A clone-directed approach may improve diagnosis and treatment of proliferative glomerulonephritis with monoclonal immunoglobulin deposits

The optimal treatment for the monoclonal gammopathies of renal significance is not known, but there is consensus among experts that treatment should be specific for the underlying clone. The majority of patients with proliferative glomerulonephritis with monoclonal immunoglobulin deposits (PGNMID) do not have an identifiable clone, and prior studies have found poor renal outcomes for patients with PGNMID treated with a variety of regimens. Here we present a retrospective case series of 19 patients with PGNMID with a more uniform, clone-directed approach. A circulating paraprotein was detected in 37% of patients, and the overall clone detection rate was 32%. Treatment was directed at the underlying clone or, for patients without a detectable clone, empirically prescribed to target the hypothesized underlying clone. Of the 16 patients who underwent treatment, the overall renal response rate was 88%, and 38% of patients experienced complete renal response (proteinuria reduction to under 0.5 gm/24 hours) with initial treatment. All patients were End Stage Renal Disease-free at last follow-up (median 693 days after diagnosis), and treatment was well tolerated. Thus, a clone-directed approach may lead to novel, targeted treatment strategies that could significantly improve outcomes for patients with PGNMID. The optimal treatment for the monoclonal gammopathies of renal significance is not known, but there is consensus among experts that treatment should be specific for the underlying clone. The majority of patients with proliferative glomerulonephritis with monoclonal immunoglobulin deposits (PGNMID) do not have an identifiable clone, and prior studies have found poor renal outcomes for patients with PGNMID treated with a variety of regimens. Here we present a retrospective case series of 19 patients with PGNMID with a more uniform, clone-directed approach. A circulating paraprotein was detected in 37% of patients, and the overall clone detection rate was 32%. Treatment was directed at the underlying clone or, for patients without a detectable clone, empirically prescribed to target the hypothesized underlying clone. Of the 16 patients who underwent treatment, the overall renal response rate was 88%, and 38% of patients experienced complete renal response (proteinuria reduction to under 0.5 gm/24 hours) with initial treatment. All patients were End Stage Renal Disease-free at last follow-up (median 693 days after diagnosis), and treatment was well tolerated. Thus, a clone-directed approach may lead to novel, targeted treatment strategies that could significantly improve outcomes for patients with PGNMID. Proliferative glomerulonephritis with monoclonal Ig deposits (PGNMID) is a renal-limited glomerular disease diagnosed by a kidney biopsy showing membranoproliferative or endocapillary proliferative glomerulonephritis on light microscopy, monoclonal Ig and complement (commonly C3) deposition on immunofluorescence microscopy, and nonorganized electron-dense deposits on electron microscopy.1Nasr S.H. Satoskar A. Markowitz G.S. et al.Proliferative glomerulonephritis with monoclonal IgG deposits.J Am Soc Nephrol. 2009; 20: 2055-2064Google Scholar PGNMID is caused by monoclonal gammopathy of renal significance (MGRS): a hematologic disorder associated with a paraprotein causing kidney injury that does not meet the criteria for malignancy (systemic multiple myeloma [plasma cell]) or lymphoma [B cell]).2Bridoux F. Leung N. Hutchison C.A. et al.Diagnosis of monoclonal gammopathy of renal significance.Kidney Int. 2015; 87: 698-711Google Scholar Patients typically present with renal insufficiency, proteinuria, and microscopic hematuria.1Nasr S.H. Satoskar A. Markowitz G.S. et al.Proliferative glomerulonephritis with monoclonal IgG deposits.J Am Soc Nephrol. 2009; 20: 2055-2064Google Scholar The optimal treatment for most subtypes of MGRS is not known, but there is consensus among experts that treatment should be specific for the underlying clone.3Fermand J.P. Bridoux F. Kyle R.A. et al.How I treat monoclonal gammopathy of renal significance (MGRS).Blood. 2013; 122: 3583-3590Google Scholar However, the majority of patients with PGNMID do not have an identifiable clone.4Bhutani G. Nasr S.H. Said S.M. et al.Hematologic characteristics of proliferative glomerulonephritides with nonorganized monoclonal immunoglobulin deposits.Mayo Clin Proc. 2015; 90: 587-596Google Scholar Previous studies found poor renal outcomes for patients with PGNMID treated with a variety of regimens.1Nasr S.H. Satoskar A. Markowitz G.S. et al.Proliferative glomerulonephritis with monoclonal IgG deposits.J Am Soc Nephrol. 