Objectives: There are no comparable trials concerning the use of rituximab among renal transplant recipients with acute antibody-mediated rejection. Here, we compared early and late acute antibody-mediated rejection in renal transplant recipients in terms of response to rituximab therapy.
Materials and Methods: Of 1230 kidney transplants performed at Hamed Al-Essa Organ Transplant Center (Kuwait) over the past 10 years, 103 recipients developed acute antibody-mediated rejections and were subcategorized into 4 groups according to the onset of rejection and rituximab treatment. All patients received the standard treatment for acute antibody-mediated rejection according to our protocol (plasma exchange and intravenous immunoglobulin). We added rituximab to the treatment regimen in 2 groups of patients: 27 patients with early rejection (group 1) and 38 patients with late rejection (group 2). Groups 3 and 4 represented nonrituximab groups, with 20 patients with early (group 3) and 18 patients with late rejection (group 4). We compared the 4 groups regarding graft and patient outcomes.
Results: All patients were comparable regarding patient age, sex, pretransplant type of dialysis, viral profile, type of induction, donor criteria, and pretransplant comorbidities. We observed that delayed and slow graft function were significantly higher in groups 1 and 3 (P = .016); however, we found no significant differences in the 4 groups regarding new-onset diabetes after transplant, BK viral infection, and malignancy. Graft outcomes were significantly better in groups 1 and 2 than in groups 3 and 4 (P = .028). However, patient outcomes were comparable in the 4 groups (P > .05).
Conclusions: Early acute antibody-mediated rejection in renal transplant recipients had significantly better outcomes when rituximab was added to the standard treatment regimen.
Key words : Acute ABMR, Rejection, Rituximab, Timing
In the field of renal transplant, a major complication associated with anti-HLA donor-specific alloantibody (DSA) is hyperacute rejection. With new sensitive diagnostic techniques for DSA and increased understanding of the histologic changes associated with DSA, it has become apparent that acute antibody-mediated rejection (aABMR) has emerged as a major cause of graft loss in the weeks and months after transplant.1
Over the past 10 years, the diagnostic accuracy for aABMR in kidney transplant patients has improved significantly. The phenotypes of early and late aABMR may differ.2 Close monitoring of renal graft function and early graft biopsy for aABMR detection may result in satisfactory short-term allograft outcomes, especially in highly sensitized patients as repeat transplant recipients.3
Basic immunology of antibody production and aABMR is vital in understanding the various treatment approaches for aABMR and their limitations and for therapy development.1 Plasmapheresis is the physical removal of anti-HLA antibodies, and has remained the main treatment of antibody-mediated rejection for the past 2 decades. Its effect tends to be temporary until the source of antibody is controlled, with marked variability in its efficacy (from 0% to 93%).4 Randomized controlled trials have not confirmed a benefit from plasmapheresis alone,5 but this therapy remains a first-line treatment for aABMR.
Intravenous immunoglobulin (IVIg) is polyvalent IgG antibody that comes from a pool of over 1000 blood donors. This agent is approved for aABMR treatment and for desensitization of transplant recipients. The exact mechanism of the immunomodulatory action of IVIg is largely unknown. Nevertheless, it is thought to saturate the Fc receptors on macrophages, suppress the production of inflammatory mediators, modulate complements, interfere with DSA binding and/or activity, and suppress idiotypic antibodies.6 Despite several available reports on use of IVIg as treatment for aABMR and as desensitization for highly sensitized transplant candidates,7,8 no randomized controlled trials for use of IVIg in aABMR have been published. Currently, IVIg is often used as adjunctive therapy for aABMR, combined with plasmapheresis and newer biologic agents.9
Although invasive, splenectomy is a potential therapy for refractory aABMR, showing rapid recovery of renal graft function within 2 weeks.10,11 Normally, the spleen is rich in mature B cells and not plasma cells; however, spleens removed at times of aABMR exhibit a distinct increase in CD138+ plasma cells compared with spleens removed for other reasons.12 Its role in chronic antibody-mediated rejection is unknown.
