Membranous Lupus Nephritis in a Renal Allograft: Response to Mycophenolate Mofetil Therapy

Authors

  • Mark D. Denton,

    Corresponding author
    1. a Renal Division, Department of Medicine, and b Department of Pathology, Brigham and Women's Hospital, Harvard Medical School, 75 Francis Street, Boston, MA 02115, USA
    Search for more papers by this author
  • a, Eleonora G. Galvanek,

    1. a Renal Division, Department of Medicine, and b Department of Pathology, Brigham and Women's Hospital, Harvard Medical School, 75 Francis Street, Boston, MA 02115, USA
    Search for more papers by this author
  • b Ajay Singh,

    1. a Renal Division, Department of Medicine, and b Department of Pathology, Brigham and Women's Hospital, Harvard Medical School, 75 Francis Street, Boston, MA 02115, USA
    Search for more papers by this author
  • and a Mohamed H. Sayegh a

    1. a Renal Division, Department of Medicine, and b Department of Pathology, Brigham and Women's Hospital, Harvard Medical School, 75 Francis Street, Boston, MA 02115, USA
    Search for more papers by this author

Abstract

Membranous lupus nephritis in a renal allograft is considered rare. A 43-year-old man with quiescent systemic lupus erythematosus (SLE) received a HLA identical transplant from his sister and 4 years later developed persistent nephrotic range proteinuria and morphological features most compatible with membranous lupus nephritis on biopsy. Angiotensin-converting enzyme (ACE) inhibitors and angiotensin II receptor antagonists, although successful in reducing proteinuria, were associated on three occasions with acute allograft dysfunction. Sustained reduction of proteinuria and stable graft function were achieved using mycophenolate mofetil (MMF). MMF is emerging as a new therapy for primary renal disease in SLE. This is the first report of successful treatment of membranous lupus nephritis in an allograft using MMF. We review all cases of transplant-associated membranous lupus nephritis in the English literature.

Introduction

In the United States, it is estimated that systemic lupus erythematosus (SLE) accounts for 2–4% of patients entering end-stage renal disease (ESRD). The majority of these patients ultimately undergo transplantation. In general, the clinical and serologic activity of SLE diminishes with ESRD and recurrence of lupus nephritis in the renal allograft is rare: 2.7% in a recent meta-analysis (1). A report from the Renal Allograft Disease Registry showed that recurrent lupus nephritis accounted for less than 2% of all forms of recurrent glomerular disease post-transplantation (2). Furthermore, only a few cases of membranous lupus nephritis in renal transplants have been reported. Given the rarity of this condition, there is little information in the literature on the optimum management of membranous lupus nephritis in transplants. In this report, we present a case of membranous lupus nephritis with persistent nephrotic range proteinuria who had a sustained reduction in the level of proteinuria following an increase in dosage of mycophenolate mofetil (MMF). We review all cases of transplant-associated membranous lupus nephritis documented in the MEDLINE literature.