2009; 20: 2055-2064Google Scholar Here we report a retrospective case series of 19 patients with PGNMID who were managed at the University of Pennsylvania with a more uniform, clone-directed approach. Nineteen patients with PGNMID were identified, all of whom were comanaged by nephrologists and hematologists at the University of Pennsylvania. The clinical and histologic characteristics at the time of diagnosis are presented in Table 1. The mean age at diagnosis was 58 years (range, 25–83 years); 63% (12/19) of patients were male and 32% (6/19) were black. The median estimated glomerular filtration rate (eGFR) at diagnosis was 38 (interquartile range [IQR], 23–58 ml/min per 1.73 m2, and the median proteinuria was 3.6 (IQR, 2.3–8.0 g/g or g/24 h). The median follow-up time after diagnosis was 693 days (IQR, 354–1355 days). Two patients (patients 9 and 10) had PGNMID diagnosed in the allograft after kidney transplantation. The presumed cause of end-stage renal disease (ESRD) in both patients had been hypertensive nephrosclerosis, and neither patient had undergone a native kidney biopsy. No patient was on dialysis at the time of diagnosis.Table 1Baseline characteristics of patients with PGNMID the time of diagnosisPatientAgeSexRaceeGFRSCr (mg/dl)ProteinuriaaProteinuria expressed as g/g of creatinine or g/24 h.% GSLight microscopy patternIFTAKidney Bx IFCirculating paraprotein (method of detection)Clone (% bone marrow involvement)Group 1: clone detected, clone-directed therapy126MW302.86.7042MPGNMildIgGλIgGλ (sIFE, sFLC)Lympho-plasmacytic (20% B-cell aggregates, 5% plasma cells)250MW581.43.442MPGNMildIgGλNoneB cellbBone marrow biopsy results for patient 2 not available (performed before electronic medical record).351FB581.22.200Mesangioproliferative and ECPGNMildIgGκIgGκ (sIFE)B cell (>80%)453MW451.73.000MPGNModerateIgG3κIgGκ (SPEP, sFLC)Plasma cell (5%–10%)Group 2: Clone-detected, nondirected therapy572MW391.75.9050MGPNSevereIgG3κNonePlasma cell (<5%)657MW183.64.3083MPGNSevereIgGκIgGκ (SPEP, sFLC, UPEP)Plasma cell (5%)Group 3: No clone detected, empirical therapy767FW232.13.5633MGPN and ECPGNMildIgG1λNoneNone865MW342.09.5213MPGNModerateIgMκIgGκ, IgG λ (SPEP)None9cPatients 9 and 10 were diagnosed with PGNMID in the allograft after kidney transplant.57FB551.31.940MGPN and ECPGNMildIgG1κNoneNone10cPatients 9 and 10 were diagnosed with PGNMID in the allograft after kidney transplant.38FW351.80.560MPGNMildIgG3κNoneNone1134MH511.715.0039MPGNModerateIgGλNoneNone1243MB661.53.9014MPGNSevereIgGλNoneNone1325FW352.02.7020MPGN, MNModerateIgG1κNoneNone1436MB681.524.000MPGN, MNMildIgGκNoneNone1569MW381.83.4386MGPNMildIgGλIgGλ (sIFE)None1675FW143.11.4756MGPN and ECPGNSevereIgGκNoneNoneGroup 4: not treated1778MB154.08.5043MPGNSevereIgMκNoneNone1859MB581.58.0077MGPN and ECPGNModerateIgG1λNoneNone1983FW142.92.280MPGNSevereIgG3λIgGλ (sIFE)NoneB, black; Bx, biopsy; eGFR, estimated glomerular filtration rate (expressed as ml/min per 1.73 m2 calculated by the Chronic Kidney Disease Epidemiology Collaboration equation); ECPGN, endocapillary proliferative glomerulonephritis; F, female; GS, glomerulosclerosis; H, Hispanic; IFTA, interstitial fibrosis/tubular atrophy; M, male; MN, membranous; MPGN, membranoproliferative glomerulonephritis; SCr, serum creatinine; sFLC, serum κ/λ free light chain assay; sIFE, serum immunofixation; SPEP, serum protein electrophoresis; UPEP, urine protein electrophoresis; W, white.a Proteinuria expressed as g/g of creatinine or g/24 h.b Bone marrow biopsy results for patient 2 not available (performed before electronic medical record).c Patients 9 and 10 were diagnosed with PGNMID in the allograft after kidney transplant. Open table in a new tab B, black; Bx, biopsy; eGFR, estimated glomerular filtration rate (expressed as ml/min per 1.73 m2 calculated by the Chronic Kidney Disease Epidemiology Collaboration equation); ECPGN, endocapillary proliferative glomerulonephritis; F, female; GS, glomerulosclerosis; H, Hispanic; IFTA, interstitial fibrosis/tubular atrophy; M, male; MN, membranous; MPGN, membranoproliferative glomerulonephritis; SCr, serum creatinine; sFLC, serum κ/λ free light chain assay; sIFE, serum immunofixation; SPEP, serum protein electrophoresis; UPEP, urine protein electrophoresis; W, white. Kidney biopsy characteristics are listed in Table 1. Light