Rituximab, a monoclonal anti-CD20 antibody that binds to the Fc portion, leads to antibody-dependent cellular cytotoxicity, complement-dependent cytotoxicity, and a rapid 70% to 80% decline in the B-cell population. This agent may down-regulate CD40, thus inhibiting the interaction between T and B cells, and may also inhibit cytokine-dependent B-cell proliferation and differentiation.13
In a nonrandomized retrospective study, patients treated with rituximab in addition to plasmapheresis and IVIg had improved graft and renal function compared with those treated with plasmapheresis and IVIg alone. However, the response to rituximab appears to be variable. Rituximab is likely ineffective in circumstances of high antibody burden because the established short- or long-living plasma cells producing the antibody lack CD20. The variable expression of CD20 on B cells may also play a role in its variable efficacy.14
In this study, we aimed to compare early and late aABMR in terms of response to rituximab therapy.
Materials and Methods
Of 1230 renal transplant recipients who underwent renal transplant between 2000 and 2012 at Hamed Al-Essa Organ Transplant Center (Kuwait), 103 patients with biopsy-proven aABMR were included in our study. They were classified into 4 groups according to the onset of rejection episode and type of treatment. All patients received the standard treatment for aABMR according to our protocol (plasmapheresis and IVIg). Patients with early rejection (< 3 mo) who received rituximab represented group 1 (n = 27); those with late rejection and rituximab treatment represented group 2 (n = 38). In addition, those who did not receive rituximab for early rejection represented group 3 (n = 20), and those with late rejection without rituximab represented group 4 (n = 18).
For our patient groups, the following data were retrospectively analyzed: age at transplant, sex, underlying renal disease, prior renal replacement therapy, organ donor source, and HLA mismatch. To assess renal transplant outcomes, patient and graft survival rates, allograft function, and infection episodes necessitating hospitalization (especially viral infections, such as Cytomegalovirus [CMV] and polyoma virus) were analyzed. The underlying renal diseases of the examined patients were classified into groups: diabetic nephropathy, hypertension, glomerulonephritis, obstructive renal disorders, and idiopathic. Pretransplant complement-dependent cytotoxicity and flow cytometry cross-matches were negative.
All patients received triple immunosuppressive regimens consisting of a calcineurin inhibitor, mycophenolate mofetil, and corticosteroids. All induction therapy regimens were based on our protocol guidelines and transplant risk factors. Patients at high risk for acute rejection, which included retransplant patients, those with panel reactive antibodies > 20%, or those who received deceased donations, were given rabbit antithymocyte globulin (Thymoglobulin, Genzyme Corp., Cambridge, MA, USA) at a dose of 1.0 mg/kg/d (total of 5 doses).
Other patients received interleukin 2 receptor antagonist induction using basiliximab (Simulect, Novartis Pharmaceuticals, New York, NY, USA) at 10 mg/m2 intravenously as a bolus within 2 hours of engraftment on day 0, with a second dose of 10 mg/m2 on day 4 at a dose of 1 mg/kg body weight. Patients with zero HLA mismatches received neither rabbit antithymocyte globulin nor interleukin 2 receptor antagonist induction. Corticosteroids were initiated intraoperatively as methylprednisolone at 250 to 500 mg according to body weight, then at 1 mg/kg to a maximum of 60 mg/d from day 1 posttransplant, and finally tapered to low-dose prednisone (0.1-0.5 mg/kg/d) by 3 months posttransplant.
Target 12-hour whole blood trough concentrations for tacrolimus (Prograf, Astellas Pharmaceuticals, Deerfield, IL, USA) were as follows: 10 to 15 ng/mL at weeks 1 to 6, 8 to 12 ng/mL at weeks 7 to 12, 6 to 10 ng/mL at months 3 to 12, and > 5 ng/mL at > 1 year or as clinically indicated. Target 12-hour whole blood trough concentrations for cyclosporine (Neoral, Novartis Pharmaceuticals) were as follows: 200 to 275 ng/mL at weeks 1 to 6, 175 to 225 ng/mL at weeks 7 to 12, 125 to 175 ng/mL at months 3 to 12, and > 70 ng/mL at > 1 year or as clinically indicated. During thymoglobulin induction, doses of calcineurin inhibitors were minimized and then returned to full dose 2 days before discontinuation of the induction regimen. All patients received mycophenolate mofetil (CellCept, Roche Pharmaceuticals, Nutley, NJ, USA), with initial doses of 600 mg/m2 orally twice daily.
