Rituximab treatment for fibrillary glomerulonephritis

Transcription

1 Nephrol Dial Transplant (2014) 29: doi: /ndt/gfu189 Advance Access publication 27 May 2014 Rituximab treatment for fibrillary glomerulonephritis Jonathan Hogan, Michaela Restivo, Pietro A. Canetta, Leal C. Herlitz, Jai Radhakrishnan, Gerald B. Appel and Andrew S. Bomback Division of Nephrology, Department of Medicine, Columbia University Medical Center, New York, NY, USA Correspondence and offprint requests to: Jonathan Hogan; ABSTRACT Background. Approximately 50% of patients with fibrillary glomerulonephritis (GN) progress to end-stage disease () within 2 years of diagnosis, and no standard therapy exists. The data on rituximab therapy for fibrillary GN are limited and have inconsistent outcomes. Here, we report the largest case series to date using rituximab for fibrillary GN. Methods. Retrospective chart reviews were conducted on 12 patients with fibrillary GN who were treated with rituximab (1 g IV 2 doses or 375 mg/m 2 4 doses) at the Center for Glomerular Diseases at Columbia University Medical Center. Non-progression of disease was defined as stable/improved serum creatinine (SCr) with a minimum of 1 year of follow-up. Results. The median SCr was 2.1 (range ) mg/dl, median estimated glomerular filtration rate (egfr) 39 (range 21 98) ml/ min/1.73 m 2 and median proteinuria 4497 (range ) mg/ day at the time of rituximab initiation. Four patients had received immunosuppression before rituximab, and nine received immunosuppression after rituximab, with four receiving a second rituximab course. Four of 12 patients were non-progressors, 3 of 12 had progressive without reaching, and 5 patients reached. The median follow-up for patients who did not reach was 38 (range 14 76) months after rituximab treatment. Non-progressors had lower SCr values, higher egfrs and shorter median duration from diagnosis to treatment than progressors. No serious adverse events were noted. Conclusions. Rituximab therapy was associated with nonprogression of disease in 4 of 12 patients. At the time of treatment, these non-progressors had better function and shorter time from diagnosis to treatment than progressors. Keywords: Glomerular disease, nephrotic syndrome, proteinuria INTRODUCTION Fibrillary glomerulonephritis (GN) is a rare glomerular disease defined histopathologically by the presence of glomerular immunoglobulin G (IgG) deposits on immunofluorescence with characteristic randomly arranged fibrils of average diameter 20 (range 10 30) nm in the mesangium and/ or glomerular basement membrane on electron microscopy [1] (Figure 1). The majority of fibrillary GN cases exhibit polyclonal IgG deposits that contain predominantly IgG4 [2, 3]. The deposits in fibrillary GN do not exhibit the characteristic staining for Congo Red observed in amyloidosis and are distinguished from immunotactoid GN by their smaller size (average 20 nm in fibrillary GN versus 30 nm in immunotactoid GN). Moreover, monoclonal IgG deposits are more common in immunotactoid GN compared with fibrillary GN [3, 4]. Fibrillary GN comprises % of kidney biopsies and usually presents clinically with the nephrotic syndrome and/or reduced kidney function [2, 4 7]. No diagnostic serologic tests exist. The prognosis of fibrillary GN is poor, with up to 50% of patients progressing to end-stage disease () within 2 years of diagnosis [2, 4, 5, 7]. The published experience on the treatment of fibrillary GN is comprised of retrospective case series and case reports describing blockade of the renin angiotensin aldosterone axis and a variety of non-specific immunosuppressive therapies commonly used in other glomerular diseases, such as steroids, cyclophosphamide, cyclosporine and mycophenolate mofetil (MMF). However, no treatment has been shown to improve long-term outcomes. The observation that glomerular deposits in fibrillary GN are polyclonal IgG, which may be IgG4 subclass-dominant, has led some to postulate that fibrillary GN is an autoimmune condition [3] that may be treated with rituximab, a monoclonal anti-cd20 antibody. This rationale is similar to the use of rituximab in membranous nephropathy and other autoimmune glomerular diseases. A previous case series reported a reduction in proteinuria and stabilization of kidney function in three patients with fibrillary GN treated with rituximab with follow-up times ranging from 9 to 46 months after treatment [8]. The largest case series of fibrillary GN included three patients treated with regimens involving rituximab, but detailed follow- The Author Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. 1925

