EXPERIMENTAL AND CLINICAL TRANSPLANTATION

Transplant is the optimal therapy for patients with end-stage renal disease. Acute cellular rejection refractory to treatment remains a major risk factor for graft loss and poor outcomes. In this study, we describe a 39-year-old man who received a living-related kidney transplant. Two days after transplant, the patient displayed acute deterioration of graft function. Conventional anti-rejection therapy was initiated, but graft function did not improve. Biopsy revealed acute cellular rejection (grade IIA), and C4d and HLA antibodies were negative. Immunohistochemistry phenotyping revealed clusters of CD20-positive lymphocytes, with 80% being CD3 positive. Rituximab was prescribed, and graft function improved dramatically. After 1 week, a second graft biopsy was done due to lagging of graft function, shown by serum creatinine of 2.1 mg/dL. Biopsy revealed regenerating acute tubular necrosis with disappearance of the CD20-positive lymphocyte cluster infiltrates. Two year, after transplant, the patient's graft function maintained stable. Phenotyping of the cellular infiltrate is important as it may lead to a proper selection of immunosuppression and consequent improvement of graft outcome.

Key words : Acute humoral rejection, B-cell-mediated rejection, CD20-mediated rejection, Renal transplantation, Rituximab

Transplant remains the optimal treatment for patients with end-stage renal disease.¹Acute graft rejection remains a common complication and affects long-term graft survival.^2,3 Advances in immunosuppressive protocols have led to improvements in acute graft rejection episodes.^3-6 However, acute graft rejection, especially antibody-mediated rejection (AMR), can cause graft loss in some patients and can shorten graft function time in others. The presence of CD20-positive cells in the graft has been associated with poor reversibility of graft function and a trend toward poor graft survival even without AMR.^7-9 Anti-CD20 antibodies (rituximab) have been used as rescue therapy for CD20-mediated refractory acute rejection with successful outcomes.^7,10-15 Herein, we report a case of refractory acute cellular rejection with clusters of CD20-positive lymphocyte graft infiltrates, resistant to steroid pulses, plasmapheresis, and switching of immunosuppressive drug regimen, which was successfully treated with anti-CD20 antibodies.

A 39-year-old-male with type 2 insulin-dependent diabetes presented with end-stage kidney disease and with both kidneys having small size. The patient had been on regular hemodialysis for 3 years.

The patient received a right iliac renal allograft from his 27-year-old brother. The sibling donor had a different compatible blood group, one mismatched HLA DR, negative repeated complement-dependent crossmatch test, and 7% donor nonspecific panel reactive antibody (PRA) class I and 0% PRA class II. There were no complications during the transplant procedure and postoperative period, with total ischemia time of 55 minutes and immediate diuresis. Induction immunosuppression was a monoclonal anti-CD25 alpha agent (basiliximab). The maintenance immunosuppression plan included a steroid-sparing protocol with tacrolimus (FK506) and everolimus.

On day 2 posttransplant, the patient had sudden cessation of urine output volume from 10 to 3.6 L/day. Tacrolimus and everolimus trough levels were within therapeutic ranges (7.8 ng/mL and 6.4 ng/mL, respectively). Doppler ultrasonography of the graft revealed absent diastolic wave over the whole graft (Figure 1A). The next day, urine output volume dropped to less than 1 L/day. Decreases in serum creatinine levels were stationary and then became lagging. A graft biopsy was done, and methylprednisone pulse steroid therapy was started (500 mg of methylprednisolone/day) on the same day. Because of aggressive deterioration of graft function, a serum sample was sent for donor-specific anti-HLA antibody (DSA) analyses (analysis by Luminex technology, Luminex, Austin, TX, USA).

The graft biopsy by light microscopy initially revealed significant clusters of interstitial infiltrates, vasculitis, foci of severe tubulitis, and acute tubular necrosis (Figure 2A). Steroid use and intensifying immunosuppressive drugs did not improve graft function; on the contrary, the biopsy revealed an accentuation of the interstitial inﬁltrates and vascular damage. These facts prompted the suspicion of AMR and led to initiation of plasmapheresis. Several days later, the full report of the biopsy was received, which showed absence of C4d staining and no circulating DSAs against HLA antigens. In addition, PRA was negative for class I and II and complement-dependent crossmatch was negative, reasonably ruling out the hypothesis of an AMR. This unexpected situation led us to request further evaluation of the graft biopsy.

Immunohistochemistry staining of consecutive sections with anti-CD20 (formally a marker for B cells) and anti-CD3 (T cell marker) revealed 90% of the cellular infiltrates were CD20 positive, with about 80% of the inﬁltrating lymphocytes composed of clusters of CD20 and CD3 double-positive T cells (Figure 2B). Acute T-cell-mediated rejection grade IIA (per Banff 200716) was diagnosed. In view of the negative results of HLA antibodies, the second plasmapheresis session was cancelled. Antithymocyte globulin would be the drug of choice in resistant T-cell-mediated rejection. However, because 90% of the cellular infiltrates were CD20 positive, the use of an anti-CD20 monoclonal antibody would be a more specific choice of therapy. The patient therefore received 375 mg/m² of rituximab. Graft function started to improve, creatinine dropped to 2.7 mg/dL from 4.1 mg/dL, and graft Doppler improved (Figure 1B). At day 12 posttransplant, the daily drop of serum creatinine was not satisfactory, lagging at 2.1 mg/dL.

