granulomatosis if untreated4JENNETTE J.C. FALK R.J. Renal and systemic vasculitis,.2000: 5.28.1-5.28.14Google Scholar. The International Consensus Conference in 1994 made an important attempt to clarify the nomenclature for vasculitis, attaching definitions to the major recognized syndromes to facilitate international understanding and awareness3Jennette J.C. Falk R.J. Andrassy K. et al.Nomenclature of systemic vasculitides: The proposal of an International Consensus Conference.Arthritis Rheum. 1994; 37: 187-192Crossref PubMed Scopus (3362) Google Scholar. Definitions were provided for giant cell arteritis, Takayasu arteritis, polyarteritis nodosa, Kawasaki disease, Wegener's granulomatosis, microscopic polyangiitis, Churg-Strauss syndrome, Henoch-Schönlein purpura, cryoglobulinemic vasculitis, and cutaneous leukocytoclastic angiitis. Three diseases, Wegener's granulomatosis, microscopic polyangiitis (acquiring clear distinction from polyarteritis nodosa and with the preferred term “microscopic polyangiitis” over “microscopic polyarteritis”), and Churg-Strauss syndrome, were acknowledged to be “commonly associated with ANCA.” The association of Wegener's granulomatosis, microscopic polyangiitis, and Churg-Strauss syndrome with ANCA has been backed up by several large, including multinational, studies which I will discuss later. The definitions for these three entities appear in Table 1.Table 1Diseases commonly associated with ANCAWegener's granulomatosisGranulomatous inflammation involving the respiratory tract, and necrotizing vasculitis affecting small- to medium-sized vessels (e.g., capillaries, venules, arterioles, and arteries). Necrotizing glomerulonephritis is common.Microscopic polyangiitisNecrotizing vasculitis with few or no immune deposits affecting small vessels (i.e., capillaries, venules, or arterioles). Necrotizing arteritis involving small and medium-sized arteries can be present. Necrotizing glomerulonephritis is very common. Pulmonary capillaritis often occurs.Churg-Strauss syndromeEosinophil-rich and granulomatous inflammation involving the respiratory tract, and necrotizing vasculitis affecting small- to medium-sized vessels; associated with blood eosinophilia and usually asthma or other form of atopy.Modified with permission from Jennette et al3Jennette J.C. Falk R.J. Andrassy K. et al.Nomenclature of systemic vasculitides: The proposal of an International Consensus Conference.Arthritis Rheum. 1994; 37: 187-192Crossref PubMed Scopus (3362) Google Scholar. Open table in a new tab Modified with permission from Jennette et al3Jennette J.C. Falk R.J. Andrassy K. et al.Nomenclature of systemic vasculitides: The proposal of an International Consensus Conference.Arthritis Rheum. 1994; 37: 187-192Crossref PubMed Scopus (3362) Google Scholar. A vascular pathology is shared among these three disorders. Focal necrotizing lesions are the common vascular pathology that characterizes these three disorders. These lesions can affect many different vessels and lead to a variety of symptoms and signs. For example, involvement of glomerular capillaries causes nephritis, of alveolar capillaries causes lung hemorrhage, of epineural arteries causes mononeuritis multiplex, and of dermal venules causes purpura. In the kidney, early glomerular lesions have focal segmental necrosis of capillary loops with some thrombosis and neutrophil infiltration; with progression, mononuclear cells are recruited, and breaks in the glomerular basement membrane quickly lead to crescent formation4JENNETTE J.C. FALK R.J. Renal and systemic vasculitis,.2000: 5.28.1-5.28.14Google Scholar,5Adu D. Howie A.J. Vasculitis in the kidney.Curr Diag Pathol. 