microscopy showed endocapillary, mesangioproliferative, and/or membranoproliferative glomerulonephritis in all cases. Immunofluorescence microscopy showed dominance of IgGκ staining in 9 cases, IgGλ in 8 cases, and IgMκ in 2 cases. γ Heavy chain subclass staining was performed in 8 of 17 cases and revealed IgG1 in 4 cases and IgG3 in 4 cases. The median percentage of globally sclerotic glomeruli was 20% (IQR, 0–50%). Interstitial fibrosis was mild in 8 cases, moderate in 5 cases, and severe in 6 cases. Details of the hematologic evaluation are shown in Table 1. A circulating paraprotein was detected in the serum or urine of 7 patients (37%), and of these patients, there was concordance between the circulating and kidney biopsy monoclonal Ig in 6 of 7 cases. Seventeen of 19 patients underwent bone marrow biopsy. A detectable clone was found in 6 patients (32% overall, 35% of patients who underwent clone work up). The underlying clone was a plasma cell in 3 cases, a B cell in 2 cases, and a lymphoplasmacytic clone in 1 case (Table 1). Four of the 6 patients with a detectable clone also had a detectable circulating paraprotein. Sixteen patients underwent treatment. Three patients were not treated at their physician’s discretion due to severe renal insufficiency, moderate-to-severe scarring observed on kidney biopsy, and/or additional comorbid conditions (Group 4). Details regarding therapy and response to initial treatment regimens are presented in Table 2 and Supplementary Table S1. Thirteen of 17 (76%) treated patients had a response to their initial therapy, 6 (35%) of whom experienced a complete response (CR). All 4 patients in Group 1 experienced a renal response, with 3 patients achieving a CR. One patient in Group 2 experienced a PR, whereas the second patient died 2.5 months after starting treatment of complications of recurrent hemothorax that were not attributed to his kidney disease or its treatment. Eight of 10 patients (80%) in Group 3 experienced renal response with initial therapy, 3 (30%) of whom had a CR. In responders, the median time to a PR (N = 13) was 5.2 months (IQR, 2.5–10.1 months), and the median time to a CR (N = 6) was 12.5 months (IQR, 6.5–21.2 months). Complete renal response was not contingent on resolution of paraproteinemia, as was noted in patients 1 and 4, both of whom experienced a CR but still had a small but detectable monoclonal spike on serum protein electrophoresis. No patient who underwent treatment developed ESRD during follow-up. All patients who did not undergo treatment (Group 4) progressed to ESRD during the follow-up period.Table 2PGNMID outcomes with initial treatmentCloneTherapyTreatment durationResponseTime to response (mo)PRCRGroup 1: clone-detected, clone-directed therapy1Lympho-plasmacyticRTX/CY/BOR/D3 moCR14.316.32B cellChlorambucilNAPRNA–3B cellRTX/PRED6 moCR1.23.34Plasma cellCY/BOR/D6 moCR5.133.3Group 2: clone-detected, nondirected therapy5Plasma cellMMF/PRED2.5 moNone––6Plasma cellPRED22 moPR15.2–Group 3: no clone-detected, empirical therapy7NoneRTX/CY/PRED9 moCR28.78NoneRTX/CY/PRED3.5 moPR7.3–9NoneRTX1 cycle of RTX 1000 mg i.v. × 2CR3.321.110NoneCY/PRED2 moPR5.2–11NoneRTX/PRED6 moPR9.2–12NoneRTX/CY/PRED6 moNone––13NoneRTX6 moPR11–14NoneRTX/PRED6 moNone––15NoneRTX/CY/BOR/D6 moCR1.16.516NoneBOR/D6 moPR3–BOR, bortezomib; CR, complete response; CY, cyclophosphamide; D, dexamethasone; MMF, mycophenolate mofetil; MPGN, membranoproliferative glomerulonephritis; NA, data not available; PGNMID, proliferative glomerulonephritis with monoclonal Ig deposits; PR, partial response; PRED, prednisone; RTX, rituximab; (-), data not applicable. Open table in a new tab BOR, bortezomib; CR, complete response; CY, cyclophosphamide; D, dexamethasone; MMF, mycophenolate mofetil; MPGN, membranoproliferative glomerulonephritis; NA, data not available; PGNMID, proliferative glomerulonephritis with monoclonal Ig deposits; PR, partial response; PRED, prednisone; RTX, rituximab; (-), data not applicable. Response to treatment was stratified by baseline eGFR, and it was found that patients with a baseline eGFR ≥45 and 20 to 44 ml/min per 1.73 m2 had similar rates of CR and PR (Figure 1). The 2 patients treated with a baseline eGFR <20 ml/min per 1.73 m2 also achieved a PR. Response to treatment was also stratified by the degree of interstitial fibrosis on kidney biopsy (Figure 2). Seven of 8 patients (87.5%) with mild interstitial fibrosis experienced a response to initial therapy, 5 of whom (62.5%) had a CR. All 4 patients (100%) with moderate interstitial