Anti-neutrophil cytoplasmic antibodies in patients with rheumatoid arthritis: clinical, biological, and radiological correlations

Joint Bone Spine 71 (2004) 198–202 www.elsevier.com/locate/bonsoi

Original article

Anti-neutrophil cytoplasmic antibodies in patients with rheumatoid arthritis: clinical, biological, and radiological correlations Birkan Sonel Tur a,*, Nurben Süldür a, S qebnem Ataman a, Hüseyin Tutkak b, Mesut Birol Atay a, Nurs¸en Düzgün b a b

Department of Physical Medicine and Rehabilitation, Ankara University School of Medicine, Ankara, Turkey Department of Clinical Immunology and Rheumatology, Ankara University School of Medicine, Ankara, Turkey Received 14 January 2003; accepted 23 April 2003 Available online 26 September 2003

Abstract Objectives. – To determine the prevalence and the associations of anti-neutrophil cytoplasmic antibodies (ANCA) and subtypes with clinical, biological, and radiological findings in patients with rheumatoid arthritis (RA). Materials and methods. – This is a transversal study of 85 patients with RA (followed in Ibn-i Sina Hospital, Ankara University School of Medicine) with disease duration of 8.7 ± 6.4 years. Besides clinical, biological, and radiological disease activity parameters, ANCA and ANCA against myeloperoxidase (MPO) and proteinase 3 (PR3) were examined. Results. – The prevalence of ANCA, perinuclear ANCA (p-ANCA) and atypical ANCA (a-ANCA) were 18% (15/85 patients), 6% and 12%, respectively. Anti-MPO was found in six patients while anti-PR3 was not found. No significant association could be found between clinical, biological, and radiological disease activity assessments and ANCA (including indirect immunoflorescence subtypes). Similarly, ANCA were not associated with features suggestive of underlying vasculitis (noticed in 11/85 patients), and/or other extra-articular features. Conclusions. – Our results confirm that ANCA of various specificities (mainly a-ANCA) occur in a minority of RA. However, those ANCA were not associated with more severe RA, including the 6/85 patients positive for MPO (who were all free from vasculitis). The over-representation in RA sera of a-ANCA, as compared to p-ANCA, should be further studied. © 2003 Elsevier SAS. All rights reserved. Keywords: Rheumatoid arthritis; Disease activity; Anti-neutrophil cytoplasmic antibodies; Myeloperoxidase; Proteinase 3; Radiography

1. Introduction Rheumatoid arthritis (RA) is a chronic, systemic autoimmune disease characterized by inflammatory changes in joints, and sometimes extra-articular features, including vasculitis. Some autoantibodies (mainly rheumatoid factor and anti-citrulline (anti-CCP)) are specific for RA. Possibly, this does not apply for other autoantibodies much less specific for RA, like anti-nuclear antibodies (ANA) [1] and antineutrophil cytoplasmic antibodies (ANCA). ANCA are directed against constituents of neutrophil and mononuclear cell granules. ANCA have many target antigens such as lysosomal proteins, proteinase 3 (PR3), myeloperoxidase (MPO), cathepsin G, lactoferrin, elastase and lysozyme. * Corresponding author. Cetin Emec Bulvari, 2 Cadde, 78 Sokak, 19/3, 06450, Asagiovecler, Ankara, Turkey. E-mail address: [email protected] (B.S. Tur). © 2003 Elsevier SAS. All rights reserved. doi:10.1016/j.jbspin.2003.04.002

Generally, three different fluorescent patterns are observed: (1) a cytoplasmic pattern (c-ANCA), (2) a perinuclear pattern (p-ANCA), (3) apart from the well-known c-ANCA and p-ANCA, third staining pattern is atypical (a-ANCA). The main antigen producing the c-ANCA pattern is PR3, and the main antigen producing the p-ANCA pattern is MPO in azurophilic granules. Atypical ANCA have not been linked to a particular antigen [2]. In general, c-ANCA is strongly associated with vasculitic disorders and has been established as a useful diagnostic marker in primary systemic small vessel vasculitides (e.g., Wegener granulomatosis, microscopic polyangiitis, pauciimmune segmental necrotizing glomerulonephritis, and Churg–Strauss syndrome). However, the prognostic value of c-ANCA in those disorders is still controversial [2,3]. Recently p-ANCA have been also identified in several diseases such as primary and secondary vasculitis, collagen vascular diseases, inflammatory bowel diseases, rheumatoid arthritis,

