Clinical Findings, Pathology, and Outcomes of C3GN after Kidney Transplantation

CLINICAL RESEARCH www.jasn.org Clinical Findings, Pathology, and Outcomes of C3GN after Kidney ation Ladan Zand,* Elizabeth C. Lorenz,* Fernando G. Cosio,* Fernando C. Fervenza,* Samih H. Nasr, Manish J. Gandhi, Richard J. H. Smith, and Sanjeev Sethi *Department of Internal Medicine, Division of Nephrology and Hypertension, and Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, Minnesota; and Otolaryngology and Renal Research Laboratories, Departments of Internal Medicine and Pediatrics, Division of Nephrology, Carver College of Medicine, Iowa City, Iowa ABSTRACT C3 glomerulonephritis (C3GN) results from abnormalities in the alternative pathway of complement, and it is characterizedbydepositionofc3withabsentorscant Igdeposition. In many patients, C3GN progresses to ESRD. The clinical features, pathology, and outcomes of patients with C3GN receiving kidney transplantation are unknown. Between 1996 and 2010, we identified 21 patients at our institution who received a kidney transplant because of ESRD from C3GN. The median age at the time of initial diagnosis of C3GN at kidney biopsy was 20.8 years. The median time from native kidney biopsy to dialysis or transplantation was 42.3 months. Of 21 patients, 14 (66.7%) patients developed recurrent C3GN in the allograft. The median time to recurrence of disease was 28 months. Graftfailure occurred in 50% of patients with recurrent C3GN, with a median time of 77 months to graft failure post-transplantation. The remaining 50% of patients had functioning grafts, with a median follow-up of 73.9 months. The majority of patients had hematuria and proteinuria at time of recurrence. Three (21%) patients were positive for monoclonal gammopathy and had a faster rate of recurrence and graft loss. Kidney biopsy at the time of recurrence showed mesangial proliferative GN in eight patients and membranoproliferative GN in six patients. All allograft kidney biopsies showed bright C3 staining (2 3+), with six biopsies also showing trace/1+ staining for IgM and/or IgG. To summarize, C3GN recurs in 66.7% of patients, and one half of the patients experience graft failure caused by recurrence. J Am Soc Nephrol 25: 1110 1117, 2014. doi: 10.1681/ASN.2013070715 C3 glomerulonephritis (C3GN) is the result of dysregulation in the alternative pathway of complement. 1,2 The renal biopsy is characterized by dominant staining for complement factor C3 with minimal or no staining for Igs on immunofluorescence microscopy. Light microscopy reveals a proliferative pattern of injury with evidence of mesangial, subendothelial, and, occasionally, subepithelial deposits on electron microscopy. 3,4 Clinically, the disease presents with hematuria, proteinuria, and low serum C3 levels. 2 The renal outcome of patients with C3GN is not as well established. The disease seems to have a chronic, but progressive, course, with deterioration in renal function and development of ESRD in up to 50% of the patients. 5 Even less well established is the course of C3GN after kidney transplantation. The aim of this study was to examine the outcome of kidney transplantation in patients with C3GN, further characterize the histology of recurrent C3GN (rc3gn), and compare the results with recurrent dense deposit disease (DDD). RESULTS Baseline Demographics and Course of Non-rC3GN Twenty-one patients at our institution received a kidney transplant because of ESRD from C3GN. The median age at time of initial diagnosis of C3GN at kidney biopsy was 20.8 years (range=7 67 years, mean=26.9 years). The median time from native kidney biopsy to dialysis or transplantation was 42.3 months (range=0 400, mean=70.8 months). Received July 11, 2013. Accepted October 29, 2013. Published online ahead of print. Publication date available at www.jasn.org. Correspondence: Dr. Sanjeev Sethi, Division of Anatomic Pathology, Department of Laboratory Medicine and Pathology, Mayo Clinic, MN 55905. Email: Sethi.sanjeev@mayo.edu Copyright 2014 by the American Society of Nephrology 1110 ISSN : 1046-6673/2505-1110 JAmSocNephrol25: 1110 1117, 2014