2 FIGURE 1: Renal biopsy findings in fibrillary GN. Light microscopic findings can range from a mild mesangial proliferative GN to more severe forms of proliferative glomerular disease. Panel A shows expansion of the mesangial areas by increased matrix admixed with fibrillary deposits and accompanied by mild mesangial proliferation. Segmental infiltration of glomerular basement membranes by deposits can also be seen (arrows), imparting a focally membranoproliferative appearance (Periodic Acid Schiff, 600 magnification). Panel B shows compression of the glomerular tuft by an overlying cellular crescent. The remainder of the glomerulus appears hypercellular due to mesangial and endocapillary proliferation (Jones methenamine silver, 400). Immunofluorescence typically reveals polyclonal deposits that stain dominantly for IgG. As shown in Panel C, the deposits typically have a smudgy texture and they typically involve the mesangium and segmentally infiltrate glomerular basement membranes (direct immunofluorescence with anti-sera to IgG, 400). Electron microscopy is required to establish the diagnosis of fibrillary GN. Panel D shows the characteristic fibrils, which are randomly oriented, non-branching and typically measure between 16 and 24 nm in diameter ( 80,000). up was not presented [7], and a recent case series of 27 patients with fibrillary GN reported outcomes on seven patients with varying degrees of insufficiency treated with rituximab, in addition to other immunosuppression regimens [9]. Here, we present the largest series of patients with fibrillary GN treated with rituximab with follow-up for up to 6 years. MATERIAL AND METHODS We evaluated all adult cases of fibrillary GN treated with rituximab at the Center for Glomerular Diseases at Columbia University Medical Center between 2007 and Fibrillary GN was defined as the presence of characteristic, randomly arranged fibrils on electron microscopy with an average of 20 nm in size, and which were negative for staining with Congo Red. Progression of disease was the primary outcome and was subdivided into three categories: nonprogression, progression and. Non-progression of disease was defined as a serum creatinine that was stable, improved or did not increase by >25% from the time of rituximab treatment until last follow-up. Progressive disease was defined as an increase in serum creatinine by >25% but not reaching at last follow-up. was defined as initiation of replacement therapy or undergoing transplant during the follow-up period. In order to be eligible for analysis, we required a minimum follow-up time of 12 months after rituximab treatment unless was reached prior to that time point. Clinical and historical data were collected on all patients via chart review. Renal biopsy reports were assayed for diagnosis confirmation, light microscopy pattern, degree of glomerulosclerosis, interstitial fibrosis and tubular atrophy, and vascular disease. Adverse events related to rituximab use were recorded based on the content of chart reviews. Given the small number of patients in the series, no formal statistical analysis was performed. Instead, the data are presented here descriptively. The institutional review board on human research approved this retrospective study. RESULTS Between 2007 and 2012, 12 patients with fibrillary GN who received rituximab were followed at the Center for Glomerular Diseases at Columbia University. Their characteristics at the 1926 J. Hogan et al.

3 time of rituximab treatment are presented in Table 1. Median age was 62 (range 38 74) years at initiation of treatment, eight patients were female and all were white and non-hispanic. The median time from diagnosis to treatment was 6.5 (range ) months. Rituximab was prescribed as primary therapy in eight patients; the remaining four patients had received prior immunosuppression: cyclosporine (n = 1), prednisone (n = 1), prednisone + MMF (n = 1) or dexamethasone + MMF (n = 1). All patients were receiving angiotensin-converting enzyme inhibitor (ACEI) or angiotensin-ii receptor blocker (ARB) therapy. No patient had previously undergone transplantation. The median serum creatinine was 2.1 (range ) mg/ dl at the time of rituximab treatment, with a median egfr of 39 (range 21 98) ml/min/1.73 m 2 (Table 1). Only three patients had an egfr of >60 ml/min/1.73 m 2 when initiating rituximab therapy. The median pretreatment proteinuria (available in 11 of 12 patients) was 4497 (range ) mg/day as measured by 24-h urine protein measurement or spot urine protein/creatinine ratio; seven patients had >3500 mg/day of proteinuria. Serum protein electrophoresis, urine protein electrophoresis and/or serum kappa/lambda free light chain ratio