B.S. Tur et al. / Joint Bone Spine 71 (2004) 198–202

and a few infections [4–9]. It was reported that ANCA occur especially in patients with long lasting RA and severe disease [9,10], and/or positive for rheumatoid factor (RF) and ANA [11]. A higher incidence of ANCA has been reported in RA complicated by rheumatoid vasculitis [12], but later studies could not confirm this finding [13–15], making the relationship between ANCA and vessel involvement in RA quite uncertain. Similarly, although most of previous studies found little if any association between ANCA (and/or ANCA titers) and RA activity (either at clinical, biological, and radiological levels), figures of prevalence for ANCA/subtypes in RA differed widely among reports [10–16]. Hence, we aimed to determine the prevalence of ANCA (and subtypes of ANCA) in RA to look for an association with clinical, biological, and radiological findings in unselected prospective transversal cohort of RA patients seen in our unit.

2. Patients and methods 2.1. Patient population This was a transversal study of 85 patients with RA (followed in Ibn-i Sina Hospital, Ankara University School of Medicine) with disease duration of 8.7 ± 6.4 years. Patients (67 females and 18 males, aged 50.5 ± 10.1 years) with RA fulfilling the 1987 ACR criteria [17], and 30 healthy controls (23 females and 7 males, 49.1 ± 9.8 years), whose ages and genders matched with the patients, were prospectively studied. First, patients were evaluated by routine clinical and laboratory measures, and then the following assessments were made. 2.2. Clinical assessments of disease activity Morning stiffness duration was given in minutes. Pain during the preceding week was recorded on a visual analog scale of 100 mm (0 = no pain, 100 = worst pain). Fatigue was assessed by a 5-point scale (1 = none, 5 = very severe). Patient’s and physician’s general health assessment were made on a visual analog scale of 10 cm (0 = best possible, 10 = worst possible). Disease activity was assessed according to the number of tender and swollen joints (total 28 joints) [18]. Then, for each patient, the disease activity score (DAS28) was also calculated from the number of tender and swollen joints (both by 28-joint-count), erythrocyte sedimentation rate (ESR), and patient’s general health assessment by visual analog scale [19]. The Turkish version of health assessment questionnaire (HAQ) was used to evaluate the disability in patients with RA [20]. The HAQ is a short, self-administered instrument that can be completed in 5–9 min. The disability dimension has eight categories of daily life: dressing and grooming, arising, eating, walking, hygiene, reach, grip and activities (each containing two or three items). For each item, a score of 0 (no difficulty), 1 (some difficulty), 2 (much difficulty or with

199

assistance), or 3 (unable to perform) is assigned. The highest score for each of the eight areas is summed and divided by 8 to yield, on a continuous scale, a functional disability index between 0 and 3 [21]. Presence of extra-articular manifestations such as Sjögren’s syndrome, rheumatoid nodules, rheumatoid vasculitis, pleuritis, nephropathy, anemia, Raynaud’s syndrome, Felty’s syndrome was recorded and vasculitis was diagnosed when one of the following symptoms was present: polyneuropathy/mononeuritis multiplex, cutaneous vasculitis, digital gangrene and visceral infarction, not attributable to any disease [22]. 2.3. Laboratory assessments of disease activity Hemoglobin level, leukocyte and platelet counts were recorded. ESR was measured according to the Westergren method (mm/1 h). CRP level was determined by nephelometry. All blood samples were stored at –20 °C until analyzed. 2.4. Determination of anti-neutrophil cytoplasmic antibodies Serum samples of patient and healthy control populations were tested for the presence of ANCA by combination of standard indirect immunoflorescence (IIF) and specific enzyme immunoassays (EIA) for PR3 and MPO [2,23]. 2.5. Indirect immunofluorescence ANCA were detected in the sera of patients by IIF using ethanol-fixed neutrophils as a substrate (The binding Site, England). Serum samples were diluted to 1:20 in PBS and incubated with fixed neutrophils on slides for 1 h at room temperature. The slides were washed three times and fluorescein isothiocyanate-conjugated monovalaent goat antibody against human IgG was added; after 30 min the slides were washed again. Fluorescence patterns were defined as c-ANCA when giving a diffuse granular cytoplasmic staining and as p-ANCA when giving a perinuclear or nuclear staining. Atypical ANCA (a-ANCA) gives most often a combination of cytoplasmic and perinuclear fluorescence, i.e., a diffuse homogeneous staining pattern (without central accentuation) associated with perinuclear fluorescence on IIF [2]. 2.6. Enzyme immunoassay Serum PR3-ANCA and MPO-ANCA levels were determined by EIA (Imtec, Germany). The company’s suggested normal value for this test is 0–10 U/ml. The IIF and EIA results were analyzed at the same time for all serum samples. 2.7. Other autoantibodies (RF, ANA, Anti-DNA) Rheumatoid factor was measured by nephelometric assay. ANA were detected on Hep-2 cell substrate by indirect im-