fibrosis who were treated experienced a response to initial therapy, 1 of whom (25%) had a CR. Two of the 4 patients (50%) with severe interstitial fibrosis who were treated experienced a PR. Patient 12 did not respond to initial treatment with rituximab-cyclophosphamide-prednisone but then had a PR with bortezomib-dexamethasone.Figure 2Overall response to therapy stratified by degree of interstitial fibrosis on kidney biopsy. IFTA, interstitial fibrosis/tubular atrophy.View Large Image Figure ViewerDownload Hi-res image Download (PPT) Six patients relapsed during the follow-up period, with a median time-to-relapse of 15 months (range, 3–60 months) after initial treatment (Table 3, Supplementary Table S1). Relapses were more common in patients who achieved renal response and had a detectable clone (4 of 5 patients, 80%) versus those without a detectable clone (2 of 8 patients, 25%), but this difference did not reach statistical significance (P = 0.053). However, the overall median follow-up for responders with detectable clones was greater than that of those without detectable clones (1566 days [IQR, 1383–2996 days) versus 475 [IQR 344–729] days, P = 0.006). A repeat kidney biopsy was performed in 3 patients (patients 1, 4, and 10) after relapse and confirmed PGNMID as the diagnosis in each case. All patients who relapsed were treated, and 4 of 5 patients with available follow-up achieved a response (3 of 5, a CR; 1 of 5, a PR; and 1 had a stable eGFR and 40% reduction in proteinuria but did not meet the criteria for a PR). Patient 6 subsequently relapsed again, had no response to lenalidomide-dexamethasone, but had a CR to bortezomib-dexamethasone. Patient 10 was being treated for relapse at the time of publication.Table 3Patients treated with a second course of therapy for PGNMIDPatientResponse, initial treatmentTime to relapse (mo)Second treatmentResponseTreated for relapse1CR with RTX/CY/BOR/D20Ofatumumab/bendamustineCR2PR with chlorambucil60RituximabCR4CR with CY/BOR/D18BOR/DCR6aPatients 6 and 14 received additional treatment cycles (see Supplementary Table S1).PR with PRED9BOR/DPR10PR CY/PRED3Cy/BOR/DTBD11PR with RTX/PRED12RTXNoneTreated for no response12NR with RTX/CY/PREDNABOR/DPR14aPatients 6 and 14 received additional treatment cycles (see Supplementary Table S1).NR with RTX/PREDNARTX/PREDNoneBOR, bortezomib; CR, complete response; CY, cyclophosphamide; D, dexamethasone; NA, not applicable; PGNMID, proliferative glomerulonephritis with monoclonal Ig deposits; PR, partial response; PRED, prednisone; RTX, rituximab; TBD, to be determined (patient 10 undergoing treatment for relapse at the time of publication).a Patients 6 and 14 received additional treatment cycles (see Supplementary Table S1). Open table in a new tab BOR, bortezomib; CR, complete response; CY, cyclophosphamide; D, dexamethasone; NA, not applicable; PGNMID, proliferative glomerulonephritis with monoclonal Ig deposits; PR, partial response; PRED, prednisone; RTX, rituximab; TBD, to be determined (patient 10 undergoing treatment for relapse at the time of publication). Two patients who did not respond to initial therapy received a second course of treatment (Table 3). Patient 12 (Group 3) initially experienced no response to rituximab-cyclophosphamide-prednisone, and then experienced a PR with 6 months of bortezomib-dexamethasone (serum creatinine [SCr], 1.3–1.8 mg/dl; proteinuria, 2.4 g from 8.0 g) with a time to a PR of 5 months. This patient has subsequently maintained a PR on quarterly bortezomib treatment. Patient 14 (Group 3) did not respond to a second course of rituximab-prednisone and also had no response to rituximab-cyclophosphamide-bortezomib-dexamethasone. Patient 1 experienced severe volume overload requiring hospitalization during his third cycle of treatment that was attributed to a combination of medication nonadherence, congenital heart disease, nephrotic syndrome, bortezomib, and dexamethasone. He had a reduction in his serum λ free light chain level, so further treatment was withheld, and over time he experienced a renal CR. After relapsing, he experienced a severe infusion reaction when being re-treated with rituximab and so was treated with the human anti-CD20 monoclonal antibody ofatumumab without complications. Patient 4 experienced peripheral neuropathy and gastrointestinal side effects with bortezomib-dexamethasone. Given that he had achieved a CR, his treatment was held. No other patient stopped therapy due to a treatment-related adverse event. Patient 16 experienced peripheral neuropathy with