www.jasn.org CLINICAL RESEARCH Of 21 patients with C3GN who received kidney transplantation, 7 patients did not have recurrence of disease in their renal allograft after a median post-transplant follow-up of 162 months (27 361 months, mean=174 months) (Table 1). Protocol biopsies done at 0, 4, 12, 24, 60, and 120 months did not show any evidence of rc3gn. Of 7 patients, 6 patients were men, and one patient was a woman. The median age at time of native biopsy was 16.4 years (7 54 years, mean=21.4 years). Median time to developing ESRD or receiving kidney transplantation was 32 months (1 185 months, mean=45.5 months). Four patients received dialysis before transplantation, and three patients received preemptive transplants. All transplants, except for one, were from living donors. Induction immunosuppression therapy included thymoglobulin in one patient, anti-cd25 antibody in one patient, and antilymphocyte globulin in one patient. Four patients did not receive induction therapy. Two of seven patients received more than one kidney transplant, none of which failed because of recurrence of disease. Of seven patients, four patients continue to have functioning grafts at time of last follow-up, and three patients have developed allograft failure. Two of the graft losses were caused by rejection, and one graft was lost because of interstitial fibrosis and tubular atrophy after an episode of acute urinary tract infection and urinary obstruction. Average serum creatinine at time of last follow-up in patients with a functioning graft was 1.6 mg/dl (1.3 1.8 mg/dl). Of five patients with available data, none had low C3 before renal transplantation. Two patients had C3 levels evaluated after transplant (at the time of graft failure), and both patients were within normal range. Only one patient had abnormal serum monoclonal studies (IgGk). The patient continues to have a functioning graft more than 2.5 years after renal transplantation. Maintenance immunosuppression is shown in Table 1. Baseline Demographics and Pretransplant Course of rc3gn Of 21 patients with the diagnosis of C3GN based on the native kidney biopsy, 14 (66.7%) patients developed rc3gn (Figure 1). Except for one patient (patient 12) with positive hepatitis B serology, the remaining patients with rc3gn were negative for hepatitis B and C serology. Of 14 patients with rc3gn, 8 (57%) patients were women, and 6 (43%) patients were men. The median age at time of first native kidney biopsy was 24.4 years (11 67 years, mean=29.7 years). The median age at time of first renal transplantation was 35.5 years (12 69 years, mean=37.8 years). The median time from native kidney biopsy to dialysis or first transplantation was 62.4 months (range=0 400 months, mean 83.5 months). The median 24-hour urine protein before renal transplantation was 6943 mg (334 9596 mg/24 h, mean=5803 mg/24 h). The median C3 level was 46.5 mg/dl (mean=49.1 mg/dl, normal range=75 175 mg/dl), and the median C4 level was 21.5 mg/dl (mean=18.9 mg/dl, normal range=14 40 mg/dl) (Table 2). Of 14 patients, 10 patients were on dialysis before transplantation, and 4 patients received preemptive transplants; 11 of 14 transplants with rc3gn were from living donors, and 3 transplants were from deceased donors. Before transplantation, 6 patients did not receive any treatment for C3GN, 3 patients received prednisone alone, one patient received prednisone and cyclosporine, one patient received prednisone Figure 1. Incidence of recurrent C3GN. Kaplan Meier plot of the cumulative incidence of rc3gn after kidney transplantation. Table 1. Baseline patient demographics and clinical evaluation in patients with non-rc3gn before and after renal transplantation Patient Age (yr) at First Sex Type of Failed Time to Failure (mo) Hematuria at Last Follow-Up 24-h Protein at Last Follow-Up (mg) C3 Post- Immunosuppression 1 25 M Living No NA No 539 ND Tac, sirolimus, Pred 2 14 M Living No NA No 84 ND Cyclosporin, MMF, Pred 3 58 M Living No NA No 71 ND Tac, MMF, Pred 4 20 M Living Yes 330 Yes 5206 116 (N) Azathioprine, Pred 5 20 M Living Yes 185 No 5720 114 (N) Azathioprine, Pred 6 20 M Living No NA No 208 ND Cyclosporin, MMF, Pred 7 25 W Deceased Yes 27 Yes 1638 ND Sirolimus, Pred C3 normal range is 75 175 mg/dl. M, man; NA, not applicable; ND, not done; Tac, tacrolimus; Pred, prednisone; MMF, mycophenolate mofetil; N, normal; W, woman. J Am Soc Nephrol 25: 1110 1117, 2014 Recurrent C3GN 1111