were performed in six patients and revealed no evidence of paraproteinemia. No association was observed between fibrillary GN and history of autoimmune disease, malignancy or hepatitis C, as has been reported in previous case series. The glomerular histology was membranoproliferative GN in eight patients, mesangioproliferative GN in two patients and diffuse proliferative GN in two patients, including one who had cellular crescents (Table 2). One patient (Patient 2) did not have glomeruli sampled for immunofluorescence microscopy. Patient 8 had diffuse proliferative GN on light Table 1. Baseline characteristics (at time of treatment) of patients with fibrillary GN treated with rituximab Patient Age Gender Race Previous immunosuppression Time from diagnosis to rituximab treatment (months) microscopy with monotypic IgG lambda deposits on immunofluorescence, and Patient 9 had mesangioproliferative GN on light microscopy with monotypic IgG kappa deposits on immunofluorescence. Both of these patients had no monoclonal protein spike identified on serum and urine protein electrophoresis. All other biopsies exhibited polyclonal staining for kappa and lambda. IgG subclass staining was not performed. Eleven of 12 patients received two doses of rituximab 1000 mg intravenously, separated by 2 weeks, and 1 patient (Patient 9) received 4 weekly doses of 375 mg/m 2. All patients received intravenous corticosteroids as premedication, and five patients were prescribed concomitant oral corticosteroids with subsequent taper immediately after rituximab treatment (Table 3). During the follow-up period, nine patients received additional immunosuppression, including four who received a second course of rituximab. At last follow-up, four (33%) patients were non-progressors, three (25%) had progressive disease and five (42%) had reached (Table 3, Figures 1A C). The median follow-up time from rituximab infusion was 38 (range 14 76) months for those who did not reach, and 17 (range ) months for patients who reached. Non-progressors had lower median baseline serum creatinine values (median 1.0 mg/dl) and higher baseline egfrs (median 72 ml/min/ 1.73 m 2 ) at the time of rituximab administration than patients with progressive disease (median creatinine 2.1 mg/dl, egfr 38 ml/min/1.73 m 2 ) and patients who reached (median creatinine 2.5 mg/dl, egfr 19 ml/min/1.73 m 2 ). In addition, non-progressors demonstrated a shorter median duration from diagnosis to treatment (4.3 months) than those with progressive disease (8 months) and (29.3 months). No such trend was observed with regard to egfr (ml/min/1.73 m 2 ) 1 48 Male White Cyclosporine (100 mg BID for 54 months, stopped 9 months before rituximab) 2 68 Male White None Female White None Female White None Female White None NR 6 67 Female White None Female White None Male White Prednisone 120 mg every other day for months, then taper 9 58 Female White MMF mg BID from 27 months prior to the time of rituximab Dexamethasone 8 mg daily for 1 month (2 years before rituximab) with subsequent taper over 2 months Female White None Female White Prednisone 120 mg every other day MMF mg BID from 4 months prior to rituximab Male White None Proteinuria a (mg/day) SCr, serum creatinine; egfr, estimated glomerular filtration rate calculated by MDRD formula; BID, twice daily; NR, not recorded; MMF, mycophenolate mofetil. a Proteinuria was calculated either by 24 h urine protein measurement or spot urine protein/creatinine ratio. Rituximab for fi brillary glomerulonephritis 1927

4 Table 2. Light microscopy and immunofluoresence findings a on biopsy Patient LM pattern IgG IgM IgA C3 C1 Kappa Lambda 1 MPGN ± b MPGN NA NA NA NA NA NA NA 3 MPGN MPGN MPGN MPGN MPGN ± DPGN ± 3+ 9 MesProlif GN 3+ ± ± 3+ ± 10 MesProlif GN DPGN with crescents 3+ ± MPGN a The immunofluorescence staining represents glomerular staining intensity only. b The immunofluorescence microscopy for Patient 2 did not contain any glomeruli. LM, light microscopy; MPGN, membranoproliferative glomerulonephritis; DPGN, diffuse proliferative glomerulonephritis; MesProlif GN, mesangioproliferative glomerulonephritis; NA, not available. pretreatment proteinuria or any biopsy finding (Table 4). Although post-treatment proteinuria was not available for all patients, three of four non-progressors experienced improvement in proteinuria (Table 3). All five of the patients who reached underwent kidney transplant. There was no evidence of recurrent fibrillary GN in the allograft with a median follow-up of 19.5 (range 8 48) months post-transplant. Three adverse events were noted in three patients, two of which were noted during the second rituximab treatment