bortezomib requiring dose reduction, but she completed her treatment course and achieved a PR. Here we report our experience managing 19 patients with PGNMID at the University of Pennsylvania. All treated patients but 2 underwent testing with bone marrow biopsy, serum, and urine paraprotein studies. Seven of 19 (37%) had a detectable circulating paraprotein, and 6 of 19 (32%) had a detectable underlying clone. Patients with detectable clones were treated with clone-directed therapy, when possible. Patients without detectable clones were treated empirically with regimens that are appropriate for B- and/or plasma cell clones. Using this approach, 13 of 16 (81%) treated patients responded to the first treatment, 6 of whom experienced a CR, and 1 nonresponder to the first therapy experienced a PR with a second treatment regimen, leading to an overall response rate of 88% (15/17). Six patients experienced a relapse during follow-up, and the majority (4/5) responded to a second course of therapy. One death occurred that was not attributed to PGNMID or its treatment, and all other treated patients were alive and free of ESRD at last follow-up. Three patients experienced treatment-related adverse events that required modification of therapy, but nonetheless achieved a CR or PR. PGNMID as a histopathologic entity was first described in 2004.5Nasr S.H. Markowitz G.S. Stokes M.B. et al.Proliferative glomerulonephritis with monoclonal IgG deposits: a distinct entity mimicking immune-complex glomerulonephritis.Kidney Int. 2004; 65: 85-96Google Scholar PGNMID is mediated by deposition of monoclonal Ig in the kidney and is thus considered an MGRS-mediated disorder. However, there is a much lower rate of detection of circulating paraprotein and underlying clones in PGNMID compared with other renal lesions caused by MGRS.4Bhutani G. Nasr S.H. Said S.M. et al.Hematologic characteristics of proliferative glomerulonephritides with nonorganized monoclonal immunoglobulin deposits.Mayo Clin Proc. 2015; 90: 587-596Google Scholar The largest case series of patients with PGNMID describes poor renal outcomes after diagnosis, with a 38% PR or complete renal recovery, 38% persistent renal dysfunction, and 22% ESRD rates (N = 37, mean follow-up 30 months).1Nasr S.H. Satoskar A. Markowitz G.S. et al.Proliferative glomerulonephritis with monoclonal IgG deposits.J Am Soc Nephrol. 2009; 20: 2055-2064Google Scholar Our experience demonstrates that a joint onconephrologic approach focused on detection and specific treatment of an underlying clone can lead to improved renal outcomes, with an 88% overall response rate and 100% ESRD-free rate. Our case series has several strengths in the context of the published PGNMID literature. Nineteen patients is a large experience of a single center for a rare disease. After kidney biopsy, each patient was offered a comprehensive and standardized hematologic work up for both circulating paraprotein and underlying clonal cell disorder. Thus, our patients are well characterized from a histopathologic and hematologic perspective. The majority of patients received regimens used in clonal plasma (i.e., multiple myeloma) or B-cell (i.e., lymphoma) disorders. These include combination drug regimens and the use of agents such as bortezomib and ofatumumab, which are not commonly prescribed by nephrologists but which nonetheless may be better tolerated than prolonged treatment with high-dose prednisone. Our findings of 37% paraprotein and 32% clone detection rates are similar to a recent series of monoclonal Ig disease with nonorganized deposits by Bhutani et al.,4Bhutani G. Nasr S.H. Said S.M. et al.Hematologic characteristics of proliferative glomerulonephritides with nonorganized monoclonal immunoglobulin deposits.Mayo Clin Proc. 2015; 90: 587-596Google Scholar who found 25% clone and 30% paraprotein detection rates. Our experience was also similar to that of Bhutani et al. in that patients who lacked detectable paraprotein often did not have an identifiable underlying clone. The reason for the low detection rate of circulating protein and underlying clone in PGNMID compared with other MGRS disorders is not known. It is possible that the conventional detection methods used in these studies are not sensitive enough. Using serum immunoblotting has been shown to increase the sensitivity in detecting the circulating truncated heavy chain in heavy chain deposition disease.6Bridoux F. Javaugue V. Bender S. et al.Unravelling the immunopathological mechanisms of heavy chain deposition disease with implications for clinical management.Kidney Int. 2017; 91: 423-434Google Scholar Flow cytometry using minimal residual disease methodology on bone marrow and/or peripheral blood could also increase the sensitivity in detecting a clone.7Flores-Montero J. Sanoja-Flores L. Paiva B. et al.Next Generation Flow for highly sensitive and standardized detection of minimal residual disease in multiple myeloma.Leukemia. 