CLINICAL RESEARCH www.jasn.org Table 2. Patient Baseline patient demographics and clinical evaluation in patients with rc3gn before renal transplantation Sex Age (yr) at Native Biopsy Time from Biopsy to ESRD or First Preemptive 24-h Protein (mg) before C3 before C4 before Treatment before 1 M 20 3.7 No 7424 26 11 Pred 2 W 12 88.9 No NA NA NA Pred 3 W 11 22.7 Yes NA NA NA NA 4 M 45 82.6 No NA 53 18 NA 5 W 23 400.3 Yes 4557 18 NA None 6 M 67 0 No NA 87 22 None 7 W 15 36.1 No 2815 55 21 None 8 W 25 42.3 No 6463 NA 11 MMF, IVIG 9 M 39 29.1 No NA NA NA None 10 W 34 109 Yes 7736 14 NA CYC, Pred 11 M 14 93.1 No 9596 NA 23 Pred 12 W 34 100.7 No NA NA NA None 13 M 19 7.5 Yes 7503 100 22 None 14 W 54 153.7 No 334 40 23 Pred, cyclosporine C3 normal range is 75 175 mg/dl, and C4 normal range is 14 40 mg/dl. M, man; Pred, prednisone; W, woman; NA, not applicable; MMF, mycophenolate mofetil; IVIG, intravenous immunoglobulin; CYC, cyclophosphamide. and cyclophosphamide, one patient received mycophenolate mofetil and intravenous Ig, and information was not available for two patients. Post- Course of rc3gn Seven patients continue to have functional grafts with a median follow-up of 73.9 months (32 162.7 months, mean=78.1 months), and 7 (50%) patients have lost their graft after recurrence of the disease (Figure 2). The median serum creatinine at time of last follow-up in those patients with a functioning graft was 1.6 mg/dl (1.0 2.1 mg/dl, mean=1.67 mg/dl) (Table 3). The median time from transplantation to recurrence of disease was 28 months (ranging from 9 days to over 11 years, mean=46 months). The median graft survival from time of transplant to failure was 77 months (0 159 months, mean=90.1 months), whereas median time from recurrence of disease to failure was 18 months (range=0 71 months, mean=21.1 months). In patients with graft failure, six grafts were lost because of rc3gn, and one graft was lost because of delayed graft function with evidence of rc3gn 9 days after kidney transplantation (patient with monoclonal gammopathy; see below). Of 14 patients with rc3gn, induction immunosuppression included thymoglobulin (8 patients), anti-cd25 antibody (1 patient), anti-cd52 antibody (1 patient), and antilymphocyte globulin (1 patient). Three patients did not receive induction therapy. Two patients received prednisone, cyclosporine, and azathioprine for maintenance immunosuppression. The remaining patients received prednisone, mycophenolate mofetil, and tacrolimus. At the time of recurrence, 6 patients had evidence of hematuria, and 5 (45%) of 11 patients with available data had Figure 2. Graft failure in rc3gn. Incidence of graft failure of rc3gn (solid line) and non-rc3gn (nrc3gn; dashed line) after kidney transplant (P=0.04, log rank). Median time to graft failure of rc3gn was 136 months, and median time to graft failure of nrc3gn was 330 months. low C3 levels. The median 24-hour urinary protein at time of recurrence was 799 mg (22 4288 mg/24 h, mean=1629.3 mg/ 24 h). Ten of the renal biopsies were performed for clinical reasons, including hematuria, proteinuria, or elevated serum creatinine; four renal biopsies were performed as protocol biopsies (patients 1, 3, 11, and 13). Of 4 patients who were incidentally found to have rc3gn, all 4 patients continue to have a functioning graft at last follow-up. Of 10 patients who had biopsies done for clinical reasons, 7 (70%) patients have lost their grafts. Five of fourteen patients with rc3gn received more than one kidney transplant. Of 5 patients, 2 patients had rc3gn in the first renal allograft, and 3 patients had recurrence of disease in the second renal allograft. 1112 Journal of the American Society of Nephrology J Am Soc Nephrol 25: 1110 1117, 2014

