Repeat protocol renal biopsy in ANCA-associated renal vasculitis

NDT Advance Access published February 26, 2014 Nephrol Dial Transplant (2014) 0: 1 5 doi: 10.1093/ndt/gfu042 Original Article Repeat protocol renal biopsy in ANCA-associated renal vasculitis Zdenka Hruskova 1, Eva Honsova 2, Annelies E. Berden 3, Ivan Rychlik 4, Vera Lanska 5, Jiri Zabka 6, Ingeborg M. Bajema 3 and Vladimir Tesar 1 1 Department of Nephrology, General University Hospital and First Faculty of Medicine, Charles University in Prague, Prague, Czech Republic, 2 Department of Pathology, Institute for Clinical and Experimental Medicine, Prague, Czech Republic, 3 Department of Pathology, Leiden University Medical Center, Leiden, The Netherlands, 4 2nd Department of Internal Medicine, Third Faculty of Medicine, Charles University in Prague, Prague, Czech Republic, 5 Statistical Unit, Institute for Clinical and Experimental Medicine, Prague, Czech Republic and 6 1st Department of Internal Medicine, Third Faculty of Medicine, Charles University in Prague, Prague, Czech Republic Correspondence and offprint requests to: Zdenka Hruskova; E-mail: zd.hruskova@gmail.com ABSTRACT Background. Histopathological lesions in renal biopsy (RB) at presentation of ANCA-associated vasculitis (AAV) have been described in depth but repeat protocolized renal biopsies are seldomly performed in AAV. In this study, we present a group of AAV patients with repeat protocolized biopsies, and we evaluate their clinical significance. Methods. A total of 17 consecutive patients diagnosed between 1991 and 1995 with AAV and renal involvement confirmed by biopsy at presentation in a single center underwent a protocol planned rebiopsy in remission after a median of 13 months (range 11 28) from diagnosis. Biopsies were assessed by two independent pathologists, blinded to patient data. Clinical data were collected retrospectively. Results. Patients were followed-up for a median of 189 months from diagnosis. Renal relapse was observed in eight patients (47.1%), seven patients died, three patients reached end-stage renal failure. There was a significant decrease in the percentage of acute lesions (cellular crescents, fibrinoid necrosis, P < 0.001) and a significant increase in chronic lesions (glomerulosclerosis, interstitial fibrosis, P 0.01) in the repeat RB compared with the first RB. This resulted in a class change over the biopsies within most patients. The percentage of normal glomeruli in the first biopsy positively correlated with estimated GFR at the end of follow-up (r s = 0.509, P = 0.05). Conclusions. This is the first study on protocolized repeat biopsies in AAV, giving insight into disease activity under immunosuppressive treatment. Apparently, many AAV patients have grumbling disease with ongoing activity, eventually leading to an increased amount of chronic lesions. Keywords: ANCA, outcome, rebiopsy, renal biopsy, vasculitis INTRODUCTION Renal involvement is known to be an important prognostic factor in patients with anti-neutrophil cytoplasmic antibodies (ANCA) associated systemic vasculitides (AAV) [1], particularly granulomatosis with polyangiitis (GPA, formerly Wegener s) and microscopic polyangiitis (MPA). Even in patients with positive ANCA and typical clinical features, renal biopsy (RB) is widely recommended [2] not only in the diagnostic work-up but also to help to predict the renal outcome of these patients. Untreated, patients with renal vasculitis follow an unfavourable disease course towards end-stage renal disease (ESRD), with high mortality. Conventional combined immunosuppressive treatment consisting of cyclophosphamide and corticosteroids helps to achieve remission in most patients with AAV but is associated with substantial unwanted sideeffects. To avoid or at least reduce drug-related morbidity and mortality, there is a long-standing need to individualize immunosuppressive treatment. Acute renal lesions observed in AAV typically include extracapillary proliferation (crescents) and segmental fibrinoid necrosis. However, chronic lesions, such as sclerosis or fibrosis, The Author 2014. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. 1

