see commentary on page 13

Membranous-like glomerulopathy with masked IgGkappa depositsChristopher P. Larsen1, Josephine M. Ambruzs1, Stephen M. Bonsib1, Christie L. Boils1, Larry N. Cossey1,Nidia C. Messias1, Fred G. Silva1, Yihan H. Wang1, Neriman Gokden2 and Patrick D. Walker1

1Nephropath, Little Rock, Arkansas, USA and 2University of Arkansas for Medical Sciences, Little Rock, Arkansas, USA

The diagnostic classification of glomerulonephritis is

determined by the interplay of changes seen using light,

immunofluorescence, and electron microscopy of the renal

biopsy. Routine direct immunofluorescence on fresh tissue is

currently considered the gold standard for the detection and

characterization of immune deposits. We recently found a

peculiar form of glomerular immune complex deposition

in which masked deposits required an antigen-retrieval

step to be visualized. Over a 2-year period, 14 cases were

characterized by numerous, large subepithelial deposits

visualized by electron microscopy and C3-predominant

staining by routine immunofluorescence on fresh tissue with

weak to negative immunoglobulin staining. Repeat

immunofluorescence after digestion of the formalin-fixed

paraffin-embedded tissue with pronase elicited strong IgG-jstaining restricted within the deposits. The patients were

often young with a mean age of 26 years and commonly had

clinical evidence of vague autoimmune phenomenon.

The clinicopathologic findings in this unusual form of

glomerulopathy do not fit neatly into any currently existing

diagnostic category. We have termed this unique form of

glomerulopathy membranous-like glomerulopathy with

masked IgG-j deposits.

Kidney International (2014) 86, 154–161; doi:10.1038/ki.2013.548;

published online 15 January 2014

KEYWORDS: glomerular disease; glomerulonephritis; glomerulopathy;

kidney biopsy; membranous nephropathy; proteinuria

Glomerular nephropathies with immune-complex depositionare placed in diagnostic categories on the basis of themorphologic findings on renal biopsy using light microscopy(LM), immunofluorescence (IF), and electron microscopy (EM).These disease categories are useful in that they are reproducibleand group patients according to unique pathogenic mechanisms,and they provide both treatment and prognostic information tothe treating clinician. Although many of these diseases such asmembranous glomerulopathy were first described 440 yearsago,1 others such as C3 glomerulopathy2 and proliferativeglomerulonephritis with monoclonal IgG deposits (PGNMIDs)3

have been recognized for o10 years.We have recently observed a peculiar form of glomerular

immune-complex deposition with unique clinical and patho-logic characteristics. The pattern of deposition is most akin tomembranous glomerulopathy by LM and EM in that most casesshow no evidence of endocapillary proliferation, and the depo-sits are largely subepithelial by EM. Routine IF studies showisolated C3 staining in these deposits. However, the depositsappear to be ‘masked’ in that the true nature of the deposits isnot evident until an antigen-retrieval step (pronase digestion) isadded to reveal the presence of IgG-k restriction. Most of thepatients are relatively young women who frequently have vagueautoimmune phenomenon. The clinical and demographicfindings in these patients do not correspond with those typicallyfound in patients with monoclonal membranous glomerulo-pathy or the more recently described diagnosis of PGNMID. Toour knowledge, this is the first description of this unique formof glomerulopathy, which we have termed membranous-likeglomerulopathy with masked IgG-k deposits (MGMIDs).

RESULTSClinical features

The first observation of MGMID came about in an effort tofurther evaluate for the presence of immunoglobulins in case2 after it was found to be C3-only by routine IF. Wesubsequently found that this phenomenon was not unique tothis patient, and we began collecting these cases with maskedsubepithelial deposits in an effort to determine whether therewere any commonalities between them. This series includesall cases of MGMID for which a biopsy was received in ourlaboratory since the initial recognition 2 years ago. Weidentified 14 cases of MGMID during this 2-year period of

c l i n i c a l i n v e s t i g a t i o n http://www.kidney-international.org

& 2014 International Society of Nephrology

Correspondence: Christopher P. Larsen, Nephropath, 10180 Executive

Center Drive, Suite 100, Little Rock, Arkansas 72211, USA.