2017; 31: 2094-2103Google Scholar New mass spectrometry techniques that are more sensitive than immunofixation electrophoresis may be needed to identify a paraprotein.8Mills J.R. Barnidge D.R. Dispenzieri A. et al.High sensitivity blood-based M-protein detection in sCR patients with multiple myeloma.Blood Cancer J. 2017; 7: e590Google Scholar Our series contributes valuable observations for the treatment of these patients. Although it may be expected that patients with detectable clones will respond to clone-directed therapy (100% response rate in Group 1), a clone-directed approach also led to a 90% overall response rate for patients who did not have a detectable clone (Group 3). The median time to response (5.2 months for a PR, 12.5 months for a CR), the improvement in renal parameters even after completing therapy, and the durability of patients’ responses are all important observations for the management of patients with PGNMID. We also noted that resolution of paraproteinemia was not necessary to achieve a complete renal response, as was noted in patients 1 and 4. Patient 12 had a particularly interesting course. After presenting with moderate to severe interstitial scarring, he had no detectable clone and did not respond to 6 months of rituximab, cyclophosphamide, and prednisone. However, treatment with bortezomib and dexamethasone led to a PR, which he has maintained with quarterly doses of bortezomib and dexamethasone. Recent literature suggests that bortezomib may be an important drug in the treatment of other MGRS disorders.6Bridoux F. Javaugue V. Bender S. et al.Unravelling the immunopathological mechanisms of heavy chain deposition disease with implications for clinical management.Kidney Int. 2017; 91: 423-434Google Scholar, 9Bonaud A. Bender S. Touchard G. et al.A mouse model recapitulating human monoclonal heavy chain deposition disease evidences the relevance of proteasome inhibitor therapy.Blood. 2015; 126: 757-765Google Scholar This effect may be mediated through induction of the unfolded protein response and proapoptotic pathways in clonal cell populations9Bonaud A. Bender S. Touchard G. et al.A mouse model recapitulating human monoclonal heavy chain deposition disease evidences the relevance of proteasome inhibitor therapy.Blood. 2015; 126: 757-765Google Scholar, 10Kubiczkova L. Pour L. Sedlarikova L. et al.Proteasome inhibitors - molecular basis and current perspectives in multiple myeloma.J Cell Mol Med. 2014; 18: 947-961Google Scholar as well as antifibrotic effects in the kidney.11Zeniya M. Mori T. Yui N. et al.The proteasome inhibitor bortezomib attenuates renal fibrosis in mice via the suppression of TGF-beta1.Sci Rep. 2017; 7: 13086Google Scholar Given that bortezomib-based therapy is currently the first line of the treatment for plasma cell disorders, one may hypothesize that bortezomib responsiveness in patients without a detectable clone reflects the presence of a small plasma cell clone. However, it is important to recognize that bortezomib also has anti–B cell activity and is approved for the treatment of mantle cell lymphoma.12Tanday S. Bortezomib treatment for patients with mantle-cell lymphoma.Lancet Oncol. 2015; 16: e162Google Scholar We also found that patients with severe renal insufficiency and/or moderate to severe interstitial scarring and glomerulosclerosis may still benefit from treatment (Figures 1 and 2). This is a particularly important in PGNMID because many patients are diagnosed in their sixth and seventh decades of life and therefore may not be candidates for kidney transplantation if they progress to ESRD.13Menn-Josephy H. Lee C.S. Nolin A. et al.Renal interstitial fibrosis: an imperfect predictor of kidney disease progression in some patient cohorts.Am J Nephrol. 