200

B.S. Tur et al. / Joint Bone Spine 71 (2004) 198–202

Table 1 Demographic, clinical, biological and radiographic characteristics of patients with RA and comparison between ANCA-positive and -negative groups Characteristics

Age (years) Gender (n, % female) Disease duration (years) Morning stiffness (minutes) Fatigue (0–4) Pain (0–100 VAS) Number of tender joints (0–28) Number of swollen joints (0–28) Physician’s GHA (0–10 VAS) Patient’s GHA (0–10 VAS) DAS28 Disability (0–3 HAQ) Extra-articular features (n, %) Vasculitis (n, %) Hemoglobin level (g/dl) Leukocyte count (×109/l) Platelet count (×109/l) ESR (mm/h) CRP (g/dl) Rheumatoid factor (IU/ml)

Study population (N = 85) Mean ± S.D. 50.5 ± 10.1 67, 79% 8.7 ± 6.4 61 ± 99 1.8 ± 1.2 49 ± 26 11 ± 9 2±3 3.9 ± 2.5 4.8 ± 2.6 4.5 ± 1.2 1.3 ± 0.9 49, 58% 11, 13% 12.1 ± 1.7 7.5 ± 2.8 318 ± 98 42 ± 24 32 ± 40 121.2 ± 156.8

Median 50 7.5 30 2 50 9.5 1 3.2 5 4.4 1.2

12.1 7.2 291 34 18 48.4

ANCA (–) group (n = 70)

ANCA (+) group (n = 15)

median 51.5 ± 13.1 ** 76% 8.3 ± 5.7 ** 30 2 50 10 1 3.5 5.1 4.6 1.3 39, 57% 9, 12.8% 12.1 7.4 314 40 19 57.9

median 50.0 ± 17.4 ** 80% 9.0 ± 5.2 ** 45 2 48 9 2 3 4.5 4.4 1.0 10, 67% 2, 13.3% 12.2 6.2 279 30 16 51.9

P* NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS NS

NS indicates no significant difference. Physician’s GHA, physician’s general health assessment; Patient’s GHA, patient’s general health assessment. * P values are from Chi-square or Mann–Whitney U Tests, where appropriate. ** Result is given as mean ± S.D.

munofluorescence technique. Fluorescence intensity was scored from 0 to ++++; staining patterns were described as homogeneous, speckled, nucleolar, and linear (peripheral). Anti-ds-DNA antibodies were measured by ELISA according to the manufacturer’s instructions. Serum immunoglobulin (IgG, IgA, IgM), and complement (C3,C4) levels were also measured by nephelometric assay. 2.8. Radiographic assessment Radiographic examinations of hands and feet were evaluated according to the modified sharp method [24] by one physician. The maximum erosion score of all joints in both hands is 160 and in both feet 120 (total erosions of hands and feet = 280); and the maximum score for joint space narrowing in all joints of both hands is 120 and in both feet 48 (total joint space narrowing of hands and feet = 168). 2.9. Statistical analyses The data were analyzed by non-parametric statistics. Correlations between the presence of ANCA and clinical, laboratory and radiological findings were assessed using ChiSquare and Fischer’s Exact Tests when appropriate for categorical variables, and Mann–Whitney Rank Sum Test for continuous variables. P < 0.05 was considered to be significant.