course. Patient 2 felt extreme fatigue after the initial infusion, leading to cancelation of the final dose. Patient 4 experienced chills and hypotension during the initial infusion of the second treatment course leading to the infusion being terminated early but tolerated the final dose without complication. Patient 3 underwent a transplant 3 months after receiving rituximab and 13 months later was diagnosed with metastatic appendiceal goblet cell adenocarcinoma. No major infections were noted during the follow-up period. DISCUSSION This case series represents the largest experience of treating fibrillary GN with rituximab. Four (33%) patients had non-progression of disease, three (25%) had progressive disease and five (42%) reached, over a median follow-up time after rituximab of 38 (range 14 76) months in those who did not reach and 17 (range ) months in those who reached. Non-progressors had a lower baseline serum creatinine, higher baseline egfr and shorter time from diagnosis to rituximab than progressors. Three of the four non-progressors had a decrease in proteinuria with rituximab therapy. Treatment was well-tolerated, with only one patient experiencing an infusion reaction, and no significant infections occurred during the follow-up period. The prognosis of fibrillary GN is poor, with up to half of all patients progressing to within 2 years [2, 4, 5, 7]. Due to the rarity of this disease, no controlled trials have been conducted in the treatment of fibrillary GN. The two largest case series of patients with fibrillary GN have described heterogeneous treatment regimens with blockers of the renin angiotensin aldosterone system (ACEIs and ARBs), corticosteroids, cyclophosphamide, MMF, melphalan, lenalidomide, azathioprine, cyclosporine and rituximab, with no association between treatment and progression to. Better outcomes have been associated with lower serum creatinine at the time of biopsy [4, 7]. Rituximab has been proposed as a treatment of fibrillary GN due to the characteristic presence of polyclonal immunoglobulin deposits in the mesangium and glomerular basement membrane. Collins et al. described three adults with the nephrotic syndrome due to fibrillary GN and with preserved function (serum creatinine mg/dl) who were treated with rituximab (375 mg/m 2 /week 4 weeks in two patients, and 1000 mg 2 doses 2 weeks apart in the third patient) [8]. All three patients had non-progression of disease and improvement in proteinuria during a follow-up period of 9 46 months. One patient experienced worsening of proteinuria during the follow-up period, and improvement in proteinuria was achieved with a second course of rituximab. Rituximab was also used in three patients in the series by Nasr et al., in combination with corticosteroids (n = 1), steroids and cyclophosphamide (n = 1) and cyclophosphamide + MMF (n = 1), with outcomes of (n = 1) and persistent (n = 2) in these patients [7]. Rituximab did not induce disease remission in two case reports of one adult and one child with fibrillary GN and compromised function [10, 11]. Most recently, the long-term follow-up of 27 patients with fibrillary GN was published, seven of whom received rituximab as part of their treatment regimen [9]. One patient had a complete response, four patients achieved partial response and two patients had no response to rituximab treatment (Table 5). The median pretreatment egfr for the five responders to rituximab in this series was 77 (range 35 96) ml/min/1.73 m 2. Most reported treatment successes of fibrillary GN with any immunosuppression regimen have been in patients with preserved function [8, 9, 12]. Three of our non-progressors (Patients 4, 10 and 12) had serum creatinine of <1.2 mg/ dl and egfr of >60 ml/min/1.73 m 2 at the time of rituximab treatment. Therefore, it may be hypothesized that rituximab may only prevent or prolong progression of disease in patients with preserved function and that it may not be effective after a point of no return of. In contrast, patient 6 0 s treatment course was arguably the most impressive as her declining function dramatically improved following her first course of rituximab. Although it should be noted that her proteinuria was decreasing prior to her first rituximab dose, during the follow-up period, her creatinine began to rise as B-cells returned but again improved after a second course of rituximab (Figure 2A). This case series is notable for its relatively large number of patients for such a rare disease and unconventional therapy, as well as for the long-term follow-up. Nonetheless, our findings are subject to important limitations. Chief among these 1928 J. Hogan et al.