1995; 2: 73-77Abstract Full Text PDF Scopus (23) Google Scholar. The acute lesions evolve into sclerotic lesions. Immunohistology shows little deposition of immune reactants; this feature distinguishes lesions due to ANCA-associated vasculitis from those of antiglomerular basement membrane disease, IgA nephropathy, and lupus nephritis. Patients with Wegener's granulomatosis and Churg-Strauss syndrome have additional granulomatous necrotizing lesions, with areas of necrosis surrounded by mixed infiltrates of neutrophils, lymphocytes, monocytes, macrophages, and scattered multinucleate giant cells. These necrotic areas are found most often in the respiratory tract and are separate from “granuloma-like” structures that can develop in the kidney from periglomerular infiltrates located around involved glomeruli. Eosinophils are very conspicuous in the lesions of Churg-Strauss syndrome. The relative rarity of vasculitis, the lack of clear definitions, and the overlap among syndromes have made the collection of epidemiologic data difficult [reviewed in 6]. In the 1970s, the overall annual incidence for all forms of systemic vasculitis occurring in the west of England was 10/million. Data collected in Leicester, United Kingdom in 1980 to 1986 and 1987 to 1989 suggested the combined annual incidence of Wegener's granulomatosis and microscopic polyangiitis to be 1.5/million and 6.1/million, respectively. In the United States over a similar period (1979 to 1988), the prevalence of Wegener's granulomatosis was approximately 30/million. During 1988 to 1998, the overall incidence of ANCA-associated vasculitides was 21.5/million within the Norwich area in east England. During 1988 to 1992 and 1993 to 1998, the incidence was 17.4/million and 23.8/million, respectively; this rate suggests an increased incidence of ANCA-associated vasculitis. Studies from Norway suggest a doubling incidence of Wegener's granulomatosis from 1992-1994 to 1995–1998. These figures probably reflect a real increase in incidence as well as improved awareness and diagnosis. A very high incidence of microscopic polyangiitis (24/million) is seen in Kuwait. Vasculitides must be recognized early if successful treatment is to be implemented. ANCA-associated vasculitides can occur at any age, including in children and the elderly; the peak occurrence is in the 55- to 70-year age group6Scott D.G.I. Epidemiology of systemic vasculitis, increasing incidence?.Clin Exp Immunol. 2000; 120: 19-20Google Scholar. Early diagnosis is important for reducing the ability of acute vasculitis to cause death from major organ failure, for example, respiratory failure from pulmonary hemorrhage, and to reduce major long-term morbidity, for example, end-stage renal failure. As in our patient at the time of her relapse, severe, biopsy-proven disease can be accompanied by apparently mild clinical organ involvement. Our patient's serum creatinine was 1.4 mg/dL. Although the causes of ANCA-associated vasculitides are unknown, the presence of ANCA suggests an autoimmune basis. Witebsky and colleagues used modifications of Koch's postulates, which had been developed to establish the pathogenicity of a transmissible agent, to define an autoimmune etiology for a disease7Witebsky E. Rose N.R. Terplan K. et al.Chronic thyroiditis and autoimmunization.JAMA. 1957; 164: 1439-1447Crossref PubMed Scopus (248) Google Scholar. Thus, an autoantibody or cell-mediated immune response is required, the corresponding antigen must be identified, an analogous autoimmune response must be induced in experimental animals, and the immunized animal must develop similar disease. Rose and Bona identified direct, indirect, and circumstantial evidence that supports an autoimmune cause for ANCA-associated vasculitis8Rose N.R. Bona C. Defining criteria for autoimmune diseases (Witebsky's postulates revisited).Immunol Today. 