ORIGINAL ARTICLE may also be found in diagnostic RB performed at first disease presentation. An international histopathologic classification of ANCA-associated glomerulonephritis was recently developed to distinguish the levels of severity [3], which may eventually help to guide therapy in individual patients. The association between histopathologic findings and longterm renal outcome of AAV was assessed previously in a number of studies [4 11], most of which demonstrated that RB findings are of predictive value, independent of baseline renal function. To the best of our knowledge, only very limited information is available on data concerning follow-up (FU) biopsies in AAV [8 11] performed outside clinical indication (suspected renal relapse or unexplained serum creatinine increase), i.e. protocolized RB. In this study, we retrospectively evaluated the clinical significance of repeat renal biopsies (rerb) in AAV. The rerb were performed at a previously specified time-point after the diagnosis, irrespective of the clinical findings. SUBJECTS AND METHODS Patients In 1991 95, a total of 17 consecutive patients newly diagnosed with ANCA-associated renal vasculitis in a single centre willing to participate in the study were included. The study conformed to generally applied ethical standards and was performed in accordance with the Declaration of Helsinki. Prior to enrolment, all patients gave informed consent to participate. Their baseline characteristics are displayed in Table 1. Diagnoses of GPA and MPA were established retrospectively using the European Medicines Agency algorithm [12]. After the first RB, all patients were treated with combined immunosuppressive treatment common at that time, consisting of oral continuous cyclophosphamide (CYC) and corticosteroids for 1 year; nine also received additional therapy with plasma exchange at entry. At the time of rebiopsy, all patients Table 1. Baseline patient characteristics (no. of patients = 17) Baseline characteristics Data Age (years) 49 (22 72) Female gender 2 (11.8%) Diagnosis GPA 12 (70.6%) MPA 5 (29.4%) ANCA-IIF result (ever) canca 14 (82.4%) panca 3 (17.6%) Organ involvement Kidney 17 (100%) ENT 6 (35.3%) Lungs 8 (47.1%) Serum creatinine (µmol/l) 281 (85 800) egfr (ml/min) a 21 (6 95) No. of patients on haemodialysis 4 (23.5%) Proteinuria (g/24 h) 2.0 (0.5 6.3) GPA, granulomatosis with polyangiitis; MPA, microscopic polyangiitis, including renal-limited forms. Data presented as median (range) or no. (%). a MDRD equation used to calculate egfr. were on CYC and corticosteroids. After the rebiopsy, combined immunosuppressive therapy was stopped in 12 patients and five were switched to azathioprine. Renal biopsy The first RB was performed at disease presentation. The rerb was planned to be performed after 1 year ± 1 month of treatment, i.e. at the time of planned withdrawal of cyclophosphamide. Eventually, ReRB was performed after a median time of 13 months (range 11 28), with only one patient biopsied after >16 months. Renal tissue taken by core biopsy was divided into three parts for immunofluorescence, light microscopy, and electron microscopy. Staining for IgG, IgA, IgM, C3, kappa and lambda light chains was routinely performed. The second part was embedded in paraffin, cut in 3-μm sections, and stained according to the protocol of our laboratory at that time haematoxylin and eosin (HE), periodic acid-schiff, trichrom, periodic acid silver-methenamine and HE with elastine. During reassessment of the samples, immunohistochemical detection of CD3- and CD20-positive lymphocytes was performed. Fifteen samples from the initial biopsies and 16 samples from the rebiopsies were available for analysis. All samples were assessed by two independent pathologists blinded to patient data and to original pathology reports. The samples were scored according to a previously described standardized protocol [13] and classified according to the recently developed classification [3]. Statistical analysis Due to non-normal distribution and small sample size, all continuous parameters are presented as median (range). The changes between first RB and rerb were tested by Wilcoxon signed-rank test or McNemar s χ 2 test. The relationship between clinical values and histopathologic findings were assessed using Spearman correlation coefficient. A P-value 0.05 was considered statistically significant in all cases. RESULTS Clinical data are displayed in Table 2 and histopathologic data in Table 3. At the time of rerb, the activity of vasculitis was assessed as complete remission in all patients even though mild erythrocyturia persisted in four (23.5%, Table 2). All patients on dialysis at entry recovered independent renal function and Table 2. Comparison of clinical renal parameters at the time of first and rerb Renal parameters 1st biopsy Re-biopsy P-value S-creatinine (μmol/l) 281 (85 800) 142 (76 260) <0.001 egfr a (ml/min) 21 (6 95) 46 (23 107) <0.001 HD (yes) 4 (23.5%) 0 (0%) <0.05 PRU (g/24 h) 2.0 (0.5 6.3) 1.5 (0 6.7) NS (0.055) eryu (yes) 17 (100%) 4 (23.5%) <0.05 HD, haemodialysis; PRU, proteinuria; eryu, erythrocyturia. (%). Data presented as median (range) or as No (%). a MDRD equation used to calculate egfr. Z. Hruskova et al. 2