2016; 44: 289-299Google Scholar Therapy was well tolerated in general, with only 1 patient (who nonetheless experienced a CR) requiring treatment modification due to treatment-related adverse events. Further study is required to assess the efficacy and safety of treating patients with severe renal insufficiency and fibrosis on kidney biopsy, given the potential of treatment regimens to cause significant hematologic and infectious complications, particularly in frail patients. Notwithstanding these strengths, our series has significant limitations. These include small overall number of cases and heterogeneity in the treatment regimens prescribed. The definitions of renal response have not been validated as a surrogate outcome for long-term renal outcomes in PGNMID, and follow-up was limited in some patients. Adverse events were detected based on chart review rather than a prospective registry, which may lead to underdetection of important adverse events. IgG subclass staining on kidney biopsy tissue, which would provide further evidence that these were monotypic deposits, was not performed in all biopsies. This was due to heterogeneity in the practice of IgG subclass staining among the pathology labs where biopsy findings were read, lack of sufficient tissue for further analysis, and/or the biopsy having occurred before the routine use of IgG subclass staining. In this retrospective study, evaluation for a clone was not uniform across subjects, in part due to different technologies having been developed during our observation period (such as positron emission tomography scanning and plasma cell flow cytometry on bone marrow biopsies), which could not be uniformly applied to our study population. In conclusion, we report that our clone-directed approach to PGNMID led to renal response in 88% of treated patients, and an ESRD-free rate of 100% during follow-up. Based on these data, we recommend that all patients with PGNMID undergo joint renal and hematology evaluation, detectable clones should be treated with clone-directed therapy, and patients without detectable clones should receive empirical therapy targeting the hypothesized underlying clone. Larger, prospective, multicenter studies comparing the safety and tolerability of different treatment regimens would address an unmet need in PGNMID, particularly in patients without a detectable clone. Last, the use of more sensitive technologies for detecting paraproteins and clones needs further investigation. We identified all patients with biopsy-proven PGNMID diagnosed between 2000 and 2016 who were managed at the University of Pennsylvania and who had a minimum of 6 months of follow-up. The diagnosis of PGNMID was established based on renal biopsy findings of proliferative glomerulonephritis (membrano-, endocapillary, or mesangial proliferative), immunofluorescence showing monotypic Ig and complement deposition, and electron microscopy demonstrating immune complex-type deposits in subendothelial, mesangial, and/or subepithelial locations.1Nasr S.H. Satoskar A. Markowitz G.S. et al.Proliferative glomerulonephritis with monoclonal IgG deposits.J Am Soc Nephrol. 2009; 20: 2055-2064Google Scholar, 5Nasr S.H. Markowitz G.S. Stokes M.B. et al.Proliferative glomerulonephritis with monoclonal IgG deposits: a distinct entity mimicking immune-complex glomerulonephritis.Kidney Int. 2004; 65: 85-96Google Scholar Charts were retrospectively reviewed for clinical and histologic data. Renal parameters were serum creatinine (SCr), eGFR (calculated by the CKD-EPI equation), and proteinuria (by 24-hour urine collection or urine protein:creatinine ratio). Hematologic parameters included testing for an underlying clone (bone marrow biopsy and/or peripheral blood flow cytometry) and circulating paraprotein (serum protein electrophoresis, serum immunofixation, serum κ/λ free light chain assay, urine protein electrophoresis, and urine immunofixation). Kidney biopsy reports were reviewed by our renal pathologist (MBP) to confirm diagnosis, identification of paraprotein, degree of glomerulosclerosis and interstitial fibrosis, and that there was no evidence of organized deposits on electron microscopy. Our patients were evaluated individually based on history, physical and laboratory testing, and imaging, and our hematologists’ evaluations concluded that these patients did not meet criteria for malignancy. Patients were then divided into categories by clone-detection and treatment status as follows: Group 1 (N = 4): clone detected, treated with clone-directed therapy; Group 2 (N = 2): clone detected, treated with nonclone-directed therapy; Group 3 (N = 10): no clone detected, empirically treated; Group 4: not treated (N = 3). For patients with a detectable clone, clone-directed therapy was defined as a regimen that would be used for that clone’s malignant counterpart (i.e., antilymphoma regimen for B-cell clones, antimyeloma regimen for