www.jasn.org CLINICAL RESEARCH Table 3. Patient Clinical findings in patients with rc3gn after renal transplantation Age (yr) at First Type of Time (mo) to Recurrence in Failed Time to Failure from Recurrence Hematuria at Relapse 24-h Protein at Recurrence (mg) C3 at Recurrence C4 at Recurrence 1 21 Living 6.3 No NA No 97 ND ND 2 20 Living 113.6 Yes 23 No 2975 63 (L) 19 (N) 3 12 Living 12.1 No NA No 105 ND ND 4 56 Deceased 0.3 Yes 0 Yes 941 38 (L) 8 (L) 5 56 Living 133.1 Yes 17 Yes 3824 78 (N) 13 (L) 6 68 Living 49.9 No NA Yes 68 40 (L) 21 (N) 7 19 Living 89.3 No NA Yes 4271 126 (N) 28 (N) 8 29 Living 6.1 Yes 71 No 22 93 (N) ND 9 41 Deceased 48 Yes 18 Yes 4288 53 (L) 21 (N) 10 43 Living 3.6 No NA Yes 3272 81 (N) 13 (L) 11 22 Living 3.7 No NA No 199 ND ND 12 48 Deceased 139 Yes 20 No 677 75 (N) 20 (N) 13 20 Living 3.6 No NA No 37 43 (L) 48 (N) 14 67 Living 43.4 Yes 0.3 No 2054 96 (N) 26 (N) C3 normal range is 75 175 mg/dl, and C4 normal range is 14 40 mg/dl. NA, not applicable; ND, not determined; N, normal; L, low. rc3gn and Monoclonal Gammopathy All patients had either serum or urine electrophoresis studies before transplant to determine the presence of a monoclonal gammopathy. Three patients were positive for M spike in the g-region (patients 4, 9, and 10). Patient 4 had evidence of IgAl on serum immunofixation and a positive bone marrow biopsy with 5% 10% l-restricted plasma cells. The patient had recurrence of disease within 9 days of renal transplantation. Serum immunofixation for patient 9 showed IgGl,butthebonemarrow biopsy was normal. The patient has received a total of four kidney allografts and had evidence of rc3gn in the last three transplants, all of which have been lost because of recurrence of disease. The time to recurrence of disease was much shorter in the final allograft (48, 54, and 2.5 months after the second, third, and fourth transplant, respectively). The final graft was lost in less than 1 year (8.5 months) because of recurrence of disease, which was shorter than the time to graft loss compared with the previous transplants. Patient 10 revealed IgGl on serum immunofixation studies, and a bone marrow biopsy showed 15% 20% l-restricted plasma cells. The patient had recurrence of disease within 4 months of renal transplantation, with significant decline in kidney function. Initially, the patient was treated intermittently with plasmapheresis and prednisone, and there was some improvement in the kidney function; however, kidney function deteriorated, and the patient required dialysis for a brief period of time. The patient eventually underwent autologous peripheral stem cell transplant 2 years after kidney transplantation with significant improvement in kidney function. The patient continues to have a functioning graft more than 10 years after the kidney transplant. HLA Haplotype Seven of twenty-one patients with C3GN who received kidney transplants were typed as HLA-DR17 (3)-DQ2 haplotype, and of these patients, six patients had rc3gn. Renal Pathology Review of the native kidney biopsies of 7 patients with nonrc3gn showed C3GN in the native kidney biopsy by the criteria listed in Concise Methods. The previous diagnoses of the biopsies were as follows: membranoproliferative GN (MPGN) type I in three cases, MPGN in two cases, MPGN type III in one case, and mesangial proliferative GN with C3 in onecase(table4). Review of the native kidney biopsies of 14 patients with rc3gn showed C3GN in the native kidney biopsy by the criteria listed in Concise Methods. The diagnosis included MPGN type I in five cases, MPGN type III in four cases, postinfectious GN in two cases, proliferative and sclerosing GN of unknown etiology in two cases, and diffuse proliferative GN in one case. Kidney allograft biopsy was evaluated in all cases of rc3gn. At the time of diagnosis, 8 patients showed a mesangial proliferative pattern of injury, whereas the remaining 6 patients showed a membranoproliferative pattern of injury. The number of globally sclerosed glomeruli ranged from 0% to 60%, with an average of 16%. The tubular atrophy and interstitial fibrosis ranged from 0% to 50%, with an average of 18%. Immunofluorescence microscopy showed dominant C3 staining in all cases: eight biopsies showed exclusive staining for C3 (2 3+), whereas five biopsies showed trace/1+ IgM or IgG in addition to the C3 staining. In one biopsy (patient 12), immunofluorescence material contained only two globally sclerosed glomeruli. However, mesangial and subendothelial electron dense deposits were noted on electron microscopy in this biopsy. Also, subsequent biopsies showed the dominant C3 staining. Electron microscopy showed mesangial deposits in three biopsies and mesangial and capillary wall deposits in nine biopsies. Electron microscopy was not performed in 2 patients. Figure 3 shows representative allograft biopsy findings in a case of early mesangial proliferative GN with C3 deposits (patient 3) and MPGN with C3 deposits (patient 6). J Am Soc Nephrol 25: 1110 1117, 2014 Recurrent C3GN 1113