there was a significant improvement in serum creatinine levels and estimated glomerular filtration rate compared with baseline (P < 0.001, Table 2). Patients were followed-up for a median of 189 months from diagnosis (range 13 232). A total of eight patients (7 GPA, 1 MPA) underwent renal relapse (47.1%) at a median time of 34 (range 0.5 99) months after the rebiopsy. Seven patients (41.2%) died during the FU period at a median time of 77 (range 1 162) months after the rebiopsy; three patients (17.6%) reached renal failure (two of them died). Median serum creatinine at the end of FU was 177 µmol/l (range 63 349) in non-dialysed patients. Histopathological evaluation essentially showed a significant decrease in acute lesions and increase in chronic lesions (Table 3, Figure 1). However, there was no significant difference in the percentage of normal glomeruli between the biopsies (Table 3). Table 4 reveals that 11 patients showed progression in the rebiopsy by the histopathological classification, mostly by shifting from the crescentic to the mixed class. Two patients Table 3. Comparison of histopathologic parameters between the first and rerb significant differences Histopathologic findings (in % of total gli) 1st biopsy Re-biopsy P-value Normal glomeruli 25.0 (0 75) 26.8 (0 53.8) NS (0.90) Segm. cellular crescents 14.3 (4.7 71.4) 0 (0 6.5) <0.001 Circumf. cellular crescents 15.8 (0 88.1) 0 (0 21.1) <0.001 Total cellular crescents 52.2 (5.0 93.9) 2.0 (0 27.0) <0.001 Fibrinoid necrosis 23.2 (7.8 47.1) 0 (0 15.1) <0.001 Segm. fibrous crescents 2.6 (0 18.3) 13.9 (0 45.8) 0.01 Circumf. fibrous crescents 0 (0 25) 12.5 (0 34.3) 0.05 Total fibrous crescents 3.8 (0 38.8) 25.4 (0 51.3) 0.002 Global glomerulosclerosis 6.0 (0 46) 32.3 (0 59.5) 0.007 Segmental 3.2 (0 25) 17.2 (0 28.6) 0.03 glomerulosclerosis Total no. of sclerotic 9.0 (0 64.5) 52.5 (0 70) 0.001 glomeruli Oedema (yes) 12/15 (80%) 3/16 (18.8%) 0.01 Interstitial inflammation a 1.5 (0.5 3) 1 (0 2) 0.04 Interstitial fibrosis b 0.5 (0 2) 1.5 (0 2) 0.01 Data presented as median (range) or no. (%). a Classified as 0 1 2 3. b Classified as 0 1 2 according to the standardized protocol [13]. remained in the same class and only one patient improved (Table 4). Of note, this patient had only three glomeruli captured in the rebiopsy, which might have influenced the result even though, on the other hand, his renal function remained stable during FU. Interestingly, there was a significant decrease in the number of CD20+ cells in the rerb [median 2.3 (range 0 120) CD20+ cells/mm 2 ] compared with the biopsy at presentation [median 39.3 (range 0 192), P = 0.03], whereas the number of CD3+ cells did not significantly decrease [median 469.6 (range 28.9 1907.6) in the first biopsy and 102.5 (15 1340.0) in the rebiopsy, P = 0.08], see also Figure. 2. A positive correlation between age and the percentage of sclerotic glomeruli (r s = 0.588, P < 0.05), tubular atrophy (r s = 0.712, P < 0.01) and interstitial fibrosis (r s = 0.726, P < 0.01) in the first biopsy was observed in this study. On the other hand, there was a negative correlation between age and the percentage of cellular crescents in the first biopsy (r s = 0.590, P < 0.05). Estimated glomerular filtration rate at the end of FU positively correlated with the percentage of normal glomeruli in the first biopsy (r s = 0.509, P = 0.05) and negatively correlated with Table 4. Comparison of histopathologic classes [3] in the biopsies Patient no. 1st biopsy Re-biopsy Outcome (time to the event a ) 1 N/A Mixed Surviving 2 Crescentic Mixed Surviving 3 N/A Sclerotic Surviving 4 Crescentic Mixed Surviving 5 Focal Focal Surviving 6 Crescentic Mixed Surviving 7 Crescentic Mixed Death (162) 8 Mixed Mixed Death (48) 9 Crescentic Sclerotic Death + ESRD (133 + 14) 10 Mixed N/A Death (1) 11 Crescentic Mixed Death (2) 12 Crescentic Mixed Surviving 13 Mixed Sclerotic ESRD (167) 14 Crescentic Mixed Surviving 15 Mixed Focal Death (77) 16 Focal Sclerotic Surviving 17 Crescentic Mixed Death + ESRD (99 + 81) a In months after the rebiopsy. ORIGINAL ARTICLE FIGURE 1: (A) Necrosis and cellular crescent in RB at presentation. (B) Segmental glomerular sclerotic lesion with adhesion to fibrous crescent in rebiopsy. 3 Repeat renal biopsy in AAV