plasma cells); other regimens were designated as nonclone-directed therapy (patient 6: unable to obtain insurance approval for bortezomib; patient 7: treated with prednisone monotherapy before seeing a hematologist). Patients without a detectable clone were prescribed therapy at the physician’s discretion that would target an underlying clone that was not detected and for which insurance approval for the regimen could be obtained. CR was defined as stabilization or improvement in SCr and eGFR and urine proteinuria improvement to <0.5 g/g on urine protein-to-creatinine ratio or <0.5 g/24-h urine collection. A PR was defined as stabilization (±25%) or improvement of SCr, but not to normal, and >50% decrease in proteinuria. If there was nephrotic range proteinuria at baseline (>3 g/g or >3 g/24 h), response required >50% reduction in urine protein-to-creatinine ratio and a urine protein-to-creatinine ratio <3 g/g or <3 g/24 h urine collection. Patients were designated to have achieved a CR or PR if they had improvement in renal parameters at any point after initiation of treatment based on the observation that patients may continue to experience improvement in renal parameters even after therapy was completed. Time from treatment initiation to a PR or CR was also noted. Patients were deemed to have no response if they did not meet criteria for a CR or PR. Relapse was defined as worsening of proteinuria or SCr after the patient had achieved a PR or CR. Changes in paraprotein levels were described but not formally assessed by formal hematologic response criteria because there was a mixture of B- and plasma cell clones, and a majority of patients did not have measurable paraprotein levels. Description of treatment-associated adverse events was based on chart review. A description of response category stratified by baseline eGFR and the degree of interstitial fibrosis on kidney biopsy was also performed, as was comparison of relapse after response by clone detection status (χ2 test). Comparison of medians for nonparametric data was performed using Wilcoxon rank sum testing. No additional formal statistical analysis was performed due to small sample size. BMW has received research funding from Janssen and Prothena and consulting fees from Alnylam and Novartis. All the other authors declared no competing interests. Download .docx (.02 MB) Help with docx files Table S1Detailed treatment regimens and responses. In This issueKidney InternationalVol. 94Issue 1PreviewProliferative glomerulonephritis with monoclonal Ig deposits (PGNMIDs) is a monoclonal gammopathy of renal significance that is difficult to treat and has a poor prognosis. In a series of 19 patients with well-characterized PGNMID, Gumber et al., working with hematologists, found that a hematological approach to these patients was beneficial. An underlying paraprotein was found in 37% of cases and the (presumably) responsible B or plasma cell clone was found in one-third of patients. Clone-directed hematological protocols were used not only in these patients, but were also used empirically in patients in whom a clone or paraprotein could not be detected. Full-Text PDF Open ArchiveProliferative glomerulonephritis with monoclonal Ig deposits (PGNMID): diagnostic and treatment challenges for the nephrologist!Kidney InternationalVol. 95Issue 2PreviewIn their study of 19 cases of proliferative glomerulonephritis with monoclonal Ig deposits (PGNMID), Gumber et al. demonstrated a low detection rate of circulating paraprotein (37%) and pathogenic clones in bone marrow biopsies (32%).1 Additionally, the authors report a 100% response rate in 4 patients with identifiable clones treated with clone-directed regimens and a 90% overall response rate in the 10 patients with undetectable clones receiving empirical therapies. Of note, the 2 patients with detectable clones receiving nondirected therapy had severe baseline glomerulosclerosis and interstitial fibrosis. Full-Text PDF Open ArchiveThe authors replyKidney InternationalVol. 95Issue 2PreviewWe have read the letter by Kousios et al., which describes the single-center experience at the Imperial College in London of 14 patients with proliferative glomerulonephritis with monoclonal Ig deposits (PGNMID)1. A monoclonal gammopathy was detected in 36% of the patients, and 42% of those who underwent a bone marrow biopsy had a detectable plasma cell or B cell clone. Patients received a variety of therapies regardless of clone detection status, and renal outcomes were heterogeneous. Full-Text PDF Open Archive