Case Report

A 43-year-old Caucasian man with ESRD secondary to SLE received a HLA identical renal allograft from his sister in July 1996. He was diagnosed with SLE in 1983 after developing lethargy, arthralgia, a typical photosensitive malar rash and positive antinuclear antibody (ANA) and dsDNA. In 1986 he presented with acute renal failure associated with a nephritic urinary sediment and reduced C3, C4 and CH50 and positive dsDNA. A diagnosis of lupus nephritis was made but a renal biopsy was not performed at this time. He required temporary hemodialysis and was treated with cyclo-phosphamide and corticosteroids. He reached end-stage renal failure 10 years later and received hemodialysis for 3 months prior to transplantation. The transplant was technically uncomplicated. He had good primary function, no acute cellular rejection and reached a baseline creatinine of 1.5 mg/dL. He was maintained on standard triple therapy immunosuppression: Neoral 200 mg b.i.d., MMF 750 mg b.i.d., prednisone 7.5 mg q.d. In March 1998 he was noted to have 3+ proteinuria on routine urinalysis. He had no symptoms of active SLE, complement levels were normal and dsDNA and lupus anticoagulant was negative. Hepatitis B and C were negative. He was hypercholesterolemic (total cholesterol 263) and atorvastatin was increased to 20 mg q.d. Enalapril was initiated but this had to be discontinued due to an acute increase in creatinine from 1.5 to 2.3. At his next clinic visit he had a urine/protein creatinine ratio of 4.6 and a 24-h urine protein excretion of 4 g. On two occasions an angiotensin II receptor antagonist (ARB) was started with initial improvement in proteinuria but deterioration in allograft function leading to discontinuation of the drug. Figure 1 illustrates changes in creatinine, proteinuria, blood pressure and medication over successive clinic visits. Cyclosporin A dosage and serum levels remained constant throughout this period. A transplant artery angiogram showed no evidence of stenosis. A transplant biopsy was performed and tissue samples taken for light microscopy, immunofluorescence and electron microscopy. The 23 glomeruli present showed normal cellularity with minimal thickening of the capillary walls (Figure 2). There was no interstitial infiltrate and only minimal local interstitial fibrosis with associated tubular atrophy involving 5–10% of the sample. Arterioles showed mild sclerosis of their walls, particularly in the glomerular hilus. By immunofluorescence, there was intense staining for IgG in a fine granular pattern along the capillary walls with a weak reaction for C3 in the same pattern and distribution. Ultrastructural studies revealed diffuse subepithelial, with occasional mesangial, paramesangial and subendothelial electron-dense deposits (Figure 3). No tubular reticular structures were seen. A diagnosis of a membranous nephropathy, most likely membranous lupus nephritis, was made. Given the intolerance to both angiotensin-converting enzyme inhibitor (ACEI) and ARB, the dose of MMF was increased to a daily dose of 3 g in an attempt to modify the disease process. The total daily dose was reduced to 2.5 g 1 month later on account of mild diarrhea. The increase in MMF resulted in a rapid and sustained reduction in proteinuria (Figure 1).

Figure 1.

Proteinuria and serum creatinine plotted against time. Horizontal bars depict time and duration of antiproteinuric therapy. All other medications remained constant over this period. There was no difference between the mean cyclosporin A (CsA) level or white blood cell count (WBC) before and after change in mycophenolate mofetil (MMF) dosage (CsA level, 268 ± 42 vs. 243 ± 36; WBC: 12 ± 2 vs. 13 ± 1.5). Mycophenolic acid levels were not measured.

Figure 2.

Glomerulus showing minimal prominence of capillary wall basement membrane, and no increase in cellularity. Hematoxylin and eosin stain (×400).

Figure 3.

Electron micrograph illustrating electron-dense deposits (arrows) on the epithelial side and within the glomerular basement membrane (BM). Few deposits are seen in paramesangial and mesangial areas (MES). Epithelial cell (EP) foot processes were diffusely effaced. CL, capillary lumen; END, endothelial cells. (×9700).

Discussion

The differential diagnosis of nephrotic range proteinuria in a transplant recipient includes transplant glomerulopathy (a manifestation of chronic rejection), de novo membranous glomerulopathy, HCV-associated membranoproliferative glomerulonephritis (MPGN) and recurrence of primary glomerular disorders. In the case reported, transplant glomerulopathy was not considered likely clinically due to the fact that graft function was well preserved and that the allograft was a HLA identical living donor transplant that exhibited good primary function and had no risk factors for chronic rejection. HCV-associated glomerular disease was excluded by negative hepatitis serology, normal complement and negative cryoglobulins. De novo membranous glomerulopathy is second only to transplant glomerulopathy as the most common cause of nephrotic syndrome in renal transplant patients. Indeed, biopsy findings were consistent with a membranous nephropathy. However, despite absent clinical and serological evidence for active SLE, we thought the morphologic findings favored a membranous lupus over de novo membranous nephropathy. This was mostly based on the presence of electron-dense deposits in the mesangial, paramesangial and occasionally subendothelial areas. In native kidney biopsies, mesangial deposits are observed in fewer than 10% of primary membranous glomerulopathies but are seen in nearly all cases of membranous lupus nephritis (3,4). The lack of staining for IgA and C1q in the case reported was not thought to exclude a diagnosis of lupus membranous. Although a ‘full house’ of immunoglobulins is typical in type IV lupus nephritis, frequently this feature is not seen in membranous lupus nephritis (5). Additional features that would favor a diagnosis of membranous lupus over de novo membranous would be the presence of tubuloreticular structures on electron microscopy or immune complex deposits along tubular basement membranes (6). In summary, the histological findings were thought to be most consistent with a membranous lupus nephropathy rather than de novo membranous nephropathy; however, the latter diagnosis could not be fully excluded.