3. Results There was no significant difference with respect to the age and gender between patients with RA and healthy controls. None of the healthy controls had positive IIF staining for ANCA, MPO and PR3. The mean disease duration was 8.7 ± 6.4 years. Seventy seven percent were seropositive, and 49 patients had extraarticular manifestations, including 11/85 classified as ‘vasculitis’ according to the criteria described in Section 2. The median morning stiffness duration of all patients was 30 min and the median fatigue score was 2. The mean DAS28 score was 4.4 ± 1.2. ANA was positive in 13 patients with RA. All ANA positive serum specimens gave homogeneous pattern on IIF. Anti-DNA was negative in all patients. The median total scores of erosion and joint space narrowing of hands and feet were 17.5 and 21, respectively. The patients were moderately active with respect to clinical, biological, and radiographic values. Demographic data and disease characteristics of the patients with RA are shown in Table 1. Sera from 15 (18%) of the 85 patients were positive for ANCA; no patient had c-ANCA, five (6%) had p-ANCA, and 10 (12 %) had a-ANCA. Anti-MPO antibodies were detected in six (7%) patients with RA by EIA, but neither had evidence of vasculitis. Anti-PR3 antibodies were not detected in the patients with RA. Three of six patients with MPO-ANCA positivity showed p-ANCA pattern, two had ANCA negativity and one had a-ANCA. None of patient had PR3-ANCA. Data are

B.S. Tur et al. / Joint Bone Spine 71 (2004) 198–202

201

Table 2 The characteristics and specificies of ANCA, its association with ANA and vasculitis ANCA positive ANA positive Anti-MPO Anti-PR3 Vasculitis

n 15 13 6 0 11

c-ANCA 0 0 0 0 0

p-ANCA 5 2 3 0 0

shown in Table 2. There was no association between ANCA positivity and clinical, biological, and radiographic variables (P > 0.05) (Table 1).

4. Discussion Although first reports (who found an association between ANCA positivity and either RA activity and/or rheumatoid vasculitis [9]) had claimed that all RA patients should be tested for ANCA, owing to prognosis value of these autoantibodies, this assumption has been seriously challenged later on. In fact, the prevalence of ANCA in RA then appeared to be highly variable (from 7% to 94%) [8,10–16,25,26], and their correlation with RA activity/severity weak or null, notwithstanding the lack of association of ANCA positivity/titers with rheumatoid vasculitis [13–15]. Our results are in line with most previous reports in this respect. Indeed we found a prevalence of 18% (0% c-ANCA, 6% p-ANCA, and 12% a-ANCA) of positive sera whose positivity was not associated either with RA activity or extraarticular features. Most of the studies have reported well-known patterns of ANCA, c-and p-ANCA [13,26–27]. Our present study and a few previous studies have evaluated a-ANCA pattern in addition to c-ANCA and p-ANCA patterns [10–13]. As IIF evaluation remains rather subjective, recent guidelines for ANCA testing have been published [2], to allow a better standardization of the methodology and to minimize discrepancies between laboratories when testing a same set of sera. Technical limitations might indeed account for many of the variations previously noticed between reports [8,13,15,25,26]. Another source of confusion might have been the mixture of results for a-ANCA and p-ANCA in a single category. Perhaps, results could have also been more consistent if separate data for a-ANCA and p-ANCA had been provided, like we did. In fact, it can be difficult to distinguish true p-ANCA reactivity from anti-nuclear activity. To better differentiate p-ANCA and ANA, sera giving p-ANCA are sometimes re-tested both on formaldehyde-fixed PMN and Hep-2 cells [26]. However, we did not use this approach, since the interpretation of all staining patterns observed with the different assay is not always decisive [15]. Csernok and Gross [8] concluded that p-ANCA in RA may be a marker for a more aggressive course of disease in respect to serological parameters and extra-articular manifestations including rheumatoid vasculitis and lung involve-