5 Table 3. Treatment regimens and outcomes in patients with fibrillary GN treated with rituximab Patient Concomitant oral steroid treatment? Change in lab values pre-/post-rituximab Variable Initial Last f/u 1 None 2 None 3 None 4 None 5 None 6 Prednisone 120 mg every other day 2 months, then tapered 7 Prednisone 120 mg every other day 1 month, then tapered 8 None 9 Dexamethasone 8 mg daily 2 weeks 10 None 11 Prednisone 120mg every other day for 2months, then tapered 12 None NA NA Immunosuppression after rituximab (SCr and proteinuria at time of initiation) MMF 2 g/day from months 13 to last follow-up (3.2 mg/dl, 8000 mg/g) MMF 1 2 g/day from months after initial rituximab dose (NA, NA) Rituximab 1000 mg 1 dose 39 months after first rituximab dose (3.8 mg/dl, 3269 mg/g) Cyclosporine 150 mg q12 h from month 0 to 17 after rituximab (NA, NA) Rituximab 1000 mg 58 and 61 months after initial rituximab dose (0.8 mg/dl, 4777 mg/24 h) Tacrolimus 1 mg q12 h from months 68 through last follow-up (0.7 mg/dl, 2793 mg/g) Follow-up time after rituximab (months) Outcome a 24 Progressive 42 Progressive Nonprogressor None 27 Rituximab 1000 mg IV 2 doses 7.5 months after initial rituximab dose, with prednisone 120 mg every other day 2 months then taper 2.6 mg/dl, 487 mg/g) Rituximab 1000 mg IV 2 at 12 months after initial dose, with prednisone 120 mg every other day 1 month then taper (2.8 mg/dl, NA) MMF 1000 mg BID twice daily and prednisone 120 mg every other day for 2 months, with taper, started 14 months after rituximab dose (NA, 2760 mg/g) MMF 500 mg BID from time of rituximab until 2 months after rituximab, dexamethasone 8 mg daily with taper from time of rituximab (2.3 mg/dl, 1817mg/g) Oral cyclophosphamide 50 mg BID for 2 weeks, started 2 months after rituximab (4.5 g/dl, mg/g) 38 Nonprogressor Progressive 3 None 25 Nonprogressor Prednisone 120 mg every other day and MMF 1000 mg BID from time of rituximab until stopped 1.5 months after rituximab (3.0 mg/dl, 4500 mg/g) 1.5 None 14 Nonprogressor All patients were treated with rituximab 1000 mg IV 2 doses except for Patient 9, who received 375 mg/m 2 mg weekly for 4 weeks. a All patients who reached received kidney transplants without evidence of recurrent disease at last follow-up. SCr, serum creatinine; egfr, estimated glomerular filtration rate calculated by MDRD formula; MMF, mycophenolate mofetil;, end-stage disease; BID: twice daily. Rituximab for fi brillary glomerulonephritis 1929

6 Table 4. Pre-rituximab clinical and histopathologic data for each outcome category Outcome Nonprogressor (n =4) Progressive (n =3) egfr (ml/min/1.73 m 2 ) Time from diagnosis to treatment (months) Proteinuria (mg/day) 1.0 ( ) 72 (22 96) ( ) 2.0 ( ) 38 (34 43) ( ) (n = 5) 2.7 ( ) 19 (17 28) ( ) Prior immunosuppression Histology None (4) MPGN (3) MesPGN (1) Cyclosporine (1) Prednisone (1) None (1) Steroids +MMF (2) None (3) MPGN (2) DPGN (1) MPGN (3) MesPGN (1) DPGN w/ crescents (1) % Global glomerulosclerosis %TAIF All data expressed as median values (range). TAIF, tubular atrophy, interstitial fibrosis;, end-stage disease; MMF, mycophenolate mofetil; MPGN, membranoproliferative glomerulonephritis; MesPGN, mesangioproliferative glomerulonephritis; DPGN, diffuse proliferative glomerulonephritis. Table 5: Published experience of rituximab treatment for fibrillary GN Number of patients treated with rituximab Rituximab regimen Renal function at rituximab treatment Follow-up after rituximab Outcome(s) Collins et al. [7] 3 adults 375 mg/m 2 weekly 4 SCr months Non-progression (n =3) doses (n =2) I000 mg IV 2 doses (n =1) Nasr et al. [6] 3 adults NR NR NR Persistent (n =2) (n =1) Javaugue et al. [8] 7 adults 375 mg/m doses Pliquett et al. [9] 1 adult 375 mg/m 2 weekly 4 doses Menon et al. [10] 1 child 375 mg/m 2 weekly 4 doses Hogan et al. 12 adults 1000 mg IV 2 doses [current study] (n = 11) 375 mg/m 2 weekly 4 doses (n =1) a Expressed as median values. NR, not recorded; egfr, estimated glomerular filtration rate; SCr, serum creatinine;, end-stage disease. egfr 66 ml/min/1.73 m 2a 44 months a No response (n =2) Complete response (n =1) Partial response (n =4) SCr 3.4 mg/dl 20 months No response SCr 2.0 mg/dl 1 month SCr 2.1 mg/dl a egfr 39 ml/min/1.73 m 2a 38 months a Non-progression (n =4) Progressive disease (n =3) (n =5) limitations are the retrospective design and the lack of a comparison group against which to interpret the results. Speculatively, the