Using a MEDLINE literature search with the key words ‘transplantation’, ‘lupus’, and ‘SLE’, we reviewed all studies published between 1985 and 1999 reporting the recurrence rate post-transplantation of patients with ESRD secondary to SLE. We identified 21 single-center studies reporting the recurrence rates of lupus nephritis in renal allografts. A total of 710 patients with ESRD secondary to SLE were transplanted. All studies had a mean follow-up period of greater than 2 years. The numbers of lupus patients transplanted within each center ranged from six to 97. A histologic diagnosis of membranous lupus nephritis post-transplantation was established in 7(1.0%) patients. An additional 4 cases of membranous lupus nephritis were identified from case reports. Histologic diagnosis was based on immunofluorescence and electron microscopy data in the majority of cases. Table 1 lists the clinical features of these 11 cases of membranous lupus nephritis in renal allografts together with the present case (7–15). Other than young age, there were no consistent characteristic features defining this patient population. Patients differed widely in the rapidity of onset of ESRD, time on hemodialysis, time to recurrence, and clinical presentation. Little information was available on the treatment and outcome of these patients. Of note, most patients did not have clinical or serologic evidence of active lupus at the time of recurrence. This emphasizes that physicians should be vigilant to the possibility of recurrent lupus nephritis even in the absence of systemic disease. Membranous lupus nephritis is less commonly reported in renal allografts than other histological variants of lupus nephritis. This may be because patients with isolated proteinuria and normal function are less likely to have a transplant biopsy. Furthermore, immunofluorescence and electron microscopy are not routinely performed on transplant biopsies. Our case highlights the importance of electron microscopy in making an accurate diagnosis of recurrent membranous lupus nephritis. An alternative explanation is that the histology of the recurrent disease tends to mirror that of the primary disease and patients with membranous lupus are less likely to reach ESRD. However, as shown in Table 1, several patients found to have membranous lupus nephritis in the allograft had diffuse proliferative lupus nephritis on native renal biopsy (four out of six in which primary renal histology was reported). Fernandez et al. (13) described transformation of recurrent membranous lupus nephritis, in an allograft, into diffuse proliferative associated with loss of allograft function.

Table 1. : Clinical features of reported cases of membranous lupus nephritis in renal allografts
AuthorAgeSexPrimary
diagnosis
Time to
ESRD
(years)
Time
on HD
(years)
TxISYears to
recurrence
Clinical
flare
SerologyLoss of
function
Proteinuria
(g/24 h)
TreatmentOutcome/
follow-up
  1. A, azathioprine; Bx, biopsy; C, cyclosporin; HD, hemodialysis; inc., increase; IS, immunosuppression; MP, methylprednisone; NR, not reported; P, prednisone; Tx, transplant.

Ward et al. (7)26MV33LRDP/A8YesPos.Yes9MP/i.v. cytoxanStable
Liano et al. (8)38MIV< 1< 1CADP+?1.5NoNoNRYesNRDeath
Lanzuica et al. (9)29MV< 14CADP/A/C< 1NoNeg.Yes6NoneStable 1 year
Stone et al. (10)30NRNRNRNRLURDNR< 1NoNeg.NRNRNRStable
Stone et al. (10)25NRNRNRNRCADNR7NoNeg.NRNRNRESRD
Azevedo et al. (11)NRNRNo Bx17NRNRP/A/CNRNoNRNRNRNRNR
Azevedo et al. (11)NRNRNo Bx12NRNRP/A/CNRNoNRNRNRNRNR
Grimbert et al. (12)32FNRNK6LRDP/A/C< 0.5NoPos.Yes20MP/inc. PRemission
Fernandez et al. (13)29FIVNR1CADP/C2YesPos.No7.4Inc. ISESRD 1 year
Sumrani et al. (14)NRNRVNRNRLRDNR8NRNRNRNRNRNR
Nyberg et al. (15)30FNR51CADP/C1.5YesPos.Yes?MP/inc. PRemission
Denton et al. (16)45MNo Bx10< 1LRDP/M/C4NoNeg.No4MMFRemission

There is no information on the treatment of membranous lupus nephritis in renal allografts. In this case, initial treatment was directed to reducing the level of proteinuria by blocking the effect of angiotensin II either with ACEI or ARB. As illustrated in Figure 1, these agents were successful in reducing proteinuria but on three occasions were associated with an acute elevation in serum creatinine. To investigate this further, angiography of the transplant artery was performed but did not show a significant stenosis. Interestingly, an incidental finding on the renal biopsy was the presence of arteriolar hyalinosis. Given the young age of the donor, these vascular changes were presumed to have arisen post-transplantation and were probably secondary to the combined effects of hypertension and cyclosporin toxicity. We speculate that this renal arteriolar disease was responsible for the reversible reduction in graft function associated with ACEI and ARB. Limited increases in serum creatinine are frequently observed following initiation of ACEI/ARB and may be tolerated if beneficial effects on urinary protein excretion are observed. However, in this case the near doubling of serum creatinine (halving of glomerular filtration rate, GFR) was considered unacceptable and long-term ACEI/ARB was avoided.

Modification of lupus activity using immunosuppressive agents was the remaining treatment option available. We found that increasing the daily dose of MMF from 1.5 g to, initially, 3 g and later 2.5 g resulted in a sustained reduction in proteinuria from over 4 g/24 h to less than 0.5 g. This was not associated with myelotoxicity and no diarrhea occurred at the 2.5 g dose level. The close temporal association between increasing MMF dose and the sustained reduction in proteinuria argues that the clinical observation was a drug effect rather than the natural history of this condition. Indeed, whereas a third of patients with idiopathic membranous glomerulonephritis may undergo remission spontaneously, this is not the natural history of membranous lupus nephritis.

Although this is the first report describing a benefit of MMF in membranous lupus nephritis in an allograft, increasing evidence supports a role for MMF in the management of SLE and glomerular diseases in general. MMF interrupts the de novo pathway of purine synthesis and is a potent inhibitor of T- and B-cell proliferation, antibody production and generation of cytotoxic T cells (16). MMF has been shown to reduce the severity of disease in an animal model of lupus nephritis (17). In a case series of patients with diffuse proliferative lupus nephritis, MMF was shown to improve creatinine and reduce proteinuria in the majority of patients (18). MMF may also reduce proteinuria in patients with idiopathic membranous nephropathy resistant to steroids, cytotoxic agents and cyclosporin A (19). Finally, a recent trial found that the combination of MMF and steroids was as effective as a regimen of cyclophosphamide and steroids in the treatment of diffuse proliferative lupus nephritis (20).

In summary, we present a case of presumed recurrent membranous lupus nephritis in a renal allograft that showed a substantial clinical response to MMF. This case provides several interesting teaching points. First, membranous lupus nephritis in an allograft frequently presents insidiously with little clinical or serologic manifestation of active lupus. Second, diagnosis requires careful attention to the immunofluorescence and electron microscopy and these should be routinely done on transplant biopsies of patients with a prior history of SLE. Third, MMF may be a valuable agent to manage both primary and recurrent lupus nephritis.

Ancillary