Advertisement
American Journal of Kidney Diseases

Recurrence of DNAJB9-Positive Fibrillary Glomerulonephritis After Kidney Transplantation: A Case Series

      Rationale & Objective

      Fibrillary glomerulonephritis (FGN) is a rare glomerular disease that often progresses to kidney failure requiring kidney replacement therapy. We have recently identified a novel biomarker of FGN, DnaJ homolog subfamily B member 9 (DNAJB9). In this study, we used sequential protocol allograft biopsies and DNAJB9 staining to help characterize a series of patients with native kidney FGN who underwent kidney transplantation.

      Study Design

      Case series.

      Setting & Participants

      Between 1996 and 2016, kidney transplantation was performed on 19 patients with a reported diagnosis of FGN in their native/transplant kidneys. Using standard diagnostic criteria and DNAJB9 staining, we excluded 5 patients (4 atypical cases diagnosed as possible FGN and 1 donor-derived FGN). Protocol allograft biopsies had been performed at 4, 12, 24, 60, and 120 months posttransplantation. DNAJB9 immunohistochemistry was performed using an anti-DNAJB9 rabbit polyclonal antibody. Pre- and posttransplantation demographic and clinical characteristics were collected. Summary statistical analysis was performed, including nonparametric statistical tests.

      Observations

      The 14 patients with FGN had a median posttransplantation follow-up of 5.7 (IQR, 2.9-13.8) years. 3 (21%) patients had recurrence of FGN, detected on the 5- (n = 1) and 10-year (n = 2) allograft biopsies. Median time to recurrence was 10.2 (IQR, 5-10.5) years. Median levels of proteinuria and iothalamate clearance at the time of recurrence were 243 mg/d and 56 mL/min. The remaining 11 patients had no evidence of histologic recurrence on the last posttransplantation biopsy, although the median time of follow-up was significantly less at 4.4 (IQR, 2.9-14.4) years. 3 (21%) patients had a monoclonal protein detectable in serum obtained pretransplantation; none of these patients had recurrent FGN.

      Limitations

      Small study sample and shorter follow-up time in the nonrecurrent versus recurrent group.

      Conclusions

      In this series, FGN had an indolent course in the kidney allograft in that detectable histologic recurrence did not appear for at least 5 years posttransplantation.

      Index Words

      Fibrillary glomerulonephritis (FGN) is a rare glomerular disease encountered in 0.4% to 1.4% of native kidney biopsies.
      • Fogo A.
      • Qureshi N.
      • Horn R.G.
      Morphologic and clinical features of fibrillary glomerulonephritis versus immunotactoid glomerulopathy.
      • Nasr S.H.
      • Valeri A.M.
      • Cornell L.D.
      • et al.
      Fibrillary glomerulonephritis: a report of 66 cases from a single institution.
      • Rosenstock J.L.
      • Markowitz G.S.
      • Valeri A.M.
      • Sacchi G.
      • Appel G.B.
      • D'Agati V.D.
      • et al.
      Fibrillary and immunotactoid glomerulonephritis: distinct entities with different clinical and pathologic features.
      • Andeen N.K.
      • Troxell M.L.
      • Riazy M.
      • et al.
      Fibrillary glomerulonephritis: clinicopathologic features and atypical cases from a multi-institutional cohort.
      It is characterized by glomerular deposition of randomly oriented nonbranching fibrils that lack a hollow center as seen by electron microscopy(EM) and stain by immunofluorescence (IF) with antisera to immunoglobulins, in particular immunoglobulin G (IgG) and κ and λ light chains. The deposits are traditionally Congo red negative.
      • Fogo A.
      • Qureshi N.
      • Horn R.G.
      Morphologic and clinical features of fibrillary glomerulonephritis versus immunotactoid glomerulopathy.
      ,
      • Iskandar S.S.
      • Falk R.J.
      • Jennette J.C.
      Clinical and pathologic features of fibrillary glomerulonephritis.
      • Pronovost P.H.
      • Brady H.R.
      • Gunning M.E.
      • Espinoza O.
      • Rennke H.G.
      Clinical features, predictors of disease progression and results of renal transplantation in fibrillary/immunotactoid glomerulopathy.
      • Alpers C.E.
      • Rennke H.G.
      • Hopper Jr., J.
      • Biava C.G.
      Fibrillary glomerulonephritis: an entity with unusual immunofluorescence features.
      The prognosis is generally poor, with nearly half the patients progressing to kidney failure within 4 years of diagnosis.
      • Nasr S.H.
      • Valeri A.M.
      • Cornell L.D.
      • et al.
      Fibrillary glomerulonephritis: a report of 66 cases from a single institution.
      There is no specific treatment for FGN and kidney transplantation is the best treatment option.
      Studies evaluating FGN in allograft kidneys are few.
      • Andeen N.K.
      • Troxell M.L.
      • Riazy M.
      • et al.
      Fibrillary glomerulonephritis: clinicopathologic features and atypical cases from a multi-institutional cohort.
      ,
      • Alpers C.E.
      • Rennke H.G.
      • Hopper Jr., J.
      • Biava C.G.
      Fibrillary glomerulonephritis: an entity with unusual immunofluorescence features.
      • Brady H.R.
      Fibrillary glomerulopathy.
      • Samaniego M.
      • Nadasdy G.M.
      • Laszik Z.
      • Nadasdy T.
      Outcome of renal transplantation in fibrillary glomerulonephritis.
      • Czarnecki P.G.
      • Lager D.J.
      • Leung N.
      • Dispenzieri A.
      • Cosio F.G.
      • Fervenza F.C.
      Long-term outcome of kidney transplantation in patients with fibrillary glomerulonephritis or monoclonal gammopathy with fibrillary deposits.
      The reported rates of recurrence range from 8%
      • Mallett A.
      • Tang W.
      • Hart G.
      • et al.
      End-stage kidney disease due to fibrillary glomerulonephritis and immunotactoid glomerulopathy - outcomes in 66 consecutive ANZDATA registry cases.
      to 50%.
      • Samaniego M.
      • Nadasdy G.M.
      • Laszik Z.
      • Nadasdy T.
      Outcome of renal transplantation in fibrillary glomerulonephritis.
      This variation is largely accounted for by differing study methodologies: most studies are published from centers in which protocol allograft biopsy is not routinely practiced and some studies define recurrence clinically without biopsy. Moreover, definitions of FGN were imprecise. Some studies combined immunotactoid glomerulopathy and FGN and others combined FGN with entities that mimic FGN. In a previous study performed at our institution, recurrence did not occur in any of the 5 patients with FGN over a median follow-up of 52 months. However, 5 of 7 (71%) patients with glomerular diseases containing vague fibrillary deposits that mimicked FGN (FGN mimics) had recurrence of deposits in the allograft: 43% underwent retransplantation and 14% lost the allograft to recurrence.
      • Czarnecki P.G.
      • Lager D.J.
      • Leung N.
      • Dispenzieri A.
      • Cosio F.G.
      • Fervenza F.C.
      Long-term outcome of kidney transplantation in patients with fibrillary glomerulonephritis or monoclonal gammopathy with fibrillary deposits.
      These results highlight the importance of distinguishing FGN from similar entities in the native kidney, especially in the setting of monoclonal gammopathy detected pretransplantation, to help clarify the risk for recurrence and prognosis.
      The aim of our study is to characterize the natural history of FGN in the allograft, specifically in regard to recurrence, using sequential protocol biopsies and the recently discovered biomarker DnaJ homolog subfamily B member 9 (DNAJB9) in a series of kidney transplant recipients with FGN.
      • Mallett A.
      • Tang W.
      • Hart G.
      • et al.
      End-stage kidney disease due to fibrillary glomerulonephritis and immunotactoid glomerulopathy - outcomes in 66 consecutive ANZDATA registry cases.
      • El-Zoghby Z.M.
      • Stegall M.D.
      • Lager D.J.
      • et al.
      Identifying specific causes of kidney allograft loss.
      • Nasr S.H.
      • Vrana J.A.
      • Dasari S.
      • et al.
      DNAJB9 is a specific immunohistochemical marker for fibrillary glomerulonephritis.
      Our study is unique in that it uses sequential protocol and clinically indicated biopsies, with late posttransplantation biopsies occurring at 5 and 10 years.

      Methods

      Study Design, Cases, and Definitions

      The study was approved by the Mayo Clinic Institutional Review Board and includes patients who provided consent to use their medical records for research purposes. The kidney transplant database at Mayo Clinic was queried for biopsies with a native and allograft diagnosis of FGN or possible FGN that were made between 1996 and 2016. Posttransplantation follow-up includes sequential protocol kidney biopsies at 4, 12, and 24 months and 5 and 10 years posttransplantation, as well as kidney function measurements with iothalamate clearance yearly and 24-hour urine collection for albumin and protein.
      Inclusion criteria were a definitive diagnosis of FGN on native/allograft biopsy and a kidney transplant with posttransplantation biopsies. The native biopsy had to demonstrate typical morphologic features of FGN as defined by Rosenstock et al.
      • Rosenstock J.L.
      • Markowitz G.S.
      • Valeri A.M.
      • Sacchi G.
      • Appel G.B.
      • D'Agati V.D.
      • et al.
      Fibrillary and immunotactoid glomerulonephritis: distinct entities with different clinical and pathologic features.
      When sufficient material was available, DNAJB9 staining was performed to support/exclude the diagnosis.
      For all native and protocol biopsies, light microscopy, IF, and EM images were reviewed. The biopsies had been stained with hematoxylin and eosin, periodic acid–Schiff, Masson trichrome, and Jones methenamine silver. For IF, 3-μm cryostat sections were stained with polyclonal fluorescein isothiocyanate–conjugated antibodies to IgG, IgM, IgA, C3, C1q, C4d, κ light chain, λ light chain, fibrinogen, and albumin. All ultrastructural images were reviewed.
      During the study period, we identified 19 patients who underwent transplantation at Mayo Clinic with a native or allograft biopsy classified as FGN or glomerulonephritis with fibrillary-like deposits. In 4 of these patients, the native biopsy did not fulfill the inclusion criteria for FGN as described by Rosenstock et al.
      • Rosenstock J.L.
      • Markowitz G.S.
      • Valeri A.M.
      • Sacchi G.
      • Appel G.B.
      • D'Agati V.D.
      • et al.
      Fibrillary and immunotactoid glomerulonephritis: distinct entities with different clinical and pathologic features.
      Three of these 4 cases had tissue available on paraffin block for DNAJB9 staining; DNAJB9 was not detected in these 3, confirming that they were not FGN but rather FGN mimics, and they were excluded from analysis. Their clinicopathological data are summarized in Table S1 and briefly discussed in the results. A fifth patient had a diagnosis of allograft FGN made at 1 year with subsequent positive DNAJB9 staining of the implantation biopsy specimen, confirming a diagnosis of donor FGN. The native biopsy was not diagnostic of FGN based on the criteria of Rosenstock et al
      • Rosenstock J.L.
      • Markowitz G.S.
      • Valeri A.M.
      • Sacchi G.
      • Appel G.B.
      • D'Agati V.D.
      • et al.
      Fibrillary and immunotactoid glomerulonephritis: distinct entities with different clinical and pathologic features.
      and negative DNAJB9 staining. This patient, also discussed later in the results, was excluded from analysis. The remaining 14 patients fulfilled the inclusion criteria.
      Allograft recurrence of FGN was defined by typical features of FGN. The last allograft biopsy, clinical or protocol, of all 14 patients underwent staining for DNAJB9. DNAJB9 staining is not routinely performed on allograft biopsies and since its discovery, its use in transplant biopsy evaluation has been only in those with a diagnosis of native FGN. Clinical data, including demographic information, native disease characteristics, recurrent disease characteristics, treatment, and follow-up, were obtained from patients’ electronic or paper medical records. The following definitions were used: proteinuria, measured or predicted 24-hour urine protein excretion > 150 mg; hematuria, more than 5 red blood cells per high-power field on microscopic examination of urinary sediment; and nephrotic syndrome, urine protein excretion ≥ 3,500 mg/d, serum albumin level < 3.5 g/dL, and peripheral edema.

      DNAJB9 Immunohistochemistry

      All instruments and reagents were purchased from Ventana Medical Systems, Inc unless otherwise specified. DNAJB9 immunohistochemistry was performed on 4-μm thick formalin-fixed paraffin-embedded tissue sections mounted on charged slides. Tissue slides were dried and melted in an oven at 68°C for 20 minutes. Slides were stained with an anti-DNAJB9 rabbit polyclonal antibody (catalog no. HPA040967 at a dilution of 1:75 [Sigma-Aldrich] or catalog no. PA5-59621 at a dilution of 1:200 [Thermo Fisher Scientific]) on a Ventana BenchMark XT system. The staining protocol included online deparaffinization, heat-induced epitope retrieval with Ventana Cell Conditioning 1 solution for 32 minutes, and incubation with the primary antibody for 32 minutes at 37°C. Antigen-antibody reactions were visualized using Ventana OptiView Universal DAB Detection and OptiView Amplification Kits. Counterstaining was performed online using Ventana Hematoxylin II for 8 minutes, followed by bluing reagent for 4 minutes. DNAJB9 staining was performed on 10 native biopsy specimens (those with available material) and on at least the last allograft biopsies of all 14 transplants.

      Statistical Analysis

      Statistical analysis was performed using nonparametric exact statistical methods. Univariate analysis was performed using Mann-Whitney-Wilcoxon test, Kruskal-Wallis test, and Fisher-Freeman-Halton exact test, as appropriate for variable type. Statistical significance was assumed at P < 0.05.

      Results

      Patient and Transplant Characteristics of Study Cohort

      Fourteen kidney transplant recipients who fulfilled the study criterion of native FGN were evaluated (Table 1). These 14 patients had a total of 17 kidney transplants. Median recipient age at time of transplantation was 67.1 (interquartile range [IQR], 53.7-68.5) years; 57% were women, and all were white. Median donor age was 48 (IQR, 35-64) years. Eight of 17 (47%) transplantions were preemptive and 9 (53%) were performed after dialysis initiation, with median dialysis vintage of 13 (range, 4-66) months. Median time from diagnosis to the development of kidney failure was 41 (range, 2-144) months. Seventy-one percent received a living donor allograft, and 29%, a deceased donor allograft. Forty-one percent of patients received antithymocyte globulin induction and 59% were maintained on triple immunosuppressive therapy with tacrolimus, mycophenolate mofetil (MMF), and prednisone. Median proteinuria pretransplantation was protein excretion of 4.2 (range, 2.4-7.7) g/d. Pretransplantation monoclonal gammopathy screening was performed in all patients, with 3 of 14 (21%) having detectable serum monoclonal protein. Median posttransplantation follow-up was 5.7 (range, 0.3-15.5) years. At the time of last follow-up, 6 of 14 (43%) patients had died and 5 of 17 (29%) allografts failed. One patient developed posttransplantation lymphoproliferative disorder. Five other patients were diagnosed with solid-organ malignancies posttransplantation (Table 1).
      Table 1Clinical Characteristics of 14 Patients With Fibrillary Glomerulonephritis Status Post 17 Kidney Transplantations
      ParameterValue
      Age at time of transplant, y (n = 17)67.1 [53.7-68.5]
      Female sex8 (57%)
      White race14 (100%)
      Associated medical conditions (n = 14)
       Hematologic condition
       MGUS2 (14%)
       PTLD1 (7%)
       Posttransplantation malignancy
      Urothelial carcinoma (n=1), renal cell carcinoma (n=1), esophageal carcinoma (n=1), lung carcinoma (n=1), and metastatic melanoma (n=1).
      5 (36%)
       Hepatitis C virus infection1 (7%)
      Preemptive transplantation (n = 17)8 (47%)
      Dialysis vintage, mo (n = 9)
       Median13 [6-26.5]
       Range4-66
      Proteinuria pretransplantation (n = 11)
       Median, mg/d4,233 [2,395-7,699]
       Nephrotic range8 (73%)
      Type of kidney transplant (n = 17)
       Living related8 (47%)
       Living unrelated4 (24%)
       Deceased donor5 (29%)
      Donor age, y (n = 17)48 [35-64]
      HLA antigen mismatches3 [1.25-4.75]
      Induction therapy
       Thymoglobulin7 (41%)
       Basiliximab4 (24%)
       Alemtuzumab1 (6%)
       Unknown5 (29%)
      Maintenance immunosuppression
       Tacrolimus/MMF/prednisone10 (59%)
       Other
      Tacrolimus/MMF (n=3), sirolimus/tacrolimus (n=1), MMF/prednisone (n=1), and unknown (n=2).
      7 (41%)
      Transplant-related complications (n = 17)
       Rejection7 (41%)
       BK nephropathy2 (12%)
      Presence of pretransplantation monoclonal protein (n = 14)3 (21%)
      Kidney transplant recurrence
       Of 17 transplants3 (18%)
       Of 14 recipients3 (21%)
      Proteinuria at time of recurrence (n = 3)
       Median, mg/d243 [98-3,403]
       Range, mg/d98-3,403
       Nephrotic range1 (33%)
      Iothalamate clearance at time of recurrence (n = 3)
       Median, mL/min56 [16-98]
       Range, mL/min16-98
      Time to recurrence, y (n = 3)
       Median10.2 [5-10.5]
       Range5-10.5
      Follow-up time, y
       Median5.7 [2.9-13.8]
       Range0.34-15.5
      Kidney allograft failure (n = 17)5 (29%)
      Death (n = 14)6 (43%)
      Age at death, y67.6 [60.2-80.2]
      Note: Unless otherwise indicated, values for continuous variables given as median [interquartile range]; for categorical variables, as count (percentage).
      Abbreviations: MGUS, monoclonal gammopathy of undetermined significance; MMF, mycophenolate mofetil; PTLD, posttransplantation lymphoproliferative disorder.
      a Urothelial carcinoma (n = 1), renal cell carcinoma (n = 1), esophageal carcinoma (n = 1), lung carcinoma (n = 1), and metastatic melanoma (n = 1).
      b Tacrolimus/MMF (n = 3), sirolimus/tacrolimus (n = 1), MMF/prednisone (n = 1), and unknown (n = 2).
      Of the 14 cases, 3 had recurrent FGN (with associated DNAJB9-positive staining of glomeruli), while 11 did not demonstrate either morphologic features of FGN or DNAJB9 positivity in the last allograft biopsy and were classified as nonrecurrent FGN. A flow chart of the study design is presented in Figure S1. Among the recurrent FGN group, 2 patients were first diagnosed on the 10-year protocol biopsies, and 1, on the 5-year biopsy. All allograft biopsies before 5 years were negative for FGN. Median proteinuria at the time of recurrence was protein excretion of 243 mg/d, and median measured glomerular filtration rate by iothalamate clearance was 56 mL/min.

      Clinical Characteristics of Recurrent Versus Nonrecurrent FGN

      Clinical characteristics of the 3 patients with recurrent FGN compared with the 11 patients without recurrence are summarized in Table 2. There was no significant difference in sex, recipient age, donor age, HLA antigen mismatches, and induction and maintenance immunosuppression. The 3 patients with recurrent FGN did not have a monoclonal protein detectable in serum. Compared with those with nonrecurrent FGN, patients with recurrent FGN had significantly longer follow-up (median time, 14.2 vs 4.4 years; P < 0.02). Median time of the last follow-up biopsy from transplantation for patients with nonrecurrent FGN was 2 (IQR, 0.01-10) years versus 10 (IQR, 5-10) years for patients with recurrent FGN. The last allograft biopsy in the nonrecurrent FGN group was performed at year 10 (n = 1), 5 (n = 2), 4 (n = 1), 2 (n = 3), and 1 or less (n = 4; Fig S1). One of the patients with an allograft biopsy at less than 1 year had lost the allograft at day 7 to thrombosis and infarction. This patient had 2 allograft biopsies, neither of which showed recurrence. None of the patients with recurrence died, as opposed to 6 of 11 (55%) patients with nonrecurrent FGN (P = 0.05). Median time from transplantation to death was 5.0 (IQR, 0.6-12.8) years; causes of death are summarized in Table S2. Five patients lost their kidney allografts, but only 1 patient lost the allograft in the setting of recurrent FGN; this patient (patient 1) also had chronic antibody-mediated rejection with transplant glomerulopathy, with the cause of graft loss presumed to be antibody-mediated rejection. Three patients (all with nonrecurrent FGN) had a circulating monoclonal protein detected pretransplantation (IgA κ light chain, IgG λ light chain, and IgA λ light chain). All 3 patients were considered to have monoclonal gammopathy of undetermined significance (MGUS), and no progression of a hematologic condition was noted in 2 patients; 1 died of sepsis as result of diverticulitis 15 years posttransplantation. The third patient with MGUS developed PTLD and died of complications within 6 months of kidney transplantation.
      Table 2Clinical Comparison of Those With Recurrent and Nonrecurrent FGN
      ParametersRecurrence (n = 3 patients and 3 transplants)No Recurrence (n = 11 patients and 14 transplants)P
      Age at time of transplantation, y67.1 [40.1-67.9]67.4 [53.3-71.4]0.5
      Female sex2 (67%)5 (46%)0.7
      White race100%100%0.9
      Preemptive transplantation2 (67%)4 (36%)0.5
      Proteinuria pretransplantation (6 missing data)
       Mean, mg/d2,655 ± 1,5135,698 ± 3,3160.1
       Median, mg/d2,655 [1,585-3,725]5,010 [2,395-7,699]0.2
       Nephrotic range1/2 (50%)5/7 (71%)0.6
      Type of kidney transplantation (n = 19)0.7
       Living related2 (67%)6 (55%)
       Living unrelated0 (0%)2 (18%)
       Deceased donor1 (33%)3 (27%)
      Donor age, y (n = 2 missing)
       Mean29 ± 14.146.3 ± 14.40.3
       Median29 [19-39]44 [34.5-57.5]0.2
      HLA antigen mismatches2 ± 1.43.5 ± 1.60.2
      Induction therapy0.5
       Thymoglobulin2 (67%)3 (27%)
       Other1 (33%)8 (73%)
      Maintenance therapy (n = 1 missing)0.9
       Triple therapy (tacrolimus/MMF/prednisone)2/3 (67%)6 (60%)
       Steroid-free regimen1/3 (33%)4 (40%)
      Presence of monoclonal protein (n = 16)0 (0%)3 (27%)0.3
      History of HCV infection0 (0%)1 (9%)0.6
      Proteinuria at 1 y, mg/d (n = 4 missing)128.5 [97-160]148.5 [57-277.5]0.9
      Proteinuria at time of recurrence, mg/d (n = 3)243 [98-3,403]
      Iothalamate clearance, mL/min
       At 1 y (n = 5 missing)82 [51-113]52 [47-60]0.5
       At time of recurrence (n = 3)56 [16-98]
      BMI at time of recurrence, kg/m233.9 [21.7-36.4]
      Kidney allograft failure1/3 (33%)4/14 (29%)0.9
      Death0 (0%)6 (55%)0.05
      Follow-up time, y (n = 14)14.2 [14-15.5]4.4 [2.9-14.4]0.02
      Follow-up time, y (n = 17)14.2 [14-15.5]4.4 [2.8-11.4]0.02
      Note: n = 14 patients.
      Abbreviations: BMI, body mass index; FGN, fibrillary glomerulonephritis; HCV, hepatitis C virus; MMF, mycophenolate mofetil.

      Clinicopathologic Course of Recurrent FGN

      The clinical course of the 3 patients with recurrent FGN is summarized next. Details of their biopsies are presented in Table S3.
      Patient 1 was diagnosed with FGN (at an outside institution) and underwent living related kidney transplantation with maintenance immunosuppression consisting of MMF and tacrolimus. Ten years posttransplantation, a progressive increase in proteinuria was noted with serum creatinine level of 1.7 mg/dL. Kidney biopsy revealed a mesangioproliferative glomerulonephritis. Smudgy mesangial deposits of IgG (1+), C1q (2+), and κ and λ light chain (trace) were noted on IF studies. IgG4 (3+) and IgG2 (trace) were identified on IgG subtype studies. Ultrastructural studies showed characterisitic fibrils of FGN in the mesangium and peripheral capillary walls. DNAJB9 staining was noted in glomeruli (involving mesangium and peripheral capillary walls) and focally involving peritubular capillaries and tubular basement membranes. A diagnosis of recurrent FGN (Fig 1) was made. Features of early transplant glomerulopathy were noted and a diagnosis of concomitant chronic antibody-mediated rejection was made. No changes were made to immunosuppressive therapy. She had a progressive decline in kidney function with nephrotic-range proteinuria over 3 years. She ultimately received a preemptive second kidney transplant 13.5 years after the first transplant. Two protocol biopsies at 4 months and 2 years showed no evidence of recurrent FGN, as determined by negative ultrastructural studies, IF, and DNAJB9 immunohistochemistry. Features of chronic antibody-mediated rejection were noted on the 2-year protocol biopsy. At last follow-up (38 months), the patient had a creatinine level of 0.8 mg/dL, iothalamate clearance of 65 mL/min, inactive urinary sediment, and proteinuria with protein excretion of 308 mg/d. She remained on a steroid-free regimen with tacrolimus and MMF.
      Figure thumbnail gr1
      Figure 1The 10-year allograft biopsy of patient 1. (A) The glomerulus shows mild mesangial sclerosis with glomerulitis (periodic acid–Schiff stain; original magnification, ×400). (B) DnaJ homolog subfamily B member 9 (DNAJB9) stain shows characteristic smudgy extracellular staining of the mesangium and segmental peripheral capillary walls; also noted is staining of the tubular basement membranes (original magnification, ×400).
      Patient 2 presented with proteinuria (protein excretion, 3.4 g/d) and rapid decline in kidney function. Kidney biopsy showed a membranoproliferative pattern of injury and smudgy glomerular deposits of IgG (3+), C3, C1q, and κ and λ light chains in glomeruli. Fibrils were noted in glomeruli and DNAJB9 staining supported the diagnosis of native FGN. She received a deceased donor kidney transplant and was maintained on triple immunosuppressive therapy with tacrolimus, MMF, and prednisone. Protocol biopsies at 4 months and 1, 2, and 5 years had no diagnostic abnormality. The 10-year protocol biopsy showed a mesangioproliferative pattern of injury. Smudgy mesangial staining with IgG (1-2+), C3 (1-2+), and λ light chain (1+) was noted. Sparse granular mesangial deposits of IgA (1+) were seen. Kappa light chain was not observed to be present. Fibrils in mesangium and segmentally in glomerular basement membrane were noted on EM. A few mesangial deposits were seen suggestive of a concurrent mild IgA nephropathy. DNAJB9 was mostly noted along glomerular capillary walls with sparse mesangial staining. A few peritubular capillaries were positive (Fig 2). Retrospective staining of the 5-year protocol allograft biopsy specimen showed no DNAJB9 staining. Urinalysis at 10 years showed no hematuria and albumin-creatinine ratio was 158 mg/g. No changes were made to immunosuppression. At follow-up 13 years posttransplantation, kidney function remained stable with no hematuria or significant proteinuria (protein excretion, 98 mg/d).
      Figure thumbnail gr2
      Figure 2The 10-year allograft biopsy of patient 2, demonstrating (A) mild mesangial sclerosis and (B) segmental mesangial hypercellularity (A: periodic acid–Schiff and B: Jones methenamine silver; original magnification, ×400). (C) DnaJ homolog subfamily B member 9 (DNAJB9) stain of the 10-year allograft biopsy demonstrates only sparse staining of the glomerular capillary walls. In contrast, (D) DNAJB9 staining of the native biopsy shows more abundant staining of the glomerulus. Focal tubular basement membrane staining is noted.
      Patient 3 presented with kidney failure secondary to FGN. The native biopsy showed characteristic findings of FGN with positive DNAJB9 staining. He underwent living related kidney transplantation and was maintained on triple immunosuppressive therapy with tacrolimus, MMF, and prednisone. Protocol biopsies performed at 4 months and 1, 2, and 5 years showed no significant diagnostic pathology. The 10-year protocol biopsy was notable for arteriolar hyalinosis with mild mesangial hypercellularity. The initial IF sample was insufficient. The biopsy showed positive DNAJB9 staining in glomeruli, peritubular capillaries, and tubular basement membranes (Fig 3). Fibrils were identified focally within the mesangium. At that time, urinalysis showed no hematuria and urine protein excretion was 240 mg/d. The 5-year protocol biopsy was retrospectively stained for DNAJB9 and showed positivity along a rare tubular basement membrane and a very small focus along a glomerular capillary wall (Fig 4). The IF slides (routine and pronase-treated) were negative and EM showed very rare fibrils suggestive of minimal and early FGN. No change was made to his immunosuppressive treatment. Follow-up urine protein excretion was 75 mg/d 11 years posttransplantation with stable kidney function and no evidence of active urinary sediment.
      Figure thumbnail gr3
      Figure 3The 5-year allograft biopsy specimen shows very minimal DnaJ homolog subfamily B member 9 (DNAJB9) staining of (A) a glomerular peripheral capillary loop (arrow) and (B) tubular basement membranes (arrow). The 10-year allograft biopsy specimen shows (C) mesangial sclerosis (periodic acid–Schiff stain; original magnification, ×400) and (D) extensive DNAJB9 staining of the glomeruli, peritubular capillaries, and tubular basement membranes (original magnification, ×200).
      Figure thumbnail gr4
      Figure 4The native biopsy of the patient with donor-derived fibrillary glomerulonephritis shows (A) a focal segmental glomerulosclerosis lesion (periodic acid–Schiff [PAS]; original magnification, ×200) and (B) no staining with DnaJ homolog subfamily B member 9 (DNAJB9; original magnification, ×200). The time zero biopsy specimen shows (C) relatively unremarkable glomeruli (PAS; original magnification, ×100) and (D) DNAJB9 staining predominantly of tubular basement membranes, glomeruli, and peritubular capillaries (original magnification, ×100). The 6-year allograft biopsy specimen shows (E) mild mesangial sclerosis (PAS; original magnification, ×400) and (F) persistent DNAJB9 staining of the glomeruli, largely limited to mesangium, without significant tubular basement membrane staining (original magnification, ×400).
      The 5 patients excluded from the study included a donor-derived FGN and 4 FGN mimics. Because DNAJB9 immunohistochemistry allowed us to accurately evaluate their biopsy specimens, a brief summary of the excluded cases is reported next.

      Donor-Derived FGN

      The patient presented in his early 20s with kidney failure and severe proteinuria (protein excretion, 20 g/d). The native kidney biopsy showed lesions of focal segmental glomerulosclerosis. Fibrils were not identified on ultrastructural studies and DNAJB9 staining was negative (Fig 4). A native biopsy diagnosis of primary focal segmental glomerulosclerosis was made and he received rituximab and plasmapheresis before receiving a living related kidney transplant and was maintained on MMF, tacrolimus, and prednisone therapy. His posttransplantation course was relatively unremarkable with a baseline creatinine level of 1.8 mg/dL and no proteinuria. The time zero and 4-month protocol biopsies showed mild mesangial expansion with segmental proliferative changes; however, IF and EM had not been performed. The 1-year protocol biopsy showed similar mild mesangioproliferative changes and IF showed smudgy deposits of IgG and κ and λ light chains. Electron microscopy showed characteristic fibrils of FGN. DNAJB9 was positive, supporting the diagnosis of FGN. The time zero biopsy was retrospectively stained for DNAJB9 and was positive, confirming the diagnosis of donor-derived FGN. Six years posttransplantation, the allograft shows persistent mesangial expansion with minimal proliferation, fibrils on EM, polyclonal immunoglobulin reactivity, and DNAJB9 in glomeruli (restricted to the mesangium), with focal staining of tubular basement membrane (Table 3).
      Table 3Kidney Pathology Characteristics of Donor FGN
      BiopsyDiagnosis and LM FindingsSclerosed GlomeruliIFEMDNAJB9
      NativeFSGS with 70% global glomerulosclerosis15 of 22 totalSegmental IgM, C3, C1qNo fibrilsNegative
      Time-zeroMinor morphologic abnormalities0 of 14 totalNDNDPositive GBM: focal; TBM: extensive; PTC: extensive
      4 moFocal tubular vacuolization1 of 16 totalNDNDND
      1 yFGN (mesangial hypercellularity)2 of 9 totalSmudgy to semi-linear staining of mesangium and PCW with IgG (2+), C3 (1+), κ and λ light chain (1+), C4d (2+)Randomly oriented fibrils with mean diameter of 17 (range, 13-23) nm; similar fibrils also segmentally thicken the lamina densa of the GBM (seen predominantly in outer lamina densa)Positive GBM: focal; TBM: extensive; PTC: extensive
      2 yFGN (mesangial hypercellularity)4 of 20 totalMesangial staining for IgG (2+), C3 (1+), C4d (2+), κ (2+) and λ (2+) light chainRandomly oriented nonbranching fibrils in glomeruliND
      6 yFGN (mesangial sclerosis)5 of 12 totalSmudgy mesangial staining with IgG (1+), κ (trace) and λ (trace to 1+) light chain; focal TBM staining; IgG4 subtype stainingRandomly oriented fibrils with mean thickness of 11 (range, 9-15) nm; similar fibrils segmentally involve and thicken the GBMPositive GBM, focal TBM
      Abbreviations: EM, electron microscopy; FGN, fibrillary glomerulonephritis; FSGS, focal segmental glomerulosclerosis; GBM, glomerular basement membrane; IF, immunofluorescence; IgM, immunogobulin M; LM, light microscopy; ND, not done; PCW, peripheral capillary wall; PTC, peritubular capillary; TBM, tulular basement membrane.
      In regard to donor outcome, the donor creatinine level increased from the predonation value of 0.6 mg/dL to 1 mg/dL immediately postdonation and has remained at 1 mg/dL 1 year postdonation with follow-up urine albumin excretion of 19 mg/d.

      Mimics of FGN

      Clinicopathologic details of the 4 FGN mimics are detailed in Table S1. Each of these patients had fibrils in glomeruli. The native biopsy of patient 1 showed thrombotic microangiopathy in the setting of IgG λ paraprotein. Polyclonal IgG deposits in glomeruli (characteristic of FGN) were absent. Short fine fibrils within areas of subendothelial lucency were seen. Neither punctate nor powdery deposits of monoclonal immunoglobulin deposition disease were identified. The allograft biopsy at 10 years showed similar findings of thrombotic microangiopathy with subendothelial fibrils. Negative IF and negative DNAJB9 staining of the allograft biopsy excluded FGN.
      Patient 2 had thrombotic microangiopathy in the setting of chronic myelomonocytic leukemia and elevated λ free light chains. The thrombotic microangiopathy recurred in the allograft. DNAJB9 was negative in both native and allograft biopsy specimens. Additionally, no typical morphologic criteria to diagnose FGN were noted.
      Patient 3 had vague noncongophilic fibrils in glomerular capillary walls noted in both the native and allograft biopsy specimens. There was no polytypic IgG staining. DNAJB9 staining of the native biopsy specimen was negative. These findings excluded a diagnosis of FGN. The findings were in the setting of elevated serum λ light chains.
      Patient 4 was an unusual case of DNAJB9-negative heavy chain FGN and has been described previously.
      • Nasr S.H.
      • Vrana J.A.
      • Dasari S.
      • et al.
      DNAJB9 is a specific immunohistochemical marker for fibrillary glomerulonephritis.
      All 4 patients had recurrent disease in the allograft with a median time of recurrence from tranplantation of 2.06 (range, 0.08-3.83) years. Two patients lost their grafts to recurrence.