A second graft biopsy was conducted, which revealed regenerating acute tubular necrosis by light microscopy and disappearance of the CD20-positive infiltrating lymphocyte clusters (Figure 3). Modification of tacrolimus and everolimus to lower therapeutic levels was done to avoid delayed recovery of acute tubular necrosis. Within 2 weeks of follow-up, serum creatinine gradually returned to a normal level of 1.2 mg/dL (Figure 4A). Two years of follow-up showed normal graft function with no protein urea (Figure 4B).

Currently, AMR is classically diagnosed by 3 criteria: detection of the complement split product C4d in peritubular capillaries, detection of DSA in the circulation, and functional and/or morphologic evidence of harm to the graft.16 Antibody-mediated rejection is not always associated with C4d, and criteria of C4d-negative AMR have been well-described in the revised version of Banff 2013.¹⁷

Refractory acute cellular rejection often presents with vascular damage and no measurable circulating DSA or C4d deposition by graft biopsy staining.^13-15 Cytotoxic T cells are usually present, and characterization of cellular graft inﬁltrate is increasingly important because presence of CD20-positive cells in the graft has been associated with poor reversibility and a trend toward poor graft survival.⁹ Studies have identiﬁed the presence of intrarenal CD20-positive B-cell clusters as a risk factor for steroid-resistant rejection and graft loss.^9,11,13,18

Traditionally, CD20 is considered a marker of active B cells. In fact, it is expressed in various types of immune cells, including Th1 T cells, Th17 effector T cells, activated B cells, and activated dendritic cells.^19-21 In this report, we presented a patient who had acute cellular rejection that was resistant to conventional antirejection therapy. Rejection was successfully reversed with speciﬁc monoclonal antibodies against CD20. As reported by Aranda and associates,²² humor rejection in a cardiac transplant patient partially responded to steroids, muromonab-CD3, cyclophosphamide, and plasmapheresis. However, the patient's condition eventually improved with the addition of monoclonal anti-CD20 antibody (rituximab therapy).

In a murine knockout model, Brandle and associates²³ reported the role of B cells in allograft rejection and concluded that B cells may play an increasing role in graft acute rejection. However, the role of speciﬁc treatment against B cells remains controversial. Pretransplant B-cell crossmatch-positive results in patients with negative T-cell crossmatch have been associated with poor graft survival.^24,25 The role of characterizing the graft inﬁltrates and modifying the course of acute rejection in recipients with CD20-positive cells has been evaluated.¹³ Rituximab has been shown to be useful in transplant recipients with posttransplant lymphoproliferative disease, recurrent glomerulonephritis, and acute rejection refractory to standard therapies.^26,27However, rituximab has been successful in only 75% of recipients with positive DSAs and C4d-positive humoral rejection, raising the question of effective use of rituximab without CD20 phenotyping of the graft cellular infiltrate.¹²

The presence of CD20-positive lymphocyte clusters in the allograft interstitium represents a severe subset of cellular rejection.⁹ The inﬁltrating CD20-positive B cells may function as efﬁcient antigen-presenting cells for indirect allorecognition, forming intrarenal ectopic lymphoid structures and leading to development of chronic allograft dysfunction.^11,15 In addition, these active B cells interact with alloreactive CD20-positive T cells and macrophages, increasing their responses¹¹ as may have occurred in our case. To date, there are no established guidelines for the ideal management of acute rejection with CD20-positive lymphocytes in the graft, and these cells have not been considered a target in the treatment of rejection. Furthermore, these may have been inaccurately considered as B cells.^11,13 Staining for CD20 in the tissue inﬁltrate has been proposed as a rapid clinical test that permits identiﬁcation of a high-risk group of patients with acute rejection who require early specific treatment.⁹

Although reported experience with rituximab in acute cellular rejection with clusters of CD20-positive cells is scarce, a complete disappearance of B cells has been observed in the inﬁltrate, which did not occur with any of the other antirejection treatments.^11,13,15 The effects of depletion are associated with lowered incidence of chronic graft rejection and improved long-term graft survival.²⁸ If rituximab depletes only B cells, thereby inhibiting their differentiation into antibody-producing plasma cells, this action would need at least several weeks to become effective; however, rituximab has been shown to improve renal allograft rejection within a few days. One possible explanation for the rapid effects of rituximab might be the depletion of locally active intrarenal B cells and a therapeutic interference with T-cell and dendritic cell interactions.¹¹

In conclusion, our report highlights the importance of cellular characterization of graft rejection in refractory cases of acute rejection and the role of optimizing therapy with monoclonal antibody against CD20-positive lymphocytes.

Volume: 17 Issue: 6 December 2019

FULL TEXT