a-ANCA 10 6 1 0 2

ANCA negative – 5 2 0 9

ment. One study reported that p-ANCA is an early marker of progressive erosive disease in RA and is detectable early in the disease course [11]. In Coremans’ study [12], ANCA were found in 43% of 49 patients with rheumatoid vasculitis, as compared to 36% of RA without vasculitis. Moreover, the authors underlined a striking increase of anti-lactoferrin antibodies in the former group (rate 11 times higher than in uncomplicated RA). However, most other reports did not make such observations, and we were also unable to find a significant correlation between disease activity and ANCA positivity. In the present study, anti-MPO activity has been demonstrated in 6/85 RA patients (7%) (3 p-ANCA, 1 aANCA, and 2 sera negative on IIF). This figure is higher than in previous reports by Cohen et al. [28] who found none out of 38 RA sera, and Schönermarck et al. [26] who found only one MPO-positive sera when testing a very large number of RA patients (n = 1412). However, there was no trend for any association between positivity for MPO and the presence of clinical features suggestive of underlying rheumatoid vasculitis in our patients (as none of our six patients positive for MPO belong to the group of 11 with possible vasculitis). Similarly, we were unable to find an association between ANCA positivity by IIF and either long disease duration or features suggestive of vasculitis in our 85 patients. Mustila et al. [27] found a correlation between p-ANCA and both RA duration/severity and underlying nephropathy (either clinically suspected or histological proven), but renal involvement was absent in all our patients, which precluded the search for any correlation with ANCA. Hence, although 6 of our 85 patients were positive for MPO by ELISA, our results suggest that ANCA positivity by IIF is lower in RA than generally reported. Conversely, our results support the conclusions of most previous studies that positivity for ANCA and/or MPO is not associated with RA severity (either for articular or extra-articular features). The over-representation of a-ANCA over p-ANCA and c-ANCA in RA sera might however deserve further studies to be elucidated.

References [1]

Baum J, Ziff M. Laboratory findings in rheumatoid arthritis. 11th ed. In: McCarty DJ, editor. Arthritis and allied conditions. Philadelphia: Lea and Febiger; 1989. p. 744–61.

202 [2]

[3]

[4]

[5]

[6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

B.S. Tur et al. / Joint Bone Spine 71 (2004) 198–202 Savige J, Gillis D, Benson E, Davies D, Esnault V, Falk RJ, et al. International consensus statement on testing and reporting of antineutrophil cytoplasmic antibodies (ANCA). Am J Clin Pathol 1999;111:507–13. Cohen Tervaert JW, Kallenberg CG. The role of autoimmunity to myeloid lysosomal ensymes in the pathogenesis of vasculitis. In: Hansson G, Libby P, editors. Immune functions of the vessel wall. Amsterdam: Harwood Academic Publishers; 1996. p. 99–120. Van der Woude FJ, Rasmussen N, Lobatto S, Wiik A, Permin H, van Es LA, et al. Autoantibodies against neutrophil and monocytes: tool for diagnosis and marker of disease activity in Wegener’s granulomatosis. Lancet 1985;1:425–9. Falk RJ, Jenette JC. Anti-neutrophil cytoplasmic antibodies with specifity for myeloperoxidase in patients with systemic vasculitis and idiopathic necrotizing and cresentic glomerulonephritis. N Engl J Med 1988;318:1651–7. Saxon A, Shanahan F, Landers C, Ganz T, Targan S. A distinct subset of antineutrophil cytoplasmic antibodies is associated with inflammatory bowel disease. J Allergy Clin Immunol 1990;86:202–10. Gross WL, Schmitt WH, Csernok E. Antineutrophil cytoplasmic autoantibody-associated diseases: a rheumatologist’s perspective. Am J Kidney Dis 1991;18:175–9. Csernok E, Gross WL. Antineutrophilic cytoplasmic antibodies (ANCA) in inflammatory rheumatic diseases: immunodiagnostic and immunopathogenetic aspects. Z Rheumatol 1995;54:26–38. Mulder AH, Horst G, van Leeween MA, Limburg PC, Kallenberg CGM. Antineutrophil cytoplasmic antibodies in rheumatoid arthritis: characterization and clinical correlations. Arthritis Rheum 1993;36:1054–60. de Bandt M, Meyer O, Haim T, Kahn MF. Antineutrophil cytoplasmic antibodies in rheumatoid arthritis patients. Br J Rheumatol 1996;35: 38–43. Mustila A, Paimela L, Leirisalo-Repo M, Huhtala H, Miettinen A. Antineutrophil cytoplasmic antibodies in patients with early rheumatoid arthritis. An early marker of progressive erosive disease. Arthritis Rheum 2000;43:1371–7. Coreman IEM, Hagen EC, Daha MR, van der Woude FJ, van der Voort EA, Kleijburg-van der Keur C, et al. Antilactoferrin antibodies in patients with rheumatoid arthritis are associated with vasculitis. Arthritis Rheum 1992;35:1466–75. Röther E, Schochat T, Peter HH. Antineutrophil cytoplasmic antibodies in rheumatoid arthritis: prospective study. Rheumatol Int 1996;15: 231–7. Braun MG, Csernok E, Schmitt WH, Gross WL. Incidence, target antigens and clinical implications of antineutrophil cytoplasmic antibodies in rheumatoid arthritis. J Rheumatol 1996;23:826–30.

