Letter to the Editor

A case of membranous nephropathy associated with chronic GVHD successfully treated with rituximab

The kidney is a rare target organ for chronic GVHD (cGVHD) in humans. Renal involvement in cGVHD has been reported to occur in 0.5–1.0% of hematopoietic cell transplantation recipients.1 Most cases present as nephrotic syndrome about 1–5 years after transplantation, mostly while having immunosuppressants tapered.2 Membranous nephropathy is the most common pathology on renal biopsy.2 Many cases respond well to corticosteroid therapy and non-responsive cases are usually treated with additional immunosuppressants (for example, mycophenolate, calcineurin inhibitors).

The etiology of chronic renal GVHD is not yet fully elucidated and the role of B cells as critical factors has been studied in the etiology. In this study, we describe a patient who developed chronic renal GVHD 4 years after transplantation and was successfully treated with rituximab, an anti-CD20 MoAb, which indicates the importance of B cells in the pathophysiology of chronic renal GVHD.

The patient is a 51-year-old Japanese man, who developed Ph+ ALL in August 2004. He underwent a BM SCT from his HLA-identical female sibling on 25 November 2004 after conditioning with cytarabine (CA)-cyclophosphamide (CY)-TBI (8 g/m2 of CA, 100 mg/kg of CY and 12 Gy of TBI). He received prednisolone (PSL), CsA and short-term MTX as GVHD prophylaxis. Neutrophil engraftment was achieved on day +15.

On day +59, he developed acute GVHD of the gastrointestinal tract, grade III, which resolved instantly with corticosteroid increase.

On day +138, the patient developed nonproductive cough/dyspnea and was diagnosed with late-onset non-infectious pulmonary complication associated with chronic GVHD. Every time his respiratory state deteriorated, he was treated with corticosteroid increase and CsA. His interstitial pneumonia, however, chronically progressed to lung fibrosis and was so severe that he started home oxygen therapy in May 2007. Thereafter, he had been treated with PSL and tacrolimus and the dose of PSL was being tapered gradually.

Four years after the transplantation, in February 2009, he was admitted to our hospital with remarkable pitting edema on both his lower legs. He had gained 2 kg in 1 week before then. At this time, the dose was PSL 5 mg and tacrolimus 0.4 mg per day. Urine protein level was 10.36 g per day and serum albumin was 1.9 g/dL, which led to the diagnosis of nephrotic syndrome. Lab data were as follows: WBC 9240/μL, Hb 13.6 g/dL, Plt 245 × 103/μL, total protein 4.3 g/dL, blood urea nitrogen 15.1 mg/dL, Cr 0.95 mg/dL and total cholesterol 352 mg/dL.

Renal biopsy revealed the histology of membranous glomerulonephritis, the most common pathology of cGVHD nephropathy (Figure 1). Although typical findings of membranous glomerulonephritis, such as bubbling or spike, were not found under light microscopy, subepithelial deposits of immunocomplex were observed under EM and we diagnosed it as membranous nephropathy, Ehrenreich−Churg stage I–II. By immunofluorescence, granular capillary wall deposition pattern of IgG, C3 and C1q was seen. C1q deposition is assumed to be characteristic to immunologic etiology, including GVHD.

Figure 1
Figure 1

Pathology of renal biopsy. (a) Light microscopy (periodic acid-Schiff stain): no typical findings of membranous nephropathy (MN). (b) EM ( × 5000): subepithelial deposits of immunocomplex (arrows), which lead to the diagnosis of MN. (c, d) Immunofluorescence: granular capillary wall deposition pattern of IgG, C3 and C1q, indicating immunological etiology, including GVHD.

We increased the PSL dose to 40 mg (1 mg/kg) per day as a standard therapy of cGVHD nephropathy, but no response was observed for 2 weeks. In this patient, deterioration of interstitial pneumonia by respiratory infection could be critical; therefore high-dose corticosteroid for a long period was undesirable.

Then we chose instead B-cell depletion therapy with rituximab (375 mg/m2 per week i.v. for 4 consecutive weeks) and decreased the PSL dose rapidly. Three weeks after the first dose of rituximab, urine protein level began to decline (Figure 2). Since 6 months after the first dose, urine protein level has been <0.5 g per day and serum albumin level has been normal so far, which indicates a CR of nephrotic syndrome. Although he underwent various complications such as mesenteric vein thrombosis, portal vein thrombosis, hemorrhagic rupture of a pancreatic cyst due to anti-coagulative therapy and pneumothorax, his general status dramatically improved as he recovered from nephrotic syndrome. He is now prescribed PSL 5 mg and tacrolimus 0.4 mg per day, the same dose as that before the development of nephrotic syndrome.