1993; 14: 426-430Abstract Full Text PDF PubMed Scopus (580) Google Scholar. Direct evidence includes transfer of patient serum, purified immunoglobulin, or autoantigen-specific T-cells to experimental animals, or transplacental transfer of pathogenic IgG. For ANCA-associated vasculitides, direct evidence of autoimmunity is lacking. Retinal vasculitis has been reported in a single individual after intravenous human immunoglobulin containing ANCA was used for treatment of a nonvasculitic condition9Ayliffe W. Haeney M. Roberts S.C. Lavin M. Uveitis after antineutrophil cytoplasmic antibody contamination of immunoglobulin replacement therapy.Lancet. 1992; 339: 558-559Abstract PubMed Scopus (33) Google Scholar, but in other studies intravenous immunoglobulin has been used successfully to treat ANCA-associated vasculitis10Jayne D. Treatment of systemic vasculitis with pooled intravenous immunoglobulin.Lancet. 1991; 337: 1137-1139Abstract PubMed Scopus (273) Google Scholar. There has been no documentation of transferred ANCA inducing disease in experimental animals, analogous to the induction of glomerulonephritis in squirrel monkeys after injection of antiglomerular basement membrane antibodies eluted from the kidneys of patients who had died from Goodpasture's disease. About 20 cases of Wegener's granulomatosis have been reported in pregnant women. The pregnancies were successful and little evidence indicates transfer of disease to the fetus [reviewed in11Haber M.A. Tso A. Taheri S. et al.Wegener's granulomatosis in pregnancy: The therapeutic dilemma.Nephrol Dial Transplant. 1999; 14: 1789-1791Crossref Scopus (33) Google Scholar. The cause of fetal death in two reported spontaneous abortions was not provided. Indirect evidence includes presence of autoantibodies or self-reactive T-cells in the target organs or lesions of the disease, or reproduction of autoimmune disease in experimental animal models. ANCA-associated vasculitides have long been associated with a lack of immunoglobulin deposition in lesions4JENNETTE J.C. FALK R.J. Renal and systemic vasculitis,.2000: 5.28.1-5.28.14Google Scholar. CD4+ and CD8+ T-cells, present in lesions, are predominantly of the CD45RO memory phenotype, but their antigen specificity is unknown12Cunningham M.A. Huang X.R. Dowling J.P. et al.Prominence of cell-mediated immunity effectors in “pauci-immune” glomerulonephritis.J Am Soc Nephrol. 1999; 10: 499-506PubMed Google Scholar. Antigen-specific, PR3- or MPO-reactive T-cells have been detected in the peripheral blood of patients with ANCA-positive vasculitis13Griffith M.E. Coulthart A. Pusey C.D. T cell responses to myeloperoxidase (MPO) and proteinase 3 (PR3) in patients with systemic vasculitis.Clin Exp Immunol. 1996; 103: 253-258Crossref PubMed Scopus (96) Google Scholar,14King W.J. Brooks C.J. Holder R. et al.T lymphocyte responses to ANCA antigens are present in patients with ANCA-associated systemic vasculitis and persist during disease remission.Clin Exp Immunol. 1998; 112: 529-546Crossref Scopus (74) Google Scholar. A common dominating T-cell receptor BV8-F/L-G-G-A/Q-G-J2S3 β chain sequence was found in CD4+ T-cells from four unrelated patients with vasculitis, all of whom were HLA-DRB1*0401 allele positive; this finding suggested that all four patients had been exposed to, and elicited a cell-mediated immune response against, a common antigen15Grunewald J. Halapi E. Wahlstrom J. et al.T-cell expansions with conserved T-cell receptor beta chain motifs in the peripheral blood of HLA-DRB1*0401 positive patients with necrotizing vasculitis.Blood. 1998; 92: 3737-3744PubMed Google Scholar. Attempts have been made to reproduce vasculitis in experimental animals [reviewed in16Heeringa P. Brouwer E. Cohen Tervaert J.W. et al.Animal models of anti-neutrophil cytoplasmic antibody associated vasculitis.Kidney Int. 1998; 53: 253-263Abstract Full Text PDF PubMed Scopus (98) Google Scholar. Brown-Norway rats immunized with human MPO develop antihuman MPO antibodies that cross-react with rat MPO, but vasculitis does not develop. However, rats subsequently subjected to other manipulations such as renal perfusion of neutrophil lysosomal extract (consisting primarily of MPO) followed by either H2O2 perfusion or clamp