E-mail: Chris.larsen@nephropath.com

Received 23 September 2013; revised 24 October 2013; accepted 14

November 2013; published online 15 January 2014

154 Kidney International (2014) 86, 154–161

study out of 10,956 native kidney biopsies (0.13%). Thedemographic and clinical details at the time of presentationare detailed in Table 1. The patients were predominantlyfemale (12/14), and multiple ethnicities were represented inthis cohort including 8 Caucasians, 2 African Americans,3 Hispanics, and 1 Pacific Islander. All patients were found tobe negative for hepatitis C, hepatitis B, and HIV. In addition,7/7 patients tested for rheumatoid factor were negative and11 patients had a serum and urine protein electrophoresisstudy with no evidence of a paraprotein. None of the patientsreported recent infections at the time of clinical presentation.Serum C3 and C4 were normal in all patients at the time ofpresentation and throughout follow-up. Anti-streptolysin Owas negative in all patients. The primary indication forbiopsy in 13/14 patients was proteinuria, with 5 of 14displaying the nephrotic syndrome. Patient 14 was theexception, in whom the primary indication for biopsy wasacute renal failure. Eleven of the patients were not on anymedication at the time of the biopsy, and there were no drugsin common among the remaining three patients.

There were nine patients (64%) with some evidence ofautoimmune disease either at the time of presentation or intheir past medical history. Although six of the patients had apositive anti-neutrophil antibody, only one of the six metARA criteria for a diagnosis of systemic lupus erythematosus.Patients 3 and 7 had a remote history of systemic lupuserythematosus but did not show systemic manifestations ofsystemic lupus erythematosus outside the kidney at the timeof renal biopsy. Patient 14 had a history of inflammatoryarthritis, but was not being treated at the time of the biopsy.Patient 2 had negative serologies but a past medical historysignificant for idiopathic thrombocytopenic purpura.

Light microscopy

The morphologic features of each biopsy are detailed inTable 2. The major finding was the presence of membranousglomerulopathy features. Silver positive glomerular basementmembrane (GBM) ‘spikes’ or silver negative GBM ‘pin holes’were present in a segmental distribution (o50% of loops) infive cases (Figure 1). The remaining nine cases showed global

Table 1 | Clinical characteristics at time of presentation for renal disease

Patient Age Gender Cr (mg/dl) Prot (g) ANA (titer) dsDNA ANCA Hematuria PMH

1 22 F 0.7 1.4 � � NP þ Proteinuria discovered during pregnancy2 17 F 0.6 3.2 � � NP þ ITP, hemolytic anemia3 27 F 0.8 1.3 þ (NT) þ � � SLE4 24 F 1.9 8 � � � þ None5 17 F 0.6 2.1 þ (NT) � NP þ Fever of unknown origin6 34 M 0.7 3.5 þ (1:80) � � þ Diabetes mellitus, type 2, and hypertension7 32 F 0.8 0.5 � � � þ Remote history of SLE8 49 F 0.9 2.9 þ (1:160) � NP þ Remote autoimmune hemolytic anemia9 21 F 0.5 5.4 � � � þ Proteinuria discovered during pregnancy10 26 M 1.5 5 � � NP þ None11 20 F 0.6 1.9 � � � þ None12 19 F 0.7 2 þ (1:80) BL þ NP þ None13 15 F 0.8 4.7 þ (1:80) � � þ None14 37 F 1.9 4.5 � � � þ Inflammatory arthritis

Abbreviations: ANA, anti-neutrophil antibody; ANCA, anti-neutrophil cytoplasmic antibody; BL, borderline; Cr, serum creatinine; dsDNA, double-stranded DNA; F, female;ITP, idiopathic thrombocytopenic purpura; M, male; NP, not performed; NT, no titer; PMH, past medical history; Prot, 24-h proteinuria; SLE, systemic lupus erythematosus.

Table 2 | Morphologic features on initial biopsy

Patient LM pattern Fresh IgG Fresh j Fresh k Pron IgG Pron j Pron k Fresh C3 Deposit locationHump-likedeposits

Hinge-regiondeposits

1 Normal 0 0 0 3 3 0 3 m, sp Yes Yes2 Normal 0 0 0 3 3 0 2 m, sp No Yes3 Normal 1 0 0 3 3 0 3 m, sp Yes Yes4 Focal crescentic 0 0 0 3 3 0 3 m, sp Yes Yes5 Normal 0 0 0 2 2 0 2 m, sp No Yes6 Normal 0 0 0 3 2 0 1 m, sp No No7 Normal 0 0 0 2 2 0 1 m, sp No Yes8 Normal 0 0 0 3 3 0 3 m, sp Yes No9 Normal 0 0 0 3 3 0 0 sp No No10 Normal 1 0 0 3 3 0 1 sp No No11 Normal 0 0 0 3 3 0 2 m, sp Yes No12 Normal 0 0 0 3 3 0 1 m, sp Yes Yes13 Focal crescentic Tr Tr 0 2 2 0 2 m, sp No No14 Focal crescentic Tr 0 0 3 3 0 2 m, sp Yes Yes

Abbreviations: LM, glomeruli on light microscopy; m, mesangial; Pron, pronase; sp, subepithelial; Tr, trace.