Doses were adjusted for efficacy and toxicity. Graft failure was defined as a situation where any other form of renal replacement therapy had to be started. Death with functioning graft was not considered to be graft failure. The deaths that were not primarily associated with renal transplant were censored in the Kaplan-Meier analyses. The glomerular filtration rate was computed using the Schwartz formula (glomerular filtration rate [mL/min/1.73 m2] = 0.55 × body length [cm]/serum creatinine [mg/dL]).
All CMV-positive recipients and CMV-negative recipients of kidneys from CMV-positive donors were given prophylaxis with valacyclovir for the first 3 months after transplant. All patients received prophylaxis against Pneumocystis jiroveci with sulfamethoxazole/trimethoprim for 6 months. All patients had blood polymerase chain reaction for BK virus at 3 months, at 1 year posttransplant, and every year thereafter.
Treatment of rejection episodes
All rejection episodes were biopsy proven and were treated initially using high-dose steroids until renal biopsy results confirmed aABMR.
Acute rejection was biopsy proven, and the diagnosis was made according to Banff classification 2007 and treated with high-dose corticosteroids. Borderline rejections were included in the analysis if treated as acute rejection. Patients were considered to have delayed graft function if they required dialysis within the first week posttransplant.
Data were manually collected in an Excel spreadsheet (Microsoft, Seattle, WA, USA). Statistical analyses were performed using software (SPSS version 11.0, SPSS Inc., Chicago, IL, USA).
Qualitative data were presented as numbers and percentages, whereas quantitative data were presented as means and standard deviation. The t test was used to compare means and standard deviations of the 3 groups. Categorical data were compared using the chi-squared test. Graft and patient survival were computed using the Kaplan-Meier technique. P values < .05 were considered significant.
Most of our patients were females with mean ages of 30.8 ± 13.6, 28.3 ± 13.3, 33.1 ± 17, and 36.7 ± 18 years in the 4 groups with or without rituximab treatment. Most patients received their grafts from male donors with mean ages of 36.7 ± 8, 34 ± 7.8, 33.8 ± 12.2, and 32.5 ± 10.7 years in the same groups (P > .05). We found no significant difference between the 4 groups regarding pretransplant comorbidities especially anemia; patients who received treatment for tuberculosis, hypertension, and diabetes mellitus; viral profile (especially anti-hepatitis C virus, anti-CMV, hepatitis B virus, and herpes viruses); and bone disease (Table 1; P > .05).
The original kidney disease was comparable in all groups (P = .30). Moreover, the number of cases who underwent preemptive renal transplant and number of patients who were on regular hemodialysis or peritoneal dialysis were comparable (Table 1; P = .28).
We observed that patients in the 4 groups were comparable regarding the source of kidney donors (Table 1; P > .05). Moreover, we observed that most patients received induction therapy with lymphocyte-depleting agents (antithymocyte globulin or thymoglobulin); however, the 4 groups were comparable (Table 2; P = .21). Although most patients with early ABMR were maintained on tacrolimus-based triple immunosuppression, those with late ABMR were kept on cyclosporine-based triple immunosuppression (Table 2; P = .43).
More than 50% of patients who develop early aABMR experienced slow or delayed graft function, which was significantly higher than the 2 groups with late aABMR (Table 1; P = .016).
Posttransplant complications and outcomes
Most of the acute rejection episodes developed during the first year posttransplant. Regarding nonimmunologic complications, we observed that new-onset diabetes after transplant, de novo hypertension, and infections (especially CMV and BK virus infection) were comparable in the 4 studied groups regardless of whether rituximab was included as part of treatment (Table 3; P > .05). We found no evidence of malignancy in any of the 4 groups of patients. We observed that graft outcomes in patients with early aABMR who were treated with rituximab (group 1) were significantly better than graft outcomes in the other groups (P = .024). However, patient outcomes were comparable in the 4 groups (P > .05).