Figure 2
Figure 2

Time line of clinical response to rituximab. Urine protein level was 6.3 g per day at the time of diagnosis. PSL increase (40 mg per day) for 2 weeks did not show any efficacy. In this case with severe interstitial pneumonia, respiratory infection could be critical and high-dose corticosteroid for a long period was undesirable. Therefore we chose rituximab (375 mg/m2 per week i.v. for 4 consecutive weeks) as a substitute therapy and decreased the PSL dose rapidly. Three weeks after the first dose of rituximab, urine protein began to decline and CR of nephrotic syndrome (urine protein level <0.5 g per day) has been maintained so far.

We have three reasons to ascribe his nephrotic syndrome to chronic GVHD. First, he originally suffered from lung cGVHD. Second, the dose of steroid was being tapered at the time of onset. The last reason is that we could detect no other cause, such as some malignancies, autoimmune disease or viral infection, including hepatitis B virus, hepatitis C virus and HIV.

With chronic renal GVHD, some patients die of deterioration in general status due to nephrotic syndrome refractory to corticosteroid. Another treatment option for such refractory cases has been anticipated and the etiology of cGVHD nephropathy is under study.

T cells have been traditionally considered to be major effectors and initiators in cGVHD. Recently B cells have been focused on as possible effectors in cGVHD, although their role is not completely understood yet.3, 4, 5, 6 Some specific autoantibodies have been found to be correlated with cGVHD, such as H-Y Ab.6 H-Y Ab to Y-chromosome-encoded mHAs, called H-Y Ags, is found in male hematopoietic cell transplant recipients with female donors.

As for cGVHD nephrotic syndrome, involvement of anti-nephrin Ab has also been suggested in an experimental mouse model.7 Nephrin is a component protein of slit diaphragms of the glomerular capillary wall and is crucial for glomerular filtration barrier. It was identified as a product of a gene linked to congenital nephrotic syndrome of the Finnish type.

These findings lead to the hypothesis that anti-B cell therapy is effective in cGVHD, whereas the standard therapy has been corticosteroid and immunosuppressants targeted at T cells. In fact, some cases of cGVHD treated with rituximab have been reported, especially in skin manifestations.2, 3, 4 Clinical observations that rituximab is useful in the treatment of cGVHD are themselves evidence for the importance of B cells in cGVHD.

Renal involvement in cGVHD is rare and the number of cases treated with rituximab has been naturally small. In our case, we chose rituximab as a substitute therapy for fear of deterioration of respiratory state due to infection, and he completely recovered from nephrotic syndrome without any infection. This is a valuable case that suggests the involvement of B cells in cGVHD nephropathy and presents the possibility of rituximab as another option for chronic renal GVHD.


  1. 1.

    , , , , , et al. Nephrotic syndrome associated with chronic graft-versus-host disease after allogeneic hematopoietic stem cell transplantation. Bone Marrow Transplant 2006; 38: 351–357.

  2. 2.

    , , , , . Renal pathology in hematopoietic cell transplantation recipients. Mod Pathol 2008; 21: 396–406.

  3. 3.

    , , , , , . Rituximab therapy for membranous nephropathy: a systematic review. Clin J Am Soc Nephrol 2009; 4: 734–744.

  4. 4.

    , , , , , et al. Rituximab for steroid-refractory chronic graft-versus-host disease. Blood 2006; 108: 756–762.

  5. 5.

    , , . B-cell involvement in chronic graft-versus-host disease. Haematologica 2008; 93: 1702–1711.

  6. 6.

    , , , . The role of B cells in the pathogenesis of graft-versus-host disease. Blood 2009; 114: 4919–4927.

  7. 7.

    , , , , , et al. Possible role of autoantibodies against nephrin in an experimental model of chronic graft-versus-host disease. Clin Exp Immunol 2005; 141: 215–222.

Download references

Author information


  1. Department of Nephrology, Osaka Red Cross Hospital, Osaka, Japan

    • E Iguchi
    •  & T Minakata
  2. Department of Hematology, Osaka Red Cross Hospital, Osaka, Japan

    • M Tsudo


  1. Search for E Iguchi in:

  2. Search for T Minakata in:

  3. Search for M Tsudo in:

Competing interests

The authors declare no conflict of interest.

Corresponding author

Correspondence to E Iguchi.