ischemia, develop a crescentic necrotizing glomerulonephritis. Attempts at reproducing this animal model have resulted in significant immune complex deposition, but the original observations, together with an ability of circulating anti-MPO antibodies to aggravate subnephritogenic anti-glomerular basement membrane (GBM) disease in rats, suggest that anti-MPO antibodies could have pathogenic co-factor potential. That MPO-ANCA alone are insufficient to induce vasculitis is also suggested by studies with SCG/Kj mice derived from BXSBx MRL/Mp-lpr/lpr F1 hybrid mice. These mice develop a crescentic glomerulonephritis together with anti-MPO antibodies that can bind to murine neutrophils. However, transfer of MPO-specific hybridomas induces proteinuria but no vasculitis or nephritis. Thus, even if MPO and neutrophils are necessary for development of vasculitis, they are not sufficient to induce it. Other experimental models such as mercuric chloride (HgCl2)-induced vasculitis or spontaneous vasculitis in autoimmune MRL/lpr mice do not truly replicate human ANCA-associated vasculitides, either in terms of clinical course or pathology, whereas the ANCA develop as part of a polyclonal antibody response. Attempts at developing PR3-ANCA-associated vasculitis have largely been unsuccessful, mainly because of the poor homology between rodent and human PR3. Altogether, perhaps the most important observation from the animal models is that ANCA alone are not pathogenic. Additional factors inducing priming or activation of neutrophils, monocytes, or vascular endothelial cells are required, after which the inflammatory process can be exacerbated by ANCA-neutrophil-endothelial cell interactions, and perpetuated by T-cells and monocytes. Circumstantial evidence includes association of autoimmune mediators such as autoantibody with disease activity, association with other autoimmune diseases, lymphocytic infiltration in target organs, major histocompatibility complex class-II associations, and response to immunosuppression. Considerable circumstantial evidence suggests an autoimmune cause in the three ANCA-associated vasculitides. A strong association exists between vasculitis and the presence of ANCA. Recent large multicenter studies combining immunofluorescence and enzyme immunoassays have demonstrated a specificity approaching 99% for Wegener's granulomatosis, microscopic polyangiitis, or their renal-limited variant17Hagen E.C. Daha M.R. Hermans J. et al.Diagnostic value of standardized assays for anti-neutrophil cytoplasmic antibodies in idiopathic systemic vasculitis.Kidney Int. 1998; 53: 743-753Abstract Full Text PDF PubMed Scopus (643) Google Scholar,18Savige J. Gillis D. Davies D. et al.International Consensus Statement on testing and reporting of antineutrophil cytoplasmic antibodies (ANCA).Am J Clin Pathol. 1999; 111: 507-513Crossref PubMed Scopus (500) Google Scholar. In a single-center study of 123 patients with Wegener's granulomatosis or microscopic polyangiitis, ANCA positivity was 97% according to indirect immunofluorescence and ELISA, the newer capture-ELISA technique19Westman K.W. Bygren P.G. Olsson H. et al.Relapse rate, renal survival, and cancer morbidity in patients with Wegener's granulomatosis or microscopic polyangiitis with renal involvement.J Am Soc Nephrol. 1998; 9: 842-852PubMed Google Scholar. In a meta-analysis of cANCA testing in Wegener's granulomatosis, the pooled sensitivity was 91% (CI, 87% to 95%) and the pooled specificity was 99% (CI, 97% to 99.9%) for the subset of patients with active disease, compared with 63% (CI, 57% to 69%) and 99.5% (CI, 99.1% to 99.7%) for those with inactive disease, where confidence intervals are given20Rao J.K. Weinberger M. Oddone E.Z. et al.The role of antineutrophil cytoplasmic antibody (cANCA) testing in the diagnosis of Wegener's granulomatosis.Ann Intern Med. 1995; 123: 925-932Crossref PubMed Scopus (340) Google Scholar. Methodologic