CP Larsen et al.: Masked glomerulopathy c l i n i c a l i n v e s t i g a t i o n

Kidney International (2014) 86, 154–161 155

involvement (X50% of loops), but in seven out of nine casesloops without spikes or pin holes were easily noted. GBMduplication was not present in any of the cases. Mesangialhypercellularity was present in 6/14 cases.

Eleven of the 14 biopsies showed no evidence of glomerularproliferation on initial biopsy. Three patients showed focalcrescent formation at presentation, with 6 of 39, 4 of 30, and19 of 45 crescentic glomeruli present in these 3 biopsies. Twoof these three with crescents had segmental necrotizing lesionswithin the crescents. No endocapillary proliferation waspresent in any of the biopsies. The trichrome stain revealedsubepithelial fuchsinophilic deposits in most glomeruli. Therewas little chronicity present in most of the biopsies, with 12/14patients showing no significant to mild interstitial fibrosis andtubular atrophy at initial biopsy. Patients 4 and 8 had severeinterstitial fibrosis and tubular atrophy, with 450% globalglomerulosclerosis on the initial biopsy. Patient 6 developedmoderate interstitial fibrosis and tubular atrophy on follow-upbiopsy (see Follow-up biopsies section below).

Immunofluorescence

Ten of the 14 patients showed completely negative staining byroutine IF for both commercial polyclonal IgG antibodiesused and k- and l-light-chain antibodies. There were fourcases with weak, trace to 1þ IgG staining that was signifi-cantly stronger after pronase digestion. All cases showedmoderate to strong staining for IgG and k after pronasedigestion (Figure 2). Twelve of the 14 cases were negative forIgG subtypes by routine IF. Cases 3 and 14 were positive forglomerular IgG 1 using the routine IF technique (i.e., frozentissue with no pronase digestion). Unfortunately, IgGsubtype staining can only be done on frozen tissue and doesnot work on pronase-digested tissue. Phospholipase A2receptor staining was negative in the glomeruli in all cases.Phospholipase A2 receptor is examined using sections thathave undergone pronase digestion, strongly supporting theconcept that these cases do not represent a primary form ofmembranous glomerulopathy.4,5 IgA and IgM were negativein all glomeruli by routine IF. C1q was negative in 13/14patients and showed 1þ staining in patient 12. IgA and IgM

were repeated after pronase digestion in the paraffin-embedded tissue and were negative in 12/14 patients. Therewas trace staining for IgA and IgM in the glomeruli of patient12 after pronase digestion, and patient 5 had trace stainingfor IgM. There was no staining of the tubular basementmembranes by immunoglobulins either before or afterpronase digestion.

The specificity of the antibodies on paraffin-embeddedtissue after pronase digestion was tested by staining caseswith other types of glomerulonephritis, including 5 cases ofIgA nephropathy, 10 cases of C3 glomerulonephritis, and5 cases of C3-only proliferative infection–associated glomer-ulonephritis. IgG staining was negative within these cases inroutine IF, as well as in the pronase-digested tissue.

Seven cases were examined by routine IF in a separatelaboratory (University of Arkansas for Medical Sciences) andall showed identical staining for IgG, k, and l, as was seen onroutine IF in our laboratory.

Electron microscopy

All patients showed evidence of subepithelial deposits, and12/14 had mesangial deposits (Figure 3). Hinge region (akamesangial waist region; 8/14) and large hump-like deposits(7/14) were present in some cases. The deposits were of theimmune complex type and none showed evidence ofsubstructure. There were no subendothelial deposits identi-fied in any of the samples. Tubuloreticular inclusions werealso not identified.

Clinical outcome

Treatment and follow-up data were available in nine patients(Table 3). Two of the patients were treated conservativelywith only an angiotensin-converting enzyme inhibitor, whereasseven of the patients received some form of immunosuppres-sion. None of the patients progressed to end-stage kidneydisease in the mean follow-up period of 14.9 months (range2–46 months). There was one patient with a completeremission (patient 3), two patients with partial remission(patients 2 and 6), and six patients with persistent kidneydisease during the time of follow-up.

Figure 1 | Glomerular deposits in membranous-like glomerulopathy with masked IgG-j deposits (MGMIDs). (a) Brightly fuchsinophilicsubepithelial deposits (arrows) are evident on this Masson trichrome stain from patient 11 (original magnification �600). (b) Segmental spikesare present along the glomerular basement membranes by Jones methenamine silver stain in patient 6 (original magnification �400).