ORIGINAL ARTICLE FIGURE 2: (A) CD20+ cells in the first biopsy. (B) CD20+ cells in the rebiopsy. (C) CD3+ cells in the first biopsy. (D) CD3+ cells in the rebiopsy. the percentage of sclerotic glomeruli both in the first biopsy and rerb (r s = 0.586 and r s = 0.545, respectively, P < 0.05). DISCUSSION ReRBs are widely used nowadays in the renal transplantation setting [14], but are rarely performed in native renal diseases. For patients with renal transplants, rerb have proven to reveal pathologic findings not yet translated into the clinical signs of renal dysfunctions [14]. Moreover, they now influence the strategy of immunosuppressive treatment. In the present study, we present the unique data of patients with AAV who underwent a protocolized repeat RB 1 year after disease presentation. Our results clearly show almost, but not complete disappearance of acute lesions such as cellular crescents and fibrinoid necrosis, with varying amounts of remaining fibrous crescents and global sclerosis. Evaluated by the recent classification for ANCA-associated glomerulonephritis, most patients moved one level up in the classification hierarchy, most typically from the crescentic to the mixed class, suggesting that many of the patients have grumbling disease with ongoing activity despite treatment. Because of the uniqueness of this study, it can hardly be compared with previously published data on repeat biopsies in AAV, since these studies reported on repeat biopsies on clinical indications [8 11], and included a total of 163 rebiopsies performed after varying time intervals from the first biopsy (mean/median values ranging from 397 days to 45 months) [8 11]. Nevertheless, even the rebiopsies performed for a suspicion of relapse [8 11], uniformly showed a decrease in activity and progression in chronicity scores (or an increase in global glomerulosclerosis) compared with the initial biopsies. In the study of Neumann et al. [8], eight patients with an increase in serum creatinine levels with the absence of clinically active disease were included and rebiopsied after a mean of 16 months, thus most closely resembling our group of patients. In line with their study, we found hardly any active changes in rebiopsies but, on the contrary, we were not able to confirm the increase in the percentage of normal glomeruli. Hauer et al. [11] previously performed a study on repeat biopsies in AAV taken on clinical indication in 31 patients. The mean percentage of normal glomeruli in the RB did not change over time (29% in the initial and 30% in the FU biopsy). The mean percentage of glomeruli with crescents, however, significantly decreased from 57 to 30% (P < 0.001), whereas the percentage of glomerulosclerosis significantly increased from 12 to 39% (P < 0.001). Their findings seem to parallel our findings, with the amount of normal glomeruli remaining relatively constant over time, and some of the active lesions being transformed to a chronic phenotype. That a relatively constant proportion of glomeruli seems to be preserved during AAV, is an interesting finding, particularly because normal glomeruli in many studies have been shown to predict for renal outcome, which is also the case in our study. The precise role of the participation of B cells and T cells in the pathogenesis and course of AAV is not clear but available data support the important role of both of these cell populations. It was shown that the number of activated B cells in the circulation in contrast to the number of activated T cells, correlates with disease activity scores of AAV [15]. Moreover, recent data on the effective treatment through rituximab Z. Hruskova et al. 4