CLINICAL RESEARCH www.jasn.org Table 4. Patient Kidney allograft biopsy findings at time of diagnosis of rc3gn Native Kidney Biopsy Diagnosis Glomeruli Sclerosed/Total Pattern of Injury Interstitial Fibrosis and Tubular Atrophy Immunofluorescence Microscopy Electron Microscopy: Location of Deposits 1 MPGN type I 1/18 Mesangial 25 C3, 3+; IgM, 1+; IgG, trace Mesangial 2 MPGN type I 4/7 MPGN 50 C3, 3+; IgM, 1 2+ No glomeruli available 3 Diffuse proliferative 0/16 Mesangial 0 C3, 2+; IgM, 1+; IgG, trace Mesangial GN 4 Postinfectious GN 2/19 Mesangial 25 C3, 3+ Mesangial, subendothelial subepithelial 5 MPGN type I 0/7 MPGN 30 C3, 2 3+; IgM, Mesangial, subendothelial 1 2+; IgG, trace 6 Postinfectious GN 1/40 MPGN 10 C3, 3+ Mesangial, subendothelial subepithelial 7 Proliferative and 0/13 Mesangial 0 C3, 2+; IgM, trace; IgG, 1+ Mesangial, few subepithelial sclerosing GN 8 MPGN type I 1/6 MPGN 0 C3, 3+ Mesangial, subendothelial 9 Proliferative and 0/16 Mesangial 5 C3, 3+ No electron microscopy sclerosing GN 10 MPGN type III 1/9 MPGN 0 C3, 3+ Mesangial, subendothelial 11 MPGN type III 0/16 Mesangial 0 C3, 3+ Mesangial, subendothelial subepithelial 12 MPGN type I 11/21 Mesangial 50 All sclerosed Mesangial, subendothelial 13 MPGN type III 3/27 Mesangial 5 C3, 3+ Mesangial 14 MPGN type III 5/8 MPGN 50 C3, 3+ Mesangial, subendothelial Treatment The majority of patients (n=10) did not receive any additional treatments for the recurrence of the disease and continued on their usual immunosuppressive therapy. Three patients (patients 7, 8, and 14) received rituximab therapy, with improvement of renal function in one patient (patient 7) and lack of response in the remaining 2 patients, which resulted in loss of graft function. As noted above, patient 10 received plasmapheresis, a higher dose of prednisone, and eventually, stem cell transplantation. Patient 9 did not receive any specific therapy for recurrence of disease in the second and third renal allografts, but underwent plasmapheresis with a higher dose of prednisone after recurrence of the disease in the fourth transplant without any response. DISCUSSION C3GN is a recently described GN that has gained recognition only over the last 5 years, with few case series describing the renal outcomes. 1,2 Some of the reports have mentioned recurrence of C3GN in few cases on the series of C3GN. However, detailed clinical features, renal biopsy findings, and renal outcomes of kidney transplant in patients with ESRD caused by C3GN are not known. This study is the first study to comprehensively evaluate the clinical findings, kidney biopsy characteristics, and renal outcomes of patients with C3GN after kidney transplantation. We identified 21 Mayo Clinic patients with ESRD caused by C3GN who received renal transplant. Fourteen of these patents had evidence of rc3gn on allograft biopsy, indicating a high rate of recurrence (66.7%). Four cases of rc3gn were identified by protocol kidney allograft biopsies, which showed C3 deposits in the absence of any clinical symptoms. Most cases, however, were identified on clinical biopsies performed for evaluation of hematuria, proteinuria, or elevated serum creatinine. Most of the patients who had evidence of recurrence based on protocol biopsy had a functioning graft at the time of last follow-up, indicating a favorable outcome. Seven patients did not have rc3gn based on clinical features and protocol biopsies, although subclinical recurrence may have been missed in this group. The median age at the time of initial diagnosis of C3GN on the native kidney biopsy was 20.8 years, and C3GN affected both sexes equally, consistent with the previous reports. 