156 Kidney International (2014) 86, 154–161

c l i n i c a l i n v e s t i g a t i o n CP Larsen et al.: Masked glomerulopathy

Follow-up biopsies

There were two cases with multiple renal biopsies. Patient8 had a previous biopsy 15 years prior at an outsideinstitution. This biopsy, which was not available for review,had a final diagnosis of membranoproliferative glomerulone-phritis, type 3 of Burkholder rendered. On the follow-upbiopsy in our laboratory, the patient had significant chronicinjury with severe interstitial fibrosis, and 34 of 47 glomeruliwith global glomerulosclerosis. Patient 5 had 4 biopsies overa 47-month period. The first biopsy showed changes typicalof MGMID without glomerular proliferation. The second

biopsy 12 months later showed similar changes withpersistence of IgG-k deposits by IF after pronase digestionof the paraffin-embedded tissue and no glomerular pro-liferation. The third biopsy at 45 months showed persistenceof masked IgG-k deposits and had evidence of diffusecrescent formation without endocapillary hypercellularity byLM. She was treated with rituximab at the time of this thirdbiopsy. The patient had a fourth biopsy 47 months after theinitial biopsy, which showed no evidence of proliferation byLM and persistence of masked IgG-k deposits by IF. Therewas increased development of chronic injury over these

Figure 3 | Electron-dense deposits of membranous-like glomerulopathy with masked IgG-j deposits (MGMIDs). (a) A large hump-likesubepithelial deposit is present along with subepithelial deposits more typical of membranous glomerulopathy in patient 4 (unstained, originalmagnification �12,000). (b) Numerous subepithelial deposits typical of membranous glomerulopathy are present in patient 9 (unstained,original magnification �12,000).

Figure 2 | Immunofluorescence findings from membranous-like glomerulopathy with masked IgG-j deposit (MGMID) patient 11.Glomeruli stain negative for IgG by routine immunofluorescence on fresh tissue (a) and positive on paraffin-embedded tissue afterpronase digestion (b) (direct immunofluorescence; original magnification �400). Glomeruli show staining for k (c) and not l (d) on thepronase-digested tissue (direct immunofluorescence; original magnification �400).

Kidney International (2014) 86, 154–161 157

CP Larsen et al.: Masked glomerulopathy c l i n i c a l i n v e s t i g a t i o n

47 months, with no interstitial fibrosis at the initial biopsyprogressing to moderate chronic injury with B30% inter-stitial fibrosis and tubular atrophy at the last biopsy.

Comparison with other glomerulonephritides showing lightchain–restricted IgG deposition

There were 19 cases with PGNMID during this same periodand 9 cases with a monoclonal membranous glomerulopathythat stained on routine IF in our files. The clinical andmorphologic features of these cases are compared withMGMID in Table 4. All of the cases classified as eitherPGNMID or monoclonal membranous glomerulopathystained by routine IF and did not require pronase digestion.The MGMID patients were significantly younger, with amean age of 26 years (range 15–49 years) and 92.8% beingdiagnosed at o40 years of age, while the mean age ofPGNMID was 67 (range 29–84) years and monoclonalmembranous was 69 (range 48–87) years. The MGMIDpatients were also significantly more likely to show someevidence of autoimmunity with autoimmune phenomenonin 64% of MGMID, 5% of PGNMID, and 0% of monoclonalmembranous glomerulopathy, Po0.001.

DISCUSSION

The classification of glomerulonephritis has evolved overthe past 70 years. The introduction of EM to augment LMfindings in renal biopsies beginning in the 1950s led to rapidand significant changes to our understanding and classifica-tion of glomerular diseases.6 The use of IF tags to identify

proteins in tissues was first described in 1950 and was soontried in diagnostic renal biopsies.7,8 The initial studies wereplagued by poor antibodies, but the reagents and the pro-cedures improved quickly such that IF became widespreadand integral to renal biopsy interpretation by the 1970s.These innovations led to a paradigm shift in the under-standing of glomerulonephritis. Not co-incidentally, many ofthe diseases commonly diagnosed on renal biopsy today werefirst recognized and named as the result of these earlyinnovations. A few renal biopsy laboratories added antibodiesto k- and l-light chains in the IF evaluation of renal biopsiesin the late 1970s and early 1980s. This led to the recognitionof new diagnostic categories of nephropathy such as light-chain deposition disease9 and PGNMIDs.3 Interestingly, theroutine evaluation of k- and l-light chains is still not carriedout in a significant number of renal pathology labs.10 Wehave identified a novel form of glomerular immune complexdeposition in which the protein epitopes are ‘masked’ andnot readily available for antibody recognition by routine IFon fresh tissue. The identification of this disease was madepossible by the advent of an IF procedure originally designedto be used as a salvage technique.11

Routine IF on fresh tissue is considered the gold-standardtechnique for the detection of immunoglobulins and otherproteins in kidney biopsies.12 Light-chain proximal tubulo-pathy is the one well-known exception in that themonoclonal crystalline inclusions in the proximal tubulecytoplasm sometimes do not react with antibodies on routineIF.In these cases, the addition of an antigen-retrieval step makes

Table 3 | Clinical follow-up

Patient Pres Cr (mg/dl) Pres Prot (g) Treatment F/U period (months) Recent Cr (mg/dl) Recent Prot (g)

1 0.7 1.4 Losartan 5 0.8 4.42 0.6 3.2 Lisinopril, tacrolimus 15 0.8 0.73 0.8 1.3 Hydroxychloroquine, prednisone 3.5 0.7 0.54 1.9 8 Mycophenolate, prednisone 3 2 65 0.6 2.1 Tacrolimus, rituximab 46 1.2 46 0.7 3.5 Lisinopril 20 0.7 1.28 0.8 0.5 Azathioprine, prednisone 24 1.2 1.210 1.5 5 Mycophenolate, prednisone 16 0.9 512 0.7 2 Prednisone 2 0.7 1.6

Abbreviations: F/U, follow-up; Pres Cr, presentation serum creatinine; Pres Prot, 24-h urine protein on presentation; Recent Cr, most recent serum creatinine; Recent Prot,most recent 24-h urine protein.

Table 4 | Comparison of MGMID with other glomerulonephritides that show IgG light-chain restriction

Clinical LM IF Deposit location

Dx (n)Mean age

(range)o40

(n; %)Autoim(n; %)

Endo(n; %)

Extra(n; %)

IgG-j(n; %)

IgG-k(n; %)

Pronaserequired (n; %)

Subendo(n; %)

Subepi(n; %)

PGNMID (19) 67.3 (29–84) 2 (11)* 1 (5)* 18 (95) 5 (26) 16 (84) 3 (16) 0 (0) 19 (100) 1 (5)MonoMG (9) 69.2 (48–87) 0 (0)* 0 (0)* 0 (0)* 0 (0) 9 (100) 0 (0) 0 (0) 0 (0) 9 (100)MGMID (14) 25.7 (15–49) 13 (93) 8 (62) 0 (0)* 4 (31) 14 (100) 0 (0) 14 (100) 0 (0) 14 (100)

Abbreviations: Autoim, autoimmune disease features; Dx, diagnosis; Endo, endocapillary proliferation; Extra, extracapillary proliferation; IF, immunofluorescence; LM, lightmicroscopy; MGMID, membranous-like glomerulopathy with masked IgG-k deposits; MonoMG, monoclonal membranous glomerulopathy with staining on routineimmunofluorescence; PGNMID, proliferative glomerulonephritis with monoclonal IgG deposits; Subendo, subendothelial; Subepi, subepithelial.*Po0.001.

158 Kidney International (2014) 86, 154–161

c l i n i c a l i n v e s t i g a t i o n CP Larsen et al.: Masked glomerulopathy

it possible to visualize the light-chain restriction within thesecrystalline inclusions.11,13 MGMID is the first report of aglomerular process, which demonstrates this same phenomenon,i.e., the use of antigen retrieval to unmask the deposits. Itseems that the masked nature of these deposits is integral tothis entity, as we detected no cases of monoclonal membra-nous glomerulopathy with similar clinical, demographic, andpathologic characteristics that required antigen retrieval inorder to demonstrate the immunoglobulin staining.

Despite the light-chain restriction within the deposits,there was no evidence of a lymphoproliferative disorder inthese patients. However, 64% of the patients did have somehistory of autoimmune disease or vaguely positive autoimmuneserologic studies. Although most cases showed no activeglomerular proliferation by biopsy, there were four patientswith evidence of crescent formation either at presentation orduring the follow-up period. All four of these patients withcrescents had negative anti-neutrophil cytoplasmic antibodyserologic studies and no systemic manifestations of lupusnephritis. Given this high proportion of patients with auto-immune phenomenon, we believe that this disease mostlikely represents a form of autoimmune glomerulonephritis.

The mechanism of the masking of these deposits isunclear. The antigenic epitopes targeted by the antibodies areclearly not available in the deposits before a retrieval step.However, we doubt that this simply represents steric hindrancerelated to a tertiary structure blocking the antibody binding,as this would require blocking multiple sites throughout theimmunoglobulin molecule. We used two different polyclonalIgG molecules. In addition, all 14 cases were negative for apolyclonal antibody directed against the k-light chain, and12/14 cases showed completely negative staining for poly-clonal antibodies directed against IgG subtypes. Given thelarge portion of the IgG molecule targeted by these polyclonalantibodies, we find it unlikely that the masking is a result ofabnormal protein folding. Other possibilities include hin-drance related to abnormal glycosylation of the IgG molecule,interference related to a charge–charge interaction betweenthe antibody and immunoglobulin, or perhaps a quaternaryarrangement of immunoglobulin molecules in a way thatimpedes antibody binding.

Lack of awareness of MGMID will almost certainly lead toplacement of these patients into uncertain or inappropriatediagnostic categories. C3 glomerulonephritis is a disease thatis defined by C3-restricted deposits that can show predomi-nance of subepithelial deposits by EM.2 This diagnosticcategory is now considered evidence for abnormalities in thealternative pathway of complement.14 MGMID could beeasily mistaken for C3 glomerulonephritis without awarenessof the potential for masked deposits. This would likely be aconfusing picture for the clinician, as none of the patients inour MGMID cohort had evidence of hypocomplementemiaat the time of clinical presentation or during follow-up.A second possible misdiagnosis given the presumed isolatedC3 staining and the presence of large hump-like depositsand/or hinge region deposits in many cases would be an

infection-associated glomerulonephritis. However, none ofour cases showed evidence of endocapillary proliferation atinitial biopsy, and none had experienced a recent previousinfection. In addition, the evidence that these deposits persistfor years in two of the patients is strongly against aninfectious etiology. The cases could also be misclassified as anatypical membranous glomerulopathy without IgG staining.We are aware of one case in the literature reported asimmunoglobulin-negative membranous glomerulopathy.15

Interestingly, the patient in this case report had a history ofan unusual form of lupus with a weakly positive anti-neutrophil antibody (1:80) at the time of presentation ofrenal disease similar to our cases of MGMID.

Recognition of the IgG-k deposition within glomeruli onpronase digestion raises the possibility of two newerdiagnostic possibilities that are somewhat similar. MGMIDcould be placed within the spectrum of PGNMID in an effortto avoid fragmenting this new diagnostic category, especiallyas the authors did include two cases without proliferation butwith a membranous pattern.3 However, PGNMID cases rarelyhave accompanying autoimmune phenomenon16 comparedwith 64% of cases with some evidence of autoimmunedisease in our present series. In addition, cases of PGNMIDfrom our case files over the same 2-year period as theMGMID cases showed a mean age of 67.3 years, with 11% ofpatients being o40 years of age. By contrast, the MGMIDpatients had a mean age of 25.7, with 93% being less than theage of 40 years. Finally, cases of PGNMID do not requirepronase digestion to reveal the presence of IgG and therestricted light-chain component. We also considered placingMGMID cases into the category of monoclonal membranousglomerulopathy, which stain on routine IF. We find thatroutine monoclonal membranous glomerulopathy appears tohave a distinct clinicopathologic picture both from theprevious 11 cases reported in the literature16–20 and from our9 cases of monoclonal membranous glomerulopathy obtainedin the same 2-year period. The patients had a mean age of69.2 years with none being o40 years of age, and none hadevidence of an autoimmune disease. Thus, MGMID shouldbe considered a new form of glomerular disease, as it is notonly unique morphologically but also has a distinct clinicalpicture.

The small samples size in this study precludes recom-mendation of a specific treatment regimen. Given thehypothesis that this likely represents an autoimmune disease,it would be reasonable to speculate that an immunosuppres-sive therapy might be beneficial. We have no evidence at thistime to suggest that this disease will spontaneously remit ifleft untreated. In fact, the one patient with follow-up biopsiesshowed persistent disease after 4 years and had significantchronic injury at the time of the last biopsy. Given thepossibility of progression and the fact that most of thesepatients are relatively young, we believe that immunosup-pressant therapy should be considered in an effort to preventdisease progression. However, the best type and duration ofthis treatment will need to be determined in future studies.

Kidney International (2014) 86, 154–161 159

CP Larsen et al.: Masked glomerulopathy c l i n i c a l i n v e s t i g a t i o n

We present what we believe to be the first description of anovel form of glomerulopathy that we have termedMGMIDs. The introduction of IF on paraffin-embeddedtissue after pronase digestion has made recognition of thisentity possible. Pronase digestion of the paraffin block for IFis most commonly implemented as a salvage technique whenno glomeruli are available in the tissue submitted for routineIF.11 However, recognition of MGMID broadens the applica-tion of this technique to include its use when deposits areclearly present by EM and no corresponding staining ispresent for immunoglobulins by routine IF. Previously,MGMID has likely been misdiagnosed as C3 glomerulo-pathy, or as an atypical form of infection-associated glomeru-lonephritis or membranous glomerulopathy. Future studiesare needed to better delineate the pathogenesis and treatmentof this entity.

MATERIALS AND METHODSAll cases were processed by light, IF, and EM using routinetechniques described below.21 Briefly, kidney biopsies were fixed inbuffered formalin, dehydrated in graded alcohols, and embedded inparaffin using standard techniques. Serial 3-mm-thick sections werecut and treated with hematoxylin and eosin, Jones methenaminesilver, Masson trichrome, and periodic acid–Schiff reagent. All datawere collected according to protocols approved by the SchulmanInstitutional Review Board.

ImmunofluorescenceSamples were transported in Michel’s medium, washed in buffer,and frozen in a cryostat. Sections, cut at 5 mm, were rinsed in bufferand reacted with fluorescein-tagged polyclonal rabbit anti-humanantibodies to IgG, IgA, IgM, C3, C4, C1q, fibrinogen, and k- andl-light chains (all from Dako, Carpenteria, CA) for 1 h, rinsed, and acoverslip applied using aqueous mounting medium. In all MGMIDcases, the IgG stain was repeated with a second fluorescein-taggedpolyclonal goat anti-human antibody to IgG (Kent Laboratories,Bellingham, WA). IgA, IgG, IgM, -k, and lambda were detected inparaffin-embedded sections using fluorescein-tagged polyclonalrabbit anti-human antibodies (all from Dako). The stains wereevaluated by standard IF microscopy using a Leica L5 filter cube.Phospholipase A2 receptor was detected in paraffin-embeddedsections following pronase digestion using rabbit polyclonal anti-phospholipase A2 receptor antibodies (Sigma-Aldrich, St Louis,MO) at a dilution of 1:50, followed by highly cross-adsorbed AlexaFluor 488 goat anti-rabbit IgG (Life Technologies, Carlsbad, CA) at adilution of 1:100, as previously described.4

Paraffin IF sections were cut at 3 mm. The sections weredeparaffinized in xylene followed by alcohol gradient, and pro-teinase K (Dako) was applied for 30 min at room temperature.The sections were then rinsed in buffer and reacted with fluo-rescein-tagged polyclonal rabbit anti-human antibodies to IgG, IgA,IgM, and k- and l-light chains (all from Dako) at room tempera-ture, rinsed, and a coverslip applied using aqueous mountingmedium.

Electron microscopyThe ends of the renal biopsy specimen were removed as 1-mmcubes, dehydrated using graded alcohols, and embedded in epon/araldite resin. Sections 1-mm-thick were cut using an ultramicrotome,

stained with toluidine blue, and examined with a light microscope.Thin sections were examined in a Jeol JEM-1011 electron micro-scope (Jeol, Tokyo, Japan). Hump-like deposits were defined as largedeposits with a rounded configuration, which set on the surface ofthe GBM with minimal-associated basement membrane response. Ingeneral, these deposits stand as high above the GBM as they are widealong its surface. Hinge deposits were defined as hump-like depositsoverlying mesangium between folds of GBMs.

Clinical definitionsHematuria was defined as greater than five red blood cells per high-power field on microscopic examination of urine sediment.Nephrotic syndrome was defined as 24-h urine protein 43.5 g perday, hypoalbuminemia (o3.5 g/dl), and peripheral edema. Thefollowing definitions were applied for the purposes of outcomeanalysis, and are similar to those used by Nasr et al.16 completeremission, remission of proteinuria to o500 mg/day with normalrenal function; partial remission, reduction in proteinuria by at least50% and to o2 g/day with stable renal function (no more than a20% increase in serum Cr); persistent kidney dysfunction, failure tomeet criteria for complete or partial remission but not reaching end-stage kidney disease (includes patients with unremitting proteinuriaor progressive chronic kidney disease); end-stage kidney disease,requiring renal replacement therapy.

DISCLOSUREAll the authors declared no competing interests.

ACKNOWLEDGMENTSWe thank Robert Yerton and Steve Strong for their valued technicalaid, and I-Shen Wen and Tina Priddy for their excellent administrativesupport.

REFERENCES1. Ehrenreich T, Churg J. Pathology of membranous nephropathy. Pathol

Ann 1968; 2: 145–186.2. Servais A, Fremeaux-Bacchi V, Lequintrec M et al. Primary

glomerulonephritis with isolated C3 deposits: a new entity which sharescommon genetic risk factors with hemolytic uremic syndrome. J MedGenet 2007; 44: 193–199.

3. Nasr SH, Markowitz GS, Stokes MB et al. Proliferative glomerulonephritiswith monoclonal IgG deposits: a distinct entity mimicking immune-complex glomerulonephritis. Kidney Int 2004; 65: 85–96.

4. Larsen CP, Messias NC, Silva FG et al. Determination of primary versussecondary membranous glomerulopathy utilizing phospholipase A2receptor (PLA2R1) staining in renal biopsies. Mod Pathol 2013; 26:709–715.

5. Beck LH, Bonegio RG, Lambeau G et al. M-type phospholipase A2receptor as target antigen in idiopathic membranous nephropathy.N Engl J Med 2009; 361: 11–21.

6. Farquhar MG, Vernier RL, Good RA. An electron microscope study of theglomerulus in nephrosis, glomerulonephritis, and lupus erythematosus.J Exp Med 1957; 106: 649–660.

7. Coons AH, Kaplan MH. Localization of antigen in tissue cells. J Exp Med1950; 91: 1–13.

8. Burkholder PM. Complement fixation in diseased tissues. J Exp Med 1961;114: 606–620.

9. Gallo GR, Feiner HD, Katz LA et al. Nodular glomerulopathy associatedwith nonamyloidotic kappa light chain deposits and excessimmunoglobulin light chain synthesis. Am J Path 1980; 99:621–644.

10. Pullman JM, Ferrario F, Nast CC. Actual practices in nephropathology: asurvey and comparison with best practices. Adv Anat Pathol 2007; 14:132–140.

11. Nasr SH, Galgano SJ, Markowitz GS et al. Immunofluorescence onpronase-digested paraffin sections: a valuable salvage technique for renalbiopsies. Kidney Int 2006; 70: 2148–2151.

160 Kidney International (2014) 86, 154–161

c l i n i c a l i n v e s t i g a t i o n CP Larsen et al.: Masked glomerulopathy

12. Molne J, Breimer ME, Svalander C. Immunoperoxidase versusimmunofluorescence in the assessment of human renal biopsies. Am JKidney Dis 2005; 45: 674–683.

13. Larsen CP, Bell JM, Harris AA et al. The morphologic spectrum and clinicalsignificance of light chain proximal tubulopathy with and without crystalformation. Mod Pathol 2011; 24: 1462–1469.

14. Servais A, Noel LH, Roumenina LT et al. Acquired and geneticcomplement abnormalities play a critical role in dense deposit diseaseand other C3 glomerulopathies. Kidney Int 2012; 82: 1–11.

15. Miura N, Mori Y, Yoshino M et al. Membranous nephropathy (bubblingappearance and spike formation) without immunoglobulin deposition in apatient with systemic lupus erythematosus. Clin Exp Nephrol 2008; 12: 475–478.

16. Nasr SH, Satoskar A, Markowitz G et al. Proliferative glomerulonephritiswith monoclonal IgG deposits. J Am Soc Nephrol 2009; 20: 2055–2064.

17. Bridoux F, Binaut R, Zanetta G et al. Glomerulopathy with non-organizedand non-Randall type monoclonal immunoglobulin deposits: a rareentity. J Am Soc Nephrol 2001; 12: 94A.

18. Komatsuda A, Masai R, Ohtani H et al. Monoclonal immunoglobulindeposition disease associated with membranous features. Nephrol DialTransplant 2008; 23: 3888–3894.

19. Touchard G, Aucouturier P, Hermine O et al. Ultrastructural pattern andclassification of renal monoclonal immunoglobulin deposits. MonoclonalGammopathies Kidney 2003; 10: 95–117.

20. Debiec H, Hanoy M, Francois A et al. Recurrent membranousnephropathy in an allograft caused by IgG3kappa targeting the PLA2receptor. J Am Soc Nephrol 2012; 23: 1949–1954.

21. Walker PD, Cavallo T, Bonsib SM. Practice guidelines for the renal biopsy.Mod Pathol 2004; 17: 1555–1563.

Kidney International (2014) 86, 154–161 161

CP Larsen et al.: Masked glomerulopathy c l i n i c a l i n v e s t i g a t i o n