clearly indicate an important role for B-cell depletion in the therapy for AAV. None of our patients received targeted B-cell depletion therapy. Nevertheless, the remission state reflected by the repeat biopsy was characterized by a significant decrease in B-cells. Despite immunosuppressive therapy directed also towards T-cells, the number of CD3 positive T-cells in repeat biopsies was not significantly decreased in the repeat biopsy. This indicates that at the tissue level, processes take place that cannot be captured by measurements for cell subpopulations in blood samples. Recently, it was suggested that under rituximab therapy, B-cells may be rescued in tissue [16], which may explain why relapses may occur whereas peripheral B-cell counts are low. The data from our study provide important baseline-values for patients treated with conventional T-celldirected therapy. In order to begin to study these processes more carefully, larger studies than ours with protocolized repeat biopsies are called for. Our study was limited by its retrospective nature and by a relatively small patient group. Nevertheless, the uniqueness of this small group of patients with AAV and rerb has revealed a number of unknown findings that provide enough support to perform prospective, larger patient studies with protocolized repeat biopsies in the near future. Only then can we begin to unravel the evolvement of pathologic findings in ANCAassociated glomerulonephritis. ACKNOWLEDGEMENTS The study was supported by the research project PRVOUK- P25/LF1/2. CONFLICT OF INTEREST STATEMENT The authors declare no conflict of interests. REFERENCES 1. Mukhtyar C, Flossmann O, Hellmich B et al. European Vasculitis Study Group (EUVAS). Outcomes from studies of antineutrophil cytoplasm antibody associated vasculitis: a systematic review by the European League Against Rheumatism systemic vasculitis task force. Ann Rheum Dis 2008; 67: 1004 1010 2. Mukhtyar C, Guillevin L, Cid MC et al. European Vasculitis Study Group. EULAR recommendations for the management of primary small and medium vessel vasculitis. Ann Rheum Dis 2009; 68: 310 317 3. Berden AE, Ferrario F, Hagen EC et al. Histopathologic classification of ANCA-associated glomerulonephritis. J Am Soc Nephrol 2010; 21: 1628 1636 4. Bajema IM, Hagen EC, Hermans J et al. Kidney biopsy as a predictor for renal outcome in ANCA-associated necrotizing glomerulonephritis. Kidney Int 1999; 56: 1751 1758 5. de Lind van Wijngaarden RA, Hauer HA, Wolterbeek R et al. Clinical and histologic determinants of renal outcome in ANCA-associated vasculitis: a prospective analysis of 100 patients with severe renal involvement. J Am Soc Nephrol 2006; 17: 2264 2274 6. Hauer HA, Bajema IM, Van Houwelingen HC et al. European Vasculitis Study Group (EUVAS). Determinants of outcome in ANCA-associated glomerulonephritis: a prospective clinico-histopathological analysis of 96 patients. Kidney Int 2002; 62: 1732 1742 7. Vergunst CE, van Gurp E, Hagen EC et al. EC/BCR Project for ANCA- Assay Standardisation. An index for renal outcome in ANCA-associated glomerulonephritis. Am J Kidney Dis 2003; 41: 532 538 8. Neumann I, Kain R, Regele H et al. Histological and clinical predictors of early and late renal outcome in ANCA-associated vasculitis. Nephrol Dial Transplant 2005; 20: 96 104 9. Aasarod K, Bostad L, Hammerstrom J et al. Renal histopathology and clinical course in 94 patients with Wegener s granulomatosis. Nephrol Dial Transplant 2001; 16: 953 960 10. Day CJ, Howie AJ, Nightingale P et al. Prediction of ESRD in pauciimmune necrotizing glomerulonephritis: quantitative histomorphometric assessment and serum creatinine. Am J Kidney Dis 2010; 55: 250 258 11. Hauer HA, Bajema IM, Hagen EC et al. Long-term renal injury in ANCAassociated vasculitis: an analysis of 31 patients with follow-up biopsies. Nephrol Dial Transplant 2002; 17: 587 596 12. Watts R, Lane S, Hanslik T et al. Development and validation of a consensus methodology for the classification of the ANCA-associated vasculitides and polyarteritis nodosa for epidemiological studies. Ann Rheum Dis 2007; 66: 222 227 13. Bajema IM, Hagen EC, Hansen BE et al. The renal histopathology in systemic vasculitis: an international survey study of inter- and intra-observer agreement. Nephrol Dial Transplant 1996; 11: 1989 1995 14. Buchmann TN, Wolff T, Bachmann A et al. Repeat true surveillance biopsies in kidney transplantation. Transplantation 2012; 93: 908 913 15. Popa ER, Stegeman CA, Bos NA et al. Differential B- and T-cell activation in Wegener s granulomatosis. J Allergy Clin Immunol 1999; 103: 885 894 16. Holden NJ, Williams JM, Morgan MD et al. ANCA-stimulated neutrophils release BLyS and promote B cell survival: a clinically relevant cellular process. Ann Rheum Dis 2011; 70: 2229 2233 Received for publication: 14.1.2014; Accepted in revised form: 30.1.2014 ORIGINAL ARTICLE 5 Repeat renal biopsy in AAV