Objectives: Data on the management of chronic antibody-mediated rejection after kidney trans-plantation are limited. We aimed to assess the impact of treatment of biopsy-proven chronic active antibody-mediated rejection with combined plasma exchange, intravenous immunoglobulin, and rituximab treatment versus intravenous immunoglobulin alone or conser-vative management on the evolution of renal function in renal transplant recipients.
Materials and Methods: In this retrospective study, we compared patients diagnosed with chronic active antibody-mediated rejection who were treated with standard of care steroids, intravenous immuno-globulin, plasma exchange, and rituximab (n = 40) at our center versus those who received intravenous immunoglobulin only or just intensified maintenance immunosuppression (n = 28). All patients were followed for 12 months clinically and by laboratory tests for graft and patient outcomes.
Results: The two groups were matched regarding mean recipient age (41.9 ± 15.4 vs 37.8 ± 15.5 y in patients with conservative versus combined treatment), recipient sex, mean body weight, and the cause of end-stage kidney disease. Most patients and their donors were males. Glomerulonephritis represented the most common cause of end-stage kidney disease in both groups followed by diabetic nephropathy. The type of induction and pretransplant comorbidities were not different between groups (P > .05) except for the significantly higher number of chronic hepatitis C infections in patients who received conservative treatment (P = .007). Mean serum creatinine values before and after treatment of chronic active antibody-mediated rejection were comparable between groups (P > .05). Active treatment with heavier immunosup-pression (rituximab and plasma exchange) was associated with posttreatment viral (cytomegalovirus and BK virus) and bacterial infections that necessitated more hospitalization (P > .05). However, graft and patient outcomes were significantly better in the active treatment group than in patients with conservative treatment (P = .002 and .028, respectively).
Conclusions: Combined treatment of chronic active antibody-mediated rejection with plasma exchange, intravenous immunoglobulin, and rituximab can significantly improve outcomes after renal transplant.
Key words : End-stage kidney disease, Kidney transplant, Rituximab
Despite chronic active antibody-mediated rejection (ABMR) being one of the main causes for late graft dysfunction,1-3 there have been no approved drugs for its prevention and/or treatment.4 It is characterized by a silent clinical evolution that can delay its diagnosis.5,6 Since its description in 2005, its diagnostic criteria have been modified.7 The diagnosis of chronic active ABMR requires C4d deposition in peritubular capillaries in addition to suggestive histopathