Membranous Nephropathy With Crescents

Membranous nephropathy (MN) is a common cause of nephrotic syndrome in adults and can be primary or secondary. Primary MN is most commonly associated with anti−M-type phospholipase A2 receptor (PLA2R) antibodies and is usually IgG4 dominant, whereas secondary MN can be seen in the setting of malignancies, infections, autoimmune diseases, or as a side effect of certain medications or toxins, and is most commonly IgG1 dominant.^1,2 The occurrence of crescents in MN is extremely rare, with a prevalence estimated between 0.39%³ and 0.26%.⁴ Crescents can be seen in both primary and secondary MN3, 4, 5 or in association with a superimposed disease process, such as anti−neutrophil cytoplasmic antibody (ANCA)−associated glomerulonephritis or anti−glomerular basement membrane (GBM) disease. Previous series in the United States have either focused on cases of MN with crescents in the absence of ANCAs or anti-GBM antibodies,³ or in association with ANCAs.⁵ In light of the recently published series of MN with crescents from the United Kingdom,⁴ and given the rarity of this entity, we analyzed its prevalence and its clinical and pathological characteristics in biopsy findings reviewed at our institution.

A total of 14,800 native and transplant renal biopsy specimens were received at the Ohio State University from 2010 to 2019. Of these, 15 cases (0.1%) showed MN with crescents (fibrous crescents only, 3; diffuse crescents [in >50% of glomeruli], 3; focal crescents [in <50% of glomeruli], 9). In addition, 3 cases had only segmental subepithelial deposits (segmental MN). Cases of lupus nephritis were excluded. PLA2R staining was performed retrospectively in 7 cases.

There were 8 female and 7 male patients. Two had anti-GBM antibodies (Figure 1a–d), 9 had ANCAs (8 MPO, 2 both MPO and PR3), and 4 showed anti-PLA2R−positive deposits in the biopsy specimen (Figure 1e–g), including 1 with positive serum anti-PLA2R. Of note, not all of these antibodies were checked in all patients. Clinical features are summarized in Table 1 and pathological features in Table 2. One patient had prior ANCA-associated cutaneous vasculitis, and 1 patient had prior biopsy-proven infection-related glomerulonephritis; both had a history of cocaine use. Potential secondary causes of MN included solid organ malignancies (n =3), hepatitis C (n = 2), and rheumatoid arthritis (n = 1). Three patients had a monoclonal gammopathy of undetermined significance, and 5 patients had positive autoantibodies (4 anti−nuclear antibodies, 2 anticardiolipin, and 2 anti-SSB). Eight patients had pulmonary symptoms, including hemoptysis in 3 and documented diffuse alveolar hemorrhage in 2. Twelve patients had hematuria, which was usually significant, and 14 had proteinuria (1 patient was anuric). Proteinuria was measured by spot urine protein-to-creatinine ratio or 24-hour collection and was >3 g in 7 of the 14 patients. Median creatinine at presentation was 2.9 mg/dl (1.07−18.6). Two patients had normal renal function on presentation (1 patient had a single fibrous and the other a single cellular crescent). The remaining 13 patients had estimated glomerular filtration rate of <60 ml/min per 1.73 m². Eleven patients presented with severe kidney dysfunction (eGFR ≤30 ml/min per 1.73 m²), 4 of whom required dialysis (2 with anti-GBM and 2 with MPO). Twelve patients received immunosuppression, 2 did not (1 had end-stage renal disease [ESRD]), and the other had a single fibrous crescent on biopsy); treatment data were not available for 1 patient. The most common drugs used were corticosteroids (n = 11) and cyclophosphamide (n = 10), followed by mycophenolate mofetil (MMF) (n = 4), azathioprine (n = 1), and rituximab (n = 1). Plasmapheresis was performed in 3 patients: 2 with anti-GBM and 1 with myeloperoxidase (MPO) antibodies and hemoptysis. Follow-up data were available for 13 patients, with a median follow-up of 38 months (4−98 months). Of these 13 patients, 5 reached ESRD (3 MPO, 2 anti-GBM), of whom 3 received a kidney transplant. Of the patients who reached ESRD, 2 had >50% crescents on biopsy, and 1 patient had >90% interstitial fibrosis and tubular atrophy (IFTA); the other 2 patients had up to 25% crescents and 25% to 35% IFTA (Table 2). Five patients developed chronic kidney disease, 2 died (both had solid organ malignancies), and only 3 (20%) had eGFR >60 ml/min 1.73 m² (all of these patients had only focal crescents and mild IFTA on biopsy). Conversely, renal survival in MN without crescents is estimated to be between 70% and 90%.⁶

Figure 1.

Top row (a–d): a case of anti–glomerular basement membrane (GBM) disease (case 12) with segmental membranous nephropathy (MN). (a) Glomerulus with a cellular crescent (hematoxylin and eosin [H&E] stain, original magnification ×400). (b) Silver stain shows a compressed glomerular tuft with no obvious spikes or lucencies along the GBM (original magnification ×4000). (c) Linear IgG staining (immunofluorescence, original magnification ×400). (d) Electron microscopic image showing segmental subepithelial deposits. Bottom row (e−h): a case of PLA2R-positive MN with concomitant p-ANCA (case 8). (e) Glomerulus with a cellular crescent (H&E stain, original magnification ×400). (f) Silver stain highlights capillary wall irregularities (fine lucencies and spikes) in the same glomerulus that contains a cellular crescent (original magnification ×400). (g) An anti-PLA2R antibody shows diffuse granular positivity along the GBM (immunofluorescence, original magnification ×400). (h) Electron microscopy confirms diffuse subepithelial deposits.

Table 1.

Clinical characteristics

Case	Sex	Age (yr)	Ethnicity	Baseline Cr (mg/dl)	Cr at biopsy (mg/dl)	GFR at diagnosis (ml/min per 1.73 m2)	Hematuria	Proteinuria at biopsy	Follow-up proteinuria	Albumin (g/dl)	Edema	Lung symptoms	Medical comorbidities	Antibodies	Other relevant labs	Treatment	Dialysis	Transplant	Cr at last follow-up (mg/dl)	GFR at follow-up (ml/min per 1.73 m2)	Follow-up duration (mo)
1	F	68	White	1.6	9	4	3+	4 g/24 h	NA	2.5	Yes	SOB	Rheumatoid arthritis, NSAID use	MPO	NA	Cyclophosphamide, prednisone complicated by aspergillus pneumonia	Yes	Yes	NA	NA	NA
2	F	21	White	NA	15	3	Anuric	Anuric	ESRD	NA	NA	DAH, acute respiratory failure	NA	Anti-GBM	NA	PLEX, prednisone, and cyclophosphamide	Yes	No	ESRD	ESRD	47
3	M	65	White	0.9	2.2	30	>100 RBCs/HPF	2.89g/24h	ESRD	3.7	No	Hemoptysis	Raynaud syndrome	pANCA, MPO	ANA 1:80	prednisone, MMF switched to cyclophosphamide, PLEX, then azathioprine and prednisone	Yes	Yes	ESRD (5.38) after transplant nephrectomy	ESRD	98
4	F	64	AA	2	2.5	24	Moderate, 20−29 RBCs/HPF	1.6 g/24h	1.1 g UPC	2.2	Yes	Hemoptysis, DAH	Diabetes, hep C cirrhosis, HCC	pANCA, MPO, PR3	ANA 1:320, +anticardiolipin IgM, c3 77, c4 14	Prednisone and cyclophosphamide then MMF	No	None	4.33 Deceased	12	13
5	F	62	Asian	NA	1.09	51	NA	16.9g/24h	NA	0.6	Yes	NA	Liver dysfunction	NA	NA	None	NA	NA	1.14	49	79
6	F	61	AA	0.8	9	5	Few RBCs	8 g UPC	1.6 g/ 24 h	1.4	Yes	Pulmonary nodule	HTN, CKD, heavy smoker	NA	NA	Prednisone and cyclophosphamide	No	No	1.32	49	10
7	F	66	AA	1.3	2.9	20	NA	3.5 g UPC	NA	NA	NA	NA	Bladder cancer, MGUS	pANCA	NA	NA	NA	NA	NA	NA	NA
8	M	70	White	1.3	2.35	28	>20 RBCs/HPF	2 g/24 h	1.3 g UPC	3.4	No	Cough	Previous MN with crescents, MGUS, resected pancreatic neoplasm, metastatic sarcomatoid carcinoma, HTN, DM	pANCA, MPO, PR3	Minute IgG K	Prednisone and cyclophosphamide	No	No	1.99 Deceased	33	14
9	M	79	White	1.5	3	21	21−30 RBCs/HPF	1 g UPC	0.2 g UPC	4.5	No	None	HTN, MGUS, diabetes	pANCA, MPO	None	Cyclophosphamide, prednisone, then MMF	No	No	2.02	32	46 mo
10	M	56	AA	1.07	1.07	>60	Large	4.7 g/24 h	0.7 g/ 24 h	2.7	Yes	None	Diabetes, HTN	NA	None	Cyclophosphamide	No	No	0.83	>60	49 mo
11	F	42	NA	1	1.8	NA	>60 RBCs/HPF	1.8 g UPC	0.2 g UPC	1.7	NA	NA	Prior MPO-ANCA cutaneous vasculitis, history of cocaine abuse, positive MRSA skin culture	MPO	Positive ANA and anti-SSB	Prednisone, MMF	NA	NA	0.8	>60	41 mo
12	M	26	White	NA	18.6	3	Large	2.8 g/24 h	0.07 g UPC	3.6	None	Hemoptysis	Obesity	Anti-GBM >8	None	PLEX, steroids, cyclophosphamide	Yes	Yes	2.01	39	38 mo
13	M	34	AA	1.03	1.18	93	45 RBCs/HPF	1.8 g UPC	1.2 g UPC	3.8	NA	NA	HTN, obesity	pANCA, MPO	ANA 1:40	Prednisone, rituximab	No	No	0.98	113	28 mo
14	M	41	White	1.8	10.9	5	Large	9.5 g UPC	ESRD	2.5	None	NA	Prior infection-related GN, i.v. drug use (including cocaine), hep C, endocarditis, seizure	pANCA, MPO	Positive anticardiolipin and anti-SSB	None	Yes	No	ESRD (10.2)	ESRD	4 mo
15	F	20	AA	0.83	3	24	3−5 RBCs/HPF	24.8 g/24 h	3.4 g UPC	1.4	Yes	SOB	Schizophrenia, bipolar disorder, Bell palsy, obesity	Anti-PLA2R antibody	Low C4 (17), normal C3	Prednisone, cyclophosphamide	No	No	1.51	52	17 mo

AA, African American; ANA, anti−nuclear antibodies; ANCA, anti−neutrophil cytoplasmic antibodies; Cr, creatinine; C3, complement factor 3; C4, complement factor 4; CKD, chronic kidney disease; DAH, diffuse alveolar hemorrhage; DM, diabetes mellitus; ESRD, end-stage renal disease; F, female; GBM, glomerular basement membrane; GFR, glomerular filtration rate; GN, glomerulonephritis; HCC, hepatocellular carcinoma; hep C, hepatitis C; HPF, high-power field; HTN, hypertension; labs, laboratory investigations; M, male; MGUS, monoclonal gammopathy of undetermined significance; MMF, mycophenolate mofetil; MN, membranous nephropathy; MPO, myeloperoxidase; MRSA, methicillin-resistant Staphylococcus aureus; NA, not available; NSAIDs, nonsteroidal anti-inflammatory drugs; PLA2R, M-type phospholipase A2 receptor; PLEX, plasma exchange; PR3, proteinase 3; RBCs, red blood cells; SOB, shortness of breath; SSB, anti-Sjögren's syndrome type B; UPC, urine protein-to-creatinine ratio.

Table 2.

Pathological characteristics

Case	Final diagnosis	Total glomeruli	Crescents/FN	GS	% Active lesionsa	IFTA%	Immunofluorescenceb	IgG1	IgG2	IgG3	IgG4	PLA2R	EM stage	Mesangial deposits	TRIs	Extraglomerular deposits
1	ANCA-associated crescentic and necrotizing GN; membranous GN	16	4	8	25	25	IgG, IgA, C3, kappa, lambda GBM	1	0	1.5	0.5	Negative	NA	NA	NA	0
2	Diffuse crescentic and necrotizing anti-GBM disease; segmental MN	21	18	0	85	0	Linear GBM IgG, kappa and lambda; linear TBM IgG (focal)	2.5 Linear	1.5 Linear	0	3 Linear	Negative	1	0	0	0
3	Focal crescentic GN with MN	30	1	10	3	35	IgG, IgA, C3, kappa, lambda GBM and mesangial; IgG, kappa, lambda TBM	1.5	1	0	0.5	Negative	1 to 2	1	0	TBM deposits by IF
4	MN with fibrous crescents; underlying diabetic glomerulosclerosis	24	4 (Fibrous)	10	0	60-70	IgG, IgM, C1q, C3, kappa and lambda GBM, IgG TBM (focal)	1.5	0.5	0.5	2	Positive	3 to 4	0	0	TBM deposits by IF
5	MN with fibrous crescents	34	1 (Fibrous)	9	0	20	IgG, C3, kappa, lambda GBM	3	2	0	2	Negative	2	0	0	0
6	MN with focal crescents	13	3	0	23	20-25	IgG, IgA, IgM, C1q, C3, Kappa and lambda GBM	2	1	2	3	Weak and segmental	2 to 3	0	0	0
7	ANCA-associated crescentic and necrotizing GN with MN	33	8	12	24	60	IgG, IgM, kappa, lambda GBM	2	1	0.5	3	Positive	2 to 3	0	0	0
8	ANCA-associated crescentic and necrotizing GN; membranous GN	12	6	0	50	30	IgG, C3, kappa, lambda GBM	1.5	0.5	1	0	Negative	1 to 4	0	0	0
9	ANCA-associated necrotizing and crescentic GN with MN and TBM deposits	5	1	0	20	35	IgG, kappa, lambda GBM; IgG, C1q, kappa, lambda TBM (focal)	1	0.5	0	2	Negative	1 to 3	0	0	TBM deposits by IF and EM
10	MN with focal crescents	12	1	2	8	5	IgG, C3, kappa and lambda	3	1	2	3	Positive	2	0	0	0
11	Immune-complex GN with segmental MN and focal necrotizing lesions	13	2	0	15	10	IgG, IgM, C3, kappa and lambda GBM and segmental mesangial	2	0.5	0.5	0	Negative	2	Rare	1	0
12	Diffuse crescentic and necrotizing anti-GBM disease; segmental MN	23	18	0	78	25	Linear IgG, IgA, C3, kappa and lambda	2 Linear	1 Linear	1 Linear	2 Linear	Negative	1 to 2	0	0	0
13	MN with focal crescents	42	1	7	2	20	IgG, C3, kappa and lambda GBM	2	1	0.5	2	Negative	1 to 2	0	0	0
14	Advanced chronic renal injury with underlying MN; acute TMA	32	5 (Fibrous)	29	0	90	IgG, IgM, C3, kappa and lambda GBM and mesangial	1	0.5	0.5	0	Negative	2	1	0	0
15	MN with focal crescents	9	3	0	33	20	IgG, C3, kappa and lambda GBM and mesangial	3	1	2	3	Positive	2 to 4	1	1	1

ANCA, anti−neutrophil cytoplasmic antibodies; C3, complement factor 3; C4, complement factor 4; Ig, immunoglobulin; EM, electron microscopy; FN, fibrinoid necrosis; GBM, glomerular basement membrane; GN, glomerulonephritis; GS, globally sclerotic glomeruli; IFTA, interstitial fibrosis and tubular atrophy; MN, membranous nephropathy; PLA2R, M-type phospholipase A2 receptor; NA, not available; TBM, tubular basement membrane; TMA, thrombotic microangiopathy; TRIs, tubuloreticular inclusions.

^aActive lesions include cellular or fibrous−cellular crescents and areas of glomerular segmental fibrinoid necrosis.

^bDeposits are granular unless otherwise specified.

Overall, our series findings are similar to those recently reported by Nikolopoulou et al.,⁴ with approximately 40% of patients reaching ESRD. We had a greater number of anti-PLA2R−positive cases (26% vs. 13%), whereas they had a greater number of anti-GBM−positive patients (33% vs. 13%). Interestingly, however, outcomes were similar, which may be due to the fact that the prognosis of anti-GBM disease tends to be better in cases with associated MN,⁷ with close to 40% of patients recovering renal function⁸ in contrast to only 15% in pure anti-GBM disease.S1 Potential explanations for that could be overall lower levels of anti-GBM antibodies in patients with concomitant MN as well as a narrower antigen reactivity spectrum of the anti-GBM antibodies present.⁷

In contrast to MN without crescents, in which case >90% of patients have normal renal function at presentation¹ and hematuria is generally microscopic and low-grade,S2 most of our patients had significant hematuria, and approximately 75% presented with severe kidney dysfunction. In addition, although approximately 80% of cases of MN without crescents are considered primary, in our series only 26% of cases were PLA2R positive, and 40% were IgG4 dominant/co-dominant. Therefore, it appears that crescents occur more often in cases of secondary MN.

In conclusion, MN with crescents is a rare and heterogeneous entity that can be associated with ANCA, anti-GBM, PLA2R, and potentially other autoantibodies. It presents more often with significant hematuria and renal dysfunction than MN without crescents and progresses more often to ESRD. Whether these cases represent a coincidental occurrence of 2 separate disease entities or whether they are pathogenically related remains to be determined,⁸ although the latter is conceivable. For example, subepithelial deposits may facilitate GBM damage, leading to anti-GBM antibody production. Conversely, GBM damage caused by anti-GBM (or other) antibodies could expose epitopes that lead to immune-complex deposition along the subepithelial aspect of the GBM.

Disclosure

All the authors declared no competing interests.