      Discussion

      In this case series that incorporated protocol biopsies and DNAJB9 staining, we have demonstrated that native kidney FGN does not usually portend a poor transplantation outcome and that early recurrence after transplantation is likely uncommon. In FGN, the presence of a pretransplantation monoclonal gammopathy does not necessarily imply early recurrence and poor allograft outcome.
      Based on this study, patients who undergo transplantation with native FGN appear to have a clinically indolent course. Histologic recurrence may occur at 5 years and beyond. In this study, recurrence was histologically noted on 5- and 10-year protocol biopsies with a median posttransplantation follow-up time of 14.2 years. It is important to note that the nonrecurent FGN cohort had shorter follow-up, suggesting that recurrence might be a function of time posttransplantation. In our case series, 6 of 14 patients died at a median of 67.6 and mean of 69.4 years. The cause of death was attributed to malignancy in half the cases (3/6). Five patients had malignancy diagnosed posttransplantation; this finding is not entirely surprising given that the association of malignancy (carcinomas and hematologic malignancies) with native FGN has been well documented.
      • Nasr S.H.
      • Valeri A.M.
      • Cornell L.D.
      • et al.
      Fibrillary glomerulonephritis: a report of 66 cases from a single institution.
      A total of 5 graft losses were noted during the median follow-up of 5.7 years. Causes of graft loss included surgical complication, rejection, infection, and interstitial fibrosis with tubular atrophy, all of which are well-recognized causes of graft loss.
      • El-Zoghby Z.M.
      • Stegall M.D.
      • Lager D.J.
      • et al.
      Identifying specific causes of kidney allograft loss.
      Unfortunately, the small number of patients in our series limits the ability to extrapolate rates of graft loss because unexpected events (such as vascular thrombosis) have a relatively big impact on these outcomes. Our study also shows that early detection of FGN in the allograft would be unusual for FGN recurrence and if present should raise the possibility of donor disease.
      Our study demonstrates that recurrent FGN, at least when first detected in the allograft, may be clinically silent: graft function at the time of diagnosis of recurrent FGN is stable, and patients with recurrent FGN typically have minimal proteinuria and no significant hematuria at the time of recurrence. In 2 of the 3 patients with recurrent FGN, recurrence was diagnosed on protocol biopsy with preserved kidney allograft function and no significant proteinuria at the time of diagnosis, as well as on follow up. In the third patient, the diagnosis was made 10 years posttransplantation on a clinical biopsy performed for significant proteinuria and kidney allograft dysfunction. This biopsy showed concurrent transplant glomerulopathy, emphasizing that concurrent disease may not only be present with recurrent FGN but may be an important contributor to graft dysfunction.
      Of note, maintenance immunosuppression was similar in patients and did not change significantly over time. Specifically, no change in immunosuppression regimen was made in cases with recurrent FGN.
      One of the strengths of our case series is the presence of protocol biopsies, which allowed for the detection of subclinical FGN (ie, before clinical manifestations of proteinuria, hematuria, and/or glomerular filtration rate decline). Patients with native FGN are typically identified at a much later stage once clinical manifestations are identified, with more advanced pathology. This can in part explain the better prognosis noted in our case series (ie, lead time bias). Another plausible hypothesis for the better prognosis is the effect of maintenance immunosuppression on modulating the disease progression.
      To define the risk for recurrence, accurate diagnosis of FGN in native biopsies is of paramount importance when considering a kidney transplant. DNAJB9 is detectable in native kidney FGN irrespective of polytypic or monotypic IgG staining and regardless of the presence of monoclonal proteins in serum.
      • Andeen N.K.
      • Troxell M.L.
      • Riazy M.
      • et al.
      Fibrillary glomerulonephritis: clinicopathologic features and atypical cases from a multi-institutional cohort.
      ,
      • Nasr S.H.
      • Vrana J.A.
      • Dasari S.
      • et al.
      DNAJB9 is a specific immunohistochemical marker for fibrillary glomerulonephritis.
      Given the high sensitivity and specificity of DNAJB9 for FGN, we argue that measuring this marker, either by immunoperoxidase staining or mass spectrometry, should occur for all native biopsies in which glomerular deposits have a fibrillary substructure (including immunotactoid glomerulopathy) as an ancillary diagnostic tool to clarify the diagnosis of FGN. In a previous study from Mayo Clinic, the 5 patients with FGN did not demonstrate allograft recurrence, whereas there were 5 recurrences among 4 patients with FGN mimics.
      • Czarnecki P.G.
      • Lager D.J.
      • Leung N.
      • Dispenzieri A.
      • Cosio F.G.
      • Fervenza F.C.
      Long-term outcome of kidney transplantation in patients with fibrillary glomerulonephritis or monoclonal gammopathy with fibrillary deposits.
      Almost 10 years later, the use of DNAJB9 in our study allowed us to re-evaluate the biopsy specimens of those 4 specific FGN mimics. The negative DNAJB9 staining and the IF staining pattern confirmed that they were not cases of FGN.
      • Nasr S.H.
      • Sirac C.
      • Bridoux F.
      • et al.
      Heavy chain fibrillary glomerulonephritis: a case report.
      The prognosis for patients with circulating monoclonal gammopathy and organized fibrillary glomerular deposits that are DNAJB9 negative may be dismal, with allograft recurrence and loss.
      We emphasize that the 3 patients with FGN in this study with monoclonal gammopathy had MGUS and not monoclonal gammopathy of renal significance (MGRS).
      • Leung N.
      • Nasr S.H.
      • Sethi S.
      How I treat amyloidosis: the importance of accurate diagnosis and amyloid typing.
      This monoclonal protein was not involved in the FGN as indicated by the IF staining, which showed polyclonal IgG. Unlike in MGRS (such as Proliferative Glomerulonephritis with Monoclonal Immunoglobulin Deposits, monoclonal immunoglobulin deposition disease, and C3 glomerulopathy with monoclonal gammopathy) in which the risk for recurrence and graft lost is high, the 3 patients with FGN with MGUS were not at increased risk for early recurrence.
      • Said S.M.
      • Cosio F.G.
      • Valeri A.M.
      • et al.
      Proliferative glomerulonephritis with monoclonal immunoglobulin G deposits is associated with high rate of early recurrence in the allograft.
      • Zand L.
      • Lorenz E.C.
      • Cosio F.G.
      • et al.
      Clinical findings, pathology, and outcomes of C3GN after kidney transplantation.
      • Leung N.
      • Lager D.J.
      • Gertz M.A.
      • Wilson K.
      • Kanakiriya S.
      • Fervenza F.C.
      Long-term outcome of renal transplantation in light-chain deposition disease.
      However, if the monoclonal gammopathy was involved in the pathogenesis of FGN (monotypic deposits), perhaps early recurrence could occur as seen in other MGRS-related diseases.
      • Leung N.
      • Nasr S.H.
      • Sethi S.
      How I treat amyloidosis: the importance of accurate diagnosis and amyloid typing.
      ,
      • Leung N.
      • Lager D.J.
      • Gertz M.A.
      • Wilson K.
      • Kanakiriya S.
      • Fervenza F.C.
      Long-term outcome of renal transplantation in light-chain deposition disease.
      DNAJB9 is an excellent marker of FGN and its utility in the evaluation of the allograft in those with native disease of FGN has several benefits. First, it allows detection of early FGN, when the deposits in glomeruli are sparse and small foci of fibrils can be missed on EM.
      • Nasr S.H.
      • Dasari S.
      • Lieske J.C.
      • et al.
      Serum levels of DNAJB9 are elevated in fibrillary glomerulonephritis patients.
      ,
      • Dasari S.
      • Alexander M.P.
      • Vrana J.A.
      • et al.
      DnaJ heat shock protein family B member 9 is a novel biomarker for fibrillary GN.
      Second, it allows distinction from diabetic mesangial fibrillosis. Ten-year allograft biopsies often show mesangial sclerosis and evidence of diabetic glomerulosclerosis, which could make the detection of recurrent FGN challenging.
      • Stegall M.D.
      • Cornell L.D.
      • Park W.D.
      • Smith B.H.
      • Cosio F.G.
      Renal allograft histology at 10 years after transplantation in the tacrolimus era: evidence of pervasive chronic injury.
      ,
      • Kronz J.D.
      • Neu A.M.
      • Nadasdy T.
      When noncongophilic glomerular fibrils do not represent fibrillary glomerulonephritis: nonspecific mesangial fibrils in sclerosing glomeruli.
      The sensitivity of the immunohistochemistry stain for DNAJB9 helps in this distinction. In all the biopsies with recurrence detected at 10 years, DNAJB9 deposits were concurrent with immune complex deposition. In the 1 case with minimal DNAJB9 detected on the 5-year protocol biopsy, there were no concurrent immune complex deposits detected by IF (routine and pronase-treated), but EM showed few fibrillary deposits. This observation might reflect the increased sensitivity of DNAJB9 as compared to IF, or sampling bias, but might also suggest that DNAJB9 is the fibrillary autoantigen, which when deposited initiates an immune reaction. If the latter were true, DNAJB9 would be the defining criterion for FGN. However, additional studies are needed to arrive at this conclusion.
      This study highlights the utility of protocol biopsies and surveillance with DNAJB9 immunohistochemistry on the allograft biopsy of patients with FGN, analogous to staining for PLA2R antigen when monitoring for recurrent PLA2R-associated membranous nephropathy.
      • Larsen C.P.
      • Walker P.D.
      Phospholipase A2 receptor (PLA2R) staining is useful in the determination of de novo versus recurrent membranous glomerulopathy.
      Table 4 outlines a proposal for the utility of DNAJB9 in the evaluation and monitoring of patients with FGN, pre- and posttransplantation.
      Table 4Proposed Use and Utility of DNAJB9 Staining for Kidney Transplantation in FGN
      TimingUtility of DNAJB9 StainConsiderations
      PretransplantationConfirms diagnosis of FGNHelpful when FGN mimic is present or diagnosis is unclear
      Time zeroIdentifies donor-derived FGNConsider when de novo FGN is identified in kidney allograft at or before 2 y to rule out donor-derived FGN
      Protocol biopsyIdentifies subclinical FGNGuides clinician vigilance in monitoring for clinical recurrence (eg, more frequent urinalysis and albuminuria measurement)
      For-cause biopsyConfirms recurrent FGNWhen there is allograft dysfunction, hematuria, proteinuria
      Abbreviations: DNAJB9, DnaJ homolog subfamily B member 9; FGN, fibrillary glomerulonephritis.
      The strength of our study is the presence of long-term protocol biopsies, which permits us to hypothesize as to the pathogenesis of allograft DNAJB9 FGN. The earliest stage of allograft FGN is characterized by deposition of DNAJB9 in peripheral glomerular capillary walls and more abundantly in the peritubular capillaries and tubular basement membranes. This localization pattern suggests that the source of DNAJB9 is extrarenal. Another pointer to an extrarenal source of FGN is the failure of histologic progression in the donor FGN. Additionally, the recurrence of FGN in the allograft supports the presence of an extrarenal source of DNAJB9.
      The main limitation is the small number of cases and of recurrence and the relatively short follow-up postrecurrence, which limited our ability to perform any survival analysis or competing-risk analysis between the recurrent and nonrecurrent groups.
      In conclusion, patients with native kidney FGN in our series have had a low rate of recurrent disease posttransplantation, at least through the first 5 posttransplantation years. If the disease recurs, its clinical and histologic expression are mild and recurrence appears independant of circulating monoclonal proteins. The allograft evaluation using protocol biopsies and staining with DNAJB9 allows the unique opportunity to study the development and progression of FGN.

      Article Information

      Authors’ Full Names and Academic Degrees

      Mireille El Ters, MD, Shane A. Bobart, MD, Lynn D. Cornell, MD, Nelson Leung, MD, Andrew Bentall, MD, Sanjeev Sethi, MD, PhD, Mary Fidler, MD, Joseph Grande, MD, PhD, Loren Herrera Hernandez, MD, Fernando G. Cosio, MD, PhD, Ladan Zand, MD, Hatem Amer, MD, Fernando C. Fervenza, MD, PhD, Samih H. Nasr, MD, and Mariam P. Alexander, MD.

      Authors’ Contributions

      Research idea and study design: MET, MPA; data acquisition: MET, SAB, MPA, AB, FCF, FGC; data analysis/interpretation: all authors; statistical analysis: MET. Each author contributed important intellectual content during manuscript drafting or revision and and agrees to be personally accountable for the individual’s own contributions and to ensure that questions pertaining to the accuracy or integrity of any portion of the work, even one in which the author was not directly involved, are appropriately investigated and resolved, including with documentation in the literature if appropriate.

      Support

      Funding for this project was made possible by Mayo Clinic Anatomic Pathology funds.

      Financial Disclosure

      The authors declare that they have no relevant financial interests.

      Acknowledgements

      The authors acknowledge Tami Schmidt and Lisa Hines for incredible administrative assistance.

      Peer Review

      Received August 13, 2019. Evaluated by 3 external peer reviewers, with direct editorial input from a Pathology Editor, an Associate Editor, and the Editor-in-Chief. Accepted in revised form January 26, 2020.

      Supplementary Material

      References

        • Fogo A.
        • Qureshi N.
        • Horn R.G.
        Morphologic and clinical features of fibrillary glomerulonephritis versus immunotactoid glomerulopathy.
        Am J Kidney Dis. 1993; 22: 367-377
        • Nasr S.H.
        • Valeri A.M.
        • Cornell L.D.
        • et al.
        Fibrillary glomerulonephritis: a report of 66 cases from a single institution.
        Clin J Am Soc Nephrol. 2011; 6: 775-784
        • Rosenstock J.L.
        • Markowitz G.S.
        • Valeri A.M.
        • Sacchi G.
        • Appel G.B.
        • D'Agati V.D.
        • et al.
        Fibrillary and immunotactoid glomerulonephritis: distinct entities with different clinical and pathologic features.
        Kidney Int. 2003; 63: 1450-1461
        • Andeen N.K.
        • Troxell M.L.
        • Riazy M.
        • et al.
        Fibrillary glomerulonephritis: clinicopathologic features and atypical cases from a multi-institutional cohort.
        Clin J Am Soc Nephrol. 2019; 14: 1741-1750
        • Iskandar S.S.
        • Falk R.J.
        • Jennette J.C.
        Clinical and pathologic features of fibrillary glomerulonephritis.
        Kidney Int. 1992; 42: 1401-1407
        • Pronovost P.H.
        • Brady H.R.
        • Gunning M.E.
        • Espinoza O.
        • Rennke H.G.
        Clinical features, predictors of disease progression and results of renal transplantation in fibrillary/immunotactoid glomerulopathy.
        Nephrol Dial Transplant. 1996; 11: 837-842
        • Alpers C.E.
        • Rennke H.G.
        • Hopper Jr., J.
        • Biava C.G.
        Fibrillary glomerulonephritis: an entity with unusual immunofluorescence features.
        Kidney Int. 1987; 31: 781-789
        • Brady H.R.
        Fibrillary glomerulopathy.
        Kidney Int. 1998; 53: 1421-1429
        • Samaniego M.
        • Nadasdy G.M.
        • Laszik Z.
        • Nadasdy T.
        Outcome of renal transplantation in fibrillary glomerulonephritis.
        Clin Nephrol. 2001; 55: 159-166
        • Czarnecki P.G.
        • Lager D.J.
        • Leung N.
        • Dispenzieri A.
        • Cosio F.G.
        • Fervenza F.C.
        Long-term outcome of kidney transplantation in patients with fibrillary glomerulonephritis or monoclonal gammopathy with fibrillary deposits.
        Kidney Int. 2009; 75: 420-427
        • Mallett A.
        • Tang W.
        • Hart G.
        • et al.
        End-stage kidney disease due to fibrillary glomerulonephritis and immunotactoid glomerulopathy - outcomes in 66 consecutive ANZDATA registry cases.
        Am J Nephrol. 2015; 42: 177-184
        • El-Zoghby Z.M.
        • Stegall M.D.
        • Lager D.J.
        • et al.
        Identifying specific causes of kidney allograft loss.
        Am J Transplant. 2009; 9: 527-535
        • Nasr S.H.
        • Vrana J.A.
        • Dasari S.
        • et al.
        DNAJB9 is a specific immunohistochemical marker for fibrillary glomerulonephritis.
        Kidney Int Rep. 2017; 3: 56-64
        • Nasr S.H.
        • Sirac C.
        • Bridoux F.
        • et al.
        Heavy chain fibrillary glomerulonephritis: a case report.
        Am J Kidney Dis. 2019; 74: 276-280
        • Leung N.
        • Nasr S.H.
        • Sethi S.
        How I treat amyloidosis: the importance of accurate diagnosis and amyloid typing.
        Blood. 2012; 120: 3206-3213
        • Said S.M.
        • Cosio F.G.
        • Valeri A.M.
        • et al.
        Proliferative glomerulonephritis with monoclonal immunoglobulin G deposits is associated with high rate of early recurrence in the allograft.
        Kidney Int. 2018; 94: 159-169
        • Zand L.
        • Lorenz E.C.
        • Cosio F.G.
        • et al.
        Clinical findings, pathology, and outcomes of C3GN after kidney transplantation.
        J Am Soc Nephrol. 2014; 25: 1110-1117
        • Leung N.
        • Lager D.J.
        • Gertz M.A.
        • Wilson K.
        • Kanakiriya S.
        • Fervenza F.C.
        Long-term outcome of renal transplantation in light-chain deposition disease.
        Am J Kidney Dis. 2004; 43: 147-153
        • Nasr S.H.
        • Dasari S.
        • Lieske J.C.
        • et al.
        Serum levels of DNAJB9 are elevated in fibrillary glomerulonephritis patients.
        Kidney Int. 2019; 95: 1269-1272
        • Dasari S.
        • Alexander M.P.
        • Vrana J.A.
        • et al.
        DnaJ heat shock protein family B member 9 is a novel biomarker for fibrillary GN.
        J Am Soc Nephrol. 2018; 29: 51-56
        • Stegall M.D.
        • Cornell L.D.
        • Park W.D.
        • Smith B.H.
        • Cosio F.G.
        Renal allograft histology at 10 years after transplantation in the tacrolimus era: evidence of pervasive chronic injury.
        Am J Transplant. 2018; 18: 180-188
        • Kronz J.D.
        • Neu A.M.
        • Nadasdy T.
        When noncongophilic glomerular fibrils do not represent fibrillary glomerulonephritis: nonspecific mesangial fibrils in sclerosing glomeruli.
        Clin Nephrol. 1998; 50: 218-223
        • Larsen C.P.
        • Walker P.D.
        Phospholipase A2 receptor (PLA2R) staining is useful in the determination of de novo versus recurrent membranous glomerulopathy.
        Transplantation. 2013; 95: 1259-1262