responders to rituximab (and, in particular, those whose pretreatment egfr was in the normal range) may have responded to other immunosuppressive regimens, as all were treatment-naïve when started on therapy. Further potential confounders include the additional immunosuppression used in four patients prior to rituximab therapy and in nine patients after rituximab therapy (including four patients who received additional courses of rituximab), and the large number of biopsies with MPGN pattern on light microscopy, which has been associated with worse clinical outcomes compared with mesangioproliferative pattern in past fibrillary GN series [4]. B-cell responses to rituximab were not monitored uniformly for all patients, and thus, we are unable to correlate treatment response and/or relapse with depletion and/or reappearance of B cells. The apparently better outcomes in patients who received rituximab with preserved egfr may be an example of lead time bias, although the long-term follow-up in this cohort reduces the chances of such bias. Moreover, patients with burn out disease and progressive failure may not respond to treatment. Finally, no cost analysis was performed, an important consideration for a medication with high treatment costs. In conclusion, we present the largest experience using rituximab in the treatment of fibrillary GN, with follow-up times that extend past 6 years. In general, non-progressors had better function and shorter time from diagnosis to rituximab treatment than progressors. While rituximab may be a promising treatment option for patients with fibrillary GN and preserved function, prospective, controlled trials are needed to further evaluate its use in this disease J. Hogan et al.

7 FIGURE 2: Serum creatinine versus time from rituximab treatment in patients with fibrillary GN. Time zero indicates the initiation of rituximab treatment. Arrows indicate when repeat rituximab courses were administered to four patients. The trends in creatinine are stratified by response to rituximab: non-progressors (A), progressors to later stages of CKD (B) and progressors to (C). ACKNOWLEDGEMENTS G.B.A has received research grants, consultantships and/or speaker honoraria from the following companies: Ardea, Alexion, Pfizer, Merck, Genentech, Aspreva (Vifor), Bristol- Myers Squibb, Questcor, FibroGen, Up-to-Date, Amgen, Centocor Ortho Biotech, Sanofi, Novartis and Teva. Genentech is the manufacturer of Rituxan (rituximab). CONFLICT OF INTEREST STATEMENT None declared. REFERENCES 1. Alpers CE, Kowalewska J. Fibrillary glomerulonephritis and immunotactoid glomerulopathy. J Am Soc Nephrol 2008; 19: Iskandar SS, Falk RJ, Jennette JC. Clinical and pathologic features of fibrillary glomerulonephritis. Kidney Int 1992; 42: Bridoux F, Hugue V, Coldefy O et al. Fibrillary glomerulonephritis and immunotactoid (microtubular) glomerulopathy are associated with distinct immunologic features. Kidney Int 2002; 62: Rosenstock JL, Markowitz GS, Valeri AM et al. Fibrillary and immunotactoid glomerulonephritis: distinct entities with different clinical and pathologic features. Kidney Int 2003; 63: Fogo A, Qureshi N, Horn RG. Morphologic and clinical features of fibrillary glomerulonephritis versus immunotactoid glomerulopathy. Am J Kidney Dis 1993; 22: Kalbermatter SA, Marone C, Casartelli D et al. Outcome of fibrillary glomerulonephritis. Swiss Med Wkly 2012; 142: w Nasr SH, Valeri AM, Cornell LD et al. Fibrillary glomerulonephritis: a report of 66 cases from a single institution. Clin J Am Soc Nephrol 2011; 6: Collins M, Navaneethan SD, Chung M et al. Rituximab treatment of fibrillary glomerulonephritis. Am J Kidney Dis 2008; 52: Javaugue V, Karras A, Glowacki F et al. Long-term Kidney disease outcomes in fibrillary glomerulonephritis: a case series of 27 patients. Am J Kidney Dis 2013; 62: Pliquett RU, Mohr P, El Din Mukhtar B et al. Plasmapheresis leading to remission of refractory nephrotic syndrome due to fibrillary glomerulonephritis: a case report. J Med Case Rep 2012; 6: Menon S, Zeng X, Valentini R. Fibrillary glomerulonephritis and failure in a child with systemic lupus erythematosus. Pediatr Nephrol 2009; 24: Dickenmann M, Schaub S, Nickeleit V et al. Fibrillary glomerulonephritis: early diagnosis associated with steroid responsiveness. Am J Kidney Dis 2002; 40: E9 Received for publication: ; Accepted in revised form: Rituximab for fi brillary glomerulonephritis 1931