Acute rejection episodes occurring late after renal transplant have long been known to be associated with poor renal allograft survival.15 These earlier studies, however, focused only on the timing of acute rejection and not on immunologic mechanisms (T cell or antibody mediated). The recent application of C4d staining and solid-phase assays for anti-HLA antibody detection has greatly enhanced aABMR detection and classification of acute rejection episodes. This has led to recent observations that aABMR significantly reduces renal allograft survival compared with T-cell-mediated rejection.16,17
Over the past decade, the diagnostic precision for aABMR in kidney transplant recipients has improved significantly. The phenotypes of early and late aABMR may differ. During the past 15 years, major advances in the understanding of the effects of antidonor antibodies on renal allografts at various stages after transplant have occurred. These advances have been due in large part to pathologic examination of both early and late renal allograft biopsies, including both routine histologic evaluation and immunohistology to detect complement split products.18
In our study, we focused on antibody-mediated rejection, its timing, and its response to therapy, especially rituximab. The 4 groups of patients included in the study were matched regarding demographic data, pretransplant comorbidities, original kidney disease, type of dialysis modality, source of kidney donors, and type of immunosuppression (both induction and maintenance). Most of the acute rejection episodes developed during the first year after transplant. We observed that more than 50% of patients with early aABMR experienced slow or delayed graft function, which was significantly higher compared with the 2 groups with late aABMR. This finding was matched with some reports denoting that slow and delayed graft function are risk factors for T-cell and antibody-mediated rejection19 and the combination of both can affect patient survival.20,21
Most of our patients were young females, and this was explained by Dörje and associates2 who found that most of their patients with rejection were young and nonadherent with suboptimal immunosuppression.
Regarding nonimmunologic complications, we observed that new-onset diabetes after transplant, de novo hypertension, and infections (especially CMV and BK virus infections) all were comparable in the studied groups regardless of whether rituximab was included in their treatment protocol (Table 3; P > .05).
The risk of serious infection was reported to be more frequent with standard rituximab (375 mg/m2) than with reduced-dose rituximab (200 mg/m2).22 In addition, Gulleroglu and associates23 reported that the combined use of rituximab with additional treatments such as antithymocyte globulin, IVIg, and repeated plasma exchange may be associated with high risk of infections. They recommended closely monitoring these patients, especially those who receive T-cell–depleting agents.
Because of close monitoring of our patients with CD count and chemoprophylaxis policy adopted against infections, the risks of infection were comparable among our studied patients. Moreover, we did not report any evidence of malignancy among patients in the 4 groups.
Tatar and associates24 found that late antibody-mediated rejection can occur soon after the modification of immunosuppressive drug dosages and may be responsible for graft dysfunction or loss.
Before aABMR, we did not convert our patients from their primary immunosuppressive regimen. In addition, we observed that graft outcomes in patients with early aABMR and treated with rituximab (group 1) were significantly better than shown in other groups (P = .024). Our finding was matched with that reported by Dörje and associates2 who reported inferior graft survival among cases with late aABMR compared with early aABMR.
However, this observation was not matched with the findings by Sautenet and
associates25 after 1-year follow-up, as they observed no additional effects of
rituximab in patients receiving plasma exchange, IVIg, and steroid pulse for
aABMR. This could
be explained by the fact that their study was underpowered and important
differences between groups may have been missed. Moreover, they recommended
complementary trials with long-term follow-up.
Inferior graft outcomes despite rituximab therapy in late rejection episodes can be explained by an accelerated B-cell response that overwhelms the capacity of plasmapheresis and IVIg to clear soluble DSA, with anti-CD20 (rituximab) being ineffective in depleting plasma cells. Fortunately, patient outcome was comparable in the 4 groups (P > .05).
Acute antibody-mediated rejection was associated with better graft outcome when rituximab was included in the treatment protocol, especially with early rejection episodes.
Volume : 15
Issue : 1
Pages : 150 - 155
DOI : 10.6002/ect.mesot2016.P32
From the 1Urology and Nephrology Center, Mansoura University, Egypt; the
Al-Essa Organ Transplant Center, Kuwait; and the 3Pathology Department, Faculty
of Medicine , Kuwait University
Acknowledgements: The authors declare that they have no sources of funding for this study, and they have no conflicts of interest to declare.
Corresponding author: Osama Gheith, Nephrology Department, Hamed Al-Essa Organ Transplant Center, Kuwait
Phone: +965 66641967
Table 1. Demographic Data of Study Patients
Table 2. Types of Immunosuppression and Posttransplant Complications Among Study Groups
Table 3. Patient and Graft Outcomes Among Study Groups