[15] Vittecoq O, Jounen-Beades F, Krzanowska K, Bichon-Tauvel I, Menard JF, Daragon A, et al. Prospective evaluation of the frequency and clinical significance of antineutrophil cytoplasmic and anticardiolipin antibodies in community cases of patients with rheumatoid arthritis. Rheumatol (Oxford) 2000;39:481–9. [16] Savige JA, Gallcchio MC, Stockman A, Cunningham TJ, Rowley MJ, Georgiou T, et al. Antineutrophil cytoplasmic antibodies in rheumatoid arthritis. Clin Exp Immunol 1991;86:92–8. [17] Arnett FC, Edworthy SM, Bloch DA, McShare DJ, Fries JF, Cooper NS, et al. American Rheumatism Association 1987 revised for the classification of rheumatoid arthritis. Arthritis Rheum 1988;31:315– 24. [18] Fuchs HA, Pincus T. Reduced joint counts in controlled clinical trials in rheumatoid arthritis. Arthritis Rheum 1994;37:470–5. [19] Prevoo MLL, van‘t Hof MA, Kuper HH, van Leeuwen MA, van de Putte LBA, van Riel PLCM. Modified disease activity scores that include twenty-eight-joint counts. Arthritis Rheum 1995;38:44–8. [20] Kucukdeveci AA, Sahin H, Ataman S, Griffiths B, Tennant A. Issues in cross-cultural validity: example from the adaptation reliability and validity testing of a Turkish version of the Stanford Health Assessment Questionnaire (HAQ). Arthritis Care Res (in press). [21] Fries JF, Spitz P, Kraines RG, Holman HR. Measurement of patient outcome in arthritis. Arthritis Rheum 1980;23:137–45. [22] Scott DGI, Bacon PA, Tribe CR. Systemic rheumatoid vasculitis: a clinical and laboratory study of 50 cases. Medicine (Baltimore) 1981; 60:288–97. [23] Wiik A, Rasmussen N, Wieslander J. Methods to detect autoantibodies to neutrophilic granulocytes. In: van Venrooij WJ, Maini RN, editors. Manual of biological markers of disease. Doordrecht: Kluwer Academic Publishers; 1993. p. A9:1–A9:14. [24] van der Heijde D. How to read radiographs according to the Sharp/van der Heijde method. J Rheumatol 1999;26:743–5. [25] Wiik AS. Clinical use of serological tests for antineutrophil cytoplasmic antibodies. What do the studies say? Rheum Dis Clin North Am 2001;27:799–813. [26] Schönermarck U, Lamprercht P, Csernok E, Gross WL. Prevalence and spectrum of rheumatic diseases associated with proteinase 3-antineutrophil cytoplasmic antibodies (ANCA) and myeloperoxidaseANCA. Rheumatology 2001;40:178–84. [27] Mustila A, Korpela M, Mustonen J, Helin H, Huhtala H, Soppi E, et al. Perinuclear antineutrophil cytoplasmic antibody in rheumatoid arthritis. A marker of severe disease with associated nephropathy. Arthritis Rheum 1997;40:710–7. [28] Cohen Tervaert JW, Goldschmeding R, Elema JD, Limburg PC, van der Giessen M, Huitema MG, et al. Association of autoantibodies to myeloperoxidase with different forms of vasculitis. Arthritis Rheum 1990;33:1264–72.