1,2 The median time to developing ESRD was over 3.5 years, suggesting a slow but progressive decline in renal function. Given that most of the patients affected are young, the majority of patients required some type of RRT in early adulthood (median age at transplantation was 36 years). In addition, some of the patients required more than one renal transplant in their lifetime. In our series, 5 patients (35%) received at least two renal transplants. C3GN has been associated with a monoclonal gammopathy. 6,7 The monoclonal Ig presumably inhibits the function of complement regulating proteins probably by acting as an autoantibody. 8 In a recent study, 31% of C3GN patients were associated 1114 Journal of the American Society of Nephrology J Am Soc Nephrol 25: 1110 1117, 2014

www.jasn.org CLINICAL RESEARCH Figure 3. Representative kidney allograft biopsies. (A C) Patient 3. (A) Light microscopy showing a mild mesangial proliferative GN (periodic acid Schiff). (B) Immunofluorescence showing mild mesangial staining for C3. (C) Electron microscopy showing mesangial electron dense deposits. (D F) Patient 6. (D) Light microscopy showing MPGN (periodic acid Schiff). (E) Immunofluorescence showing bright mesangial and capillary wall staining for C3. (F) Electron microscopy showing numerous mesangial electron dense deposits and subepithelial deposits. Note subepithelial deposits in addition to mesangial deposits. Black arrows point to mesangial deposits, and white arrow points to subepithelial deposit. Original magnification, 340 in A, B, D, and E; 317,900 in C; 378,300 in F. with a monoclonal gammopathy. 7 In keeping with these findings, we found that rc3gn was associated with a monoclonal gammopathy in 3 of 14 (21%) patients. The median time to recurrence was shorter in patients with rc3gn associated with monoclonal gammopathy (3.6 months) compared with rc3gn not associated with monoclonal gammopathy (43.3 months) (Figure 4). Also, the course was more progressive in these patients. The P value was not significant at 0.13, which is most likely because of the small numbers. In this subset of rc3gn associated with monoclonal gammopathy, targeting the monoclonal Ig may serve as a novel therapeutic approach in such patients. DDD is similar to C3GN in that it results from dysfunction of the alternative pathway of complement. 9,10 It is distinguished from C3GN on electron microscopy. Electron microscopy shows sausage-shaped large osmiophilic dense deposits within the glomerular basement membranes, whereas in C3GN, the deposits tend to be mesangial and subendothelial. However, deposits of C3GN and DDD are both composed of complement factors of the alternative and terminal pathway of complement. 11 DDD has a high rate of recurrence after kidney transplantation and results in graft failure in up to 50% of patients, with graft loss occurring within 2.5 years of transplantation. 10,12,13 With regards to C3GN, we noted a recurrence rate of 66.7%, with graft failure in 50% of rc3gn. Thus, the rates of recurrence and graft failure are comparable with Figure 4. Monoclonal gammopathy and rc3gn. Relationship between serum monoclonal proteins and graft survival in patients with rc3gn (P=0.13, log rank). Median time to failure without (solid line) monoclonal Ig was 149.7 months, and median time to failure with (dashed line) monoclonal Ig was 69.9 months. DDD. The median time to recurrence was 28 months, and the median time to graft failure of 77 months after transplantation seems to be longer than DDD. 12 Regarding pathology, native kidney biopsy of C3GN more often shows an MPGN (nine cases) and less commonly shows a diffuse proliferative/postinfectious GN (three cases) pattern of J Am Soc Nephrol 25: 1110 1117, 2014 Recurrent C3GN 1115

CLINICAL RESEARCH www.jasn.org injury. 14 However, rc3gn often shows a mesangial proliferative (eight cases) or membranoproliferative pattern (six cases) of injury. These results are in keeping with our previous findings on recurrent MPGN. 15 Mesangial proliferative GN is more common in rc3gn, which is likely because of early detection of the recurrent disease. In the native biopsy, the membranoproliferative pattern is more common, which is likely because of late detection and the chronic nature of the disease. Finally, there were more men in the nonrecurrent group compared with women (86% versus 42%, P=0.06). Whether this result suggests a higher rate of recurrence in women is difficult to determine, because the numbers are too small to draw definite conclusions. Hematuria and proteinuria were absent in patients with functioning graft and no recurrence of C3GN. The treatment for rc3gn in this series included primarily use of rituximab or plasmapheresis in addition to the standard immunosuppressive therapy. Given the small number of patients who received additional therapy, it is difficult to draw any conclusions with respect to the efficacy of rituximab or plasmapheresis. The anticomplement drug eculizumab was recently reported to have partial benefit in a case report of a patient with DDD that recurred as C3GN. 16 None of the patients received eculizumab in our series. It is of interest that the stem cell transplant in the patient with monoclonal gammopathy did result in significant improvement in the renal function after significant reduction in the light chain and Ig. The association between C3GN and monoclonal and activation of alternative pathway of complement by l-light chain has been previously reported. 6 8 In our patient, it is, therefore, likely that the improvement in kidney function was the result of reduction in the l-light chains and decreased activation of alternative pathway of complement. Finally, 6 of 14 (43%) rc3gn and only 1 of 7 (14%) nonrc3gn typed as HLA-DR17 (3)-DQ2 haplotype, which has been seen in other autoimmune disorders like type 1 diabetes mellitus. Approximately 25% of the Caucasian population shows this haplotype. 17 It is difficult to draw conclusions because of small numbers, and larger studies are needed to determine if there is correlation between rc3gn and HLA haplotypes. This study is a retrospective analysis of rc3gn patients from a single institution. The advantage of this study is the detailed availability of all records, including biopsy material. An obvious limitation of this study is the lack of evaluation of the alternative pathway of complement. Because this study is retrospective, it is difficult to obtain samples for such an analysis. In summary, C3GN recurs at a high rate (66.7%). The majority of rc3gn is detected on clinical kidney allograft biopsies performed for hematuria, proteinuria, or elevated serum creatinine. Graft loss occurs in 50% of the patients with rc3gn. The recurrence often occurs within 2 3 years, but time to graft failure from time of transplant takes longer (6 7 years). Furthermore, rc3gn associated with monoclonal gammopathy is associated with earlier and more aggressive recurrences. CONCISE METHODS Study Design Native kidney biopsies of patients who received a kidney transplant between January of 1996 and December of 2010 at Mayo Clinic (Rochester, MN) were evaluated to confirm the diagnosis of C3GN. We did not include cases of rc3gn that were available through the renal pathology biopsy consultative practice. The criteria for diagnosis of C3GN were (1) proliferative GN on light microscopy, (2) dominant C3 staining (2 3+ of 3+) and absent or minimal staining for Ig (0 1+ of 3+) on immunofluorescence microscopy, and (3) mesangial and/ or capillary wall electron dense deposits on electron microscopy. Patients with DDD were excluded. In total, 21 patients who fulfilled the criteria of C3GN were identified. Protocol renal allograft biopsies were performed at time of transplant, 4 months, and 1, 2, 5, and 10 years after transplantation. Indication renal allograft biopsies were performed at the discretion of the nephrologist. rc3gn was diagnosed according to the criteria above. Antigen-level HLA typing was performed by serological or molecular methods. Clinical information was obtained from the review of the medical records. Graft loss was defined as return to dialysis or retransplantation. The study was approved by the Institutional Review Board at Mayo Clinic. Statistical Analyses Data are expressed as medians and ranges (minimum to maximum) and compared with nonparametric tests (Kruskall Wallis). The chi-squared test is used to compare proportions. Kaplan Meier curves and log-rank tests are used to evaluate the cumulative incidence of rc3gn