discrepancies and flaws might have diluted the association between ANCA and its associated vasculitides in earlier studies. The correlation of ANCA titers with clinical disease activity supports an autoimmune cause and direct pathogenicity of ANCA21Cohen Tervaert J.W. van der Woude F.J. Fauci A.S. et al.Association between active Wegener's granulomatosis and anticytoplasmic antibodies.Arch Intern Med. 1989; 149: 2461-2465Crossref PubMed Google Scholar,22Jayne D.R. Gaskin G. Pusey C.D. Lockwood C.M. ANCA and predicting relapse in systemic vasculitis.Q J Med. 1995; 88: 127-133Google Scholar. Further, in one study, disease relapse could be predicted and prevented by intensifying immunosuppression in response to rising titers23Cohen Tervaert J.W. Huitema M.G. Hene R.J. et al.Prevention of relapses in Wegener's granulomatosis by treatment based on antineutrophil cytoplasmic antibody titre.Lancet. 1990; 336: 709-711Abstract PubMed Scopus (401) Google Scholar. Persistent or intermittent ANCA positivity is an independent risk factor for relapse24Gaskin G. Savage C.O.S. Ryan J.J. et al.Anti-neutrophil cytoplasmic antibodies and disease activity during longterm follow up of 70 patients with systemic vasculitis.Nephrol Dial Transplant. 1991; 6: 689-694Crossref PubMed Scopus (119) Google Scholar,25Stegeman C.A. Cohen Tervaert J.W. Sluiter W.J. et al.Association of chronic nasal carriage of Staphylococcus aureus and higher relapse rates in Wegener's granulomatosis.Ann Intern Med. 1994; 120: 12-17Crossref PubMed Scopus (636) Google Scholar. Disease response to immunosuppression also supports an autoimmune process. Untreated, vasculitis carries a very poor prognosis. In early studies, Wegener's granulomatosis carried a 90% two-year mortality rate26Walton E.W. Giant-cell granuloma of the respiratory tract (Wegener's granulomatosis).Br Med J. 1958; 2: 265-270Crossref PubMed Scopus (610) Google Scholar. Mortality data for untreated microscopic polyangiitis are not available, because early studies contained heterogeneous vasculitides including polyarteritis nodosa. Cytotoxic therapy using cyclophosphamide together with corticosteroids induces remission in more than 90% of patients27Hoffman G.S. Kerr G.S. Leavitt R.Y. et al.Wegener's granulomatosis: an analysis of 158 patients.Ann Intern Med. 1992; 116: 488-498Crossref PubMed Scopus (2419) Google Scholar. In addition, a role for cell-mediated immunity in systemic vasculitis is suggested by beneficial responses of some patients to treatment with monoclonal anti-T-cell antibodies (CD4, CD52)28Jayne D.R.W. Immunotherapy for ANCA-associated systemic vasculitis.Clin Exp Nephrol. 1998; 112: 12-13Google Scholar. Clustering of autoimmune diseases provides further support for the autoimmune basis of each individual disorder. Concurrence of anti-GBM disease and ANCA-associated vasculitis is recognized29Kalluri R. Meyers K. Mogyorosi A. et al.Goodpasture syndrome involving overlap with Wegener's granulomatosis and anti-glomerular basement membrane disease.J Am Soc Nephrol. 1997; 8: 1795-1800PubMed Google Scholar. In one study, ANCA occurred in 38 of 100 patients with anti-GBM antibodies30Hellmark T. Niles J.L. Collins A.B. et al.Comparison of anti-GBM antibodies in sera with or without ANCA.J Am Soc Nephrol. 1997; 8: 376-385PubMed Google Scholar. At present this clustering cannot be explained through major histocompatibility complex (MHC) associations. Anti-GBM disease is strongly associated with HLA-DRB1*1501 (a DR2 allele), and this association lends strong support to the occurrence of cell-mediated immune responses, but MHC associations have been more difficult to discern for the ANCA-associated vasculitides. The literature contains conflicting reports of positive associations with, among others, HLA-B8, -DR2, and -DQw7; negative associations with DR13DR6; or a lack of association [reviewed in31Harper L. Savage C.O.S. Pathogenesis of ANCA-associ