and graft failure over time. P,0.05 is considered significant. ACKNOWLEDGMENTS This research was supported, in part, by the Fulk Family Foundation Award (Mayo Clinic; to S.S.). DISCLOSURES None. REFERENCES 1. Servais A, Frémeaux-Bacchi V, Lequintrec M, Salomon R, Blouin J, Knebelmann B, Grünfeld JP, Lesavre P, Noël LH, Fakhouri F: Primary glomerulonephritis with isolated C3 deposits: A new entity which shares common genetic risk factors with haemolytic uraemic syndrome. JMedGenet44: 193 199, 2007 2. Sethi S, Fervenza FC, Zhang Y, Zand L, Vrana JA, Nasr SH, Theis JD, Dogan A, Smith RJ: C3 glomerulonephritis: Clinicopathological findings, complement abnormalities, glomerular proteomic profile, treatment, and follow-up. Kidney Int 82: 465 473, 2012 3. Sethi S, Nester CM, Smith RJH: Membranoproliferative glomerulonephritis and C3 glomerulopathy: Resolving the confusion. Kidney Int 81: 434 441, 2012 1116 Journal of the American Society of Nephrology J Am Soc Nephrol 25: 1110 1117, 2014

www.jasn.org CLINICAL RESEARCH 4. Sethi S, Fervenza FC: Membranoproliferative glomerulonephritis a new look at an old entity. NEnglJMed366: 1119 1131, 2012 5. Servais A, Noël LH, Frémeaux-Bacchi V, Lesavre P: C3 glomerulopathy. Contrib Nephrol 181: 185 193, 2013 6. Bridoux F, Desport E, Frémeaux-Bacchi V, Chong CF, Gombert JM, Lacombe C, Quellard N, Touchard G: Glomerulonephritis with isolated C3 deposits and monoclonal gammopathy: A fortuitous association? Clin J Am Soc Nephrol 6: 2165 2174, 2011 7. Zand L, Kattah A, Fervenza FC, Smith RJ, Nasr SH, Zhang Y, Vrana JA, Leung N, Cornell LD, Sethi S: C3 glomerulonephritis associated with monoclonal gammopathy: A case series. Am J Kidney Dis 62:506 514, 2013 8. Meri S, Koistinen V, Miettinen A, Törnroth T, Seppälä IJ: Activation of the alternative pathway of complement by monoclonal lambda light chains in membranoproliferative glomerulonephritis. JExpMed175: 939 950, 1992 9. Appel GB, Cook HT, Hageman G, Jennette JC, Kashgarian M, Kirschfink M, Lambris JD, Lanning L, Lutz HU, Meri S, Rose NR, Salant DJ, Sethi S, Smith RJ, Smoyer W, Tully HF, Tully SP, Walker P, Welsh M, Würzner R, Zipfel PF: Membranoproliferative glomerulonephritis type II (dense deposit disease): An update. J Am Soc Nephrol 16: 1392 1403, 2005 10. Smith RJH, Alexander J, Barlow PN, Botto M, Cassavant TL, Cook HT, de Córdoba SR, Hageman GS, Jokiranta TS, Kimberling WJ, Lambris JD, Lanning LD, Levidiotis V, Licht C, Lutz HU, Meri S, Pickering MC, Quigg RJ, Rops AL, Salant DJ, Sethi S, Thurman JM, Tully HF, Tully SP, van der Vlag J, Walker PD, Würzner R, Zipfel PF; Dense Deposit Disease Focus Group: New approaches to the treatment of dense deposit disease. JAmSocNephrol18: 2447 2456, 2007 11. Sethi S, Gamez JD, Vrana JA, Theis JD, Bergen HR 3rd, Zipfel PF, Dogan A, Smith RJ: Glomeruli of Dense Deposit Disease contain components of the alternative and terminal complement pathway. Kidney Int 75: 952 960, 2009 12. Braun MC, Stablein DM, Hamiwka LA, Bell L, Bartosh SM, Strife CF: Recurrence of membranoproliferative glomerulonephritis type II in renal allografts: The North American Pediatric Renal Cooperative Study experience. J Am Soc Nephrol 16: 2225 2233, 2005 13. Andresdottir MB, Assmann KJ, Hoitsma AJ, Koene RA, Wetzels JF: Renal transplantation in patients with dense deposit disease: Morphological characteristics of recurrent disease and clinical outcome. Nephrol Dial 14: 1723 1731, 1999 14. Sethi S, Fervenza FC, Zhang Y, Zand L, Meyer NC, Borsa N, Nasr SH, Smith RJ: Atypical postinfectious glomerulonephritis is associated with abnormalities in the alternative pathway of complement. Kidney Int 83: 293 299, 2013 15. Lorenz EC, Sethi S, Leung N, Dispenzieri A, Fervenza FC, Cosio FG: Recurrent membranoproliferative glomerulonephritis after kidney transplantation. Kidney Int 77: 721 728, 2010 16. Gurkan S, Fyfe B, Weiss L, Xiao X, Zhang Y, Smith RJ: Eculizumab and recurrent C3 glomerulonephritis. Pediatr Nephrol 28: 1975 1981, 2013 17. Burdett L, Smith K, Tu B, Guiterrez M, Buck K, Maiers M, Ng J, Hartzman R, Fernandez-Vina M: DRB-DQB1 diversity in the analysis of 4727 donors typed by SBT. Hum Immunol 64[10 Suppl]: S6, 2003 J Am Soc Nephrol 25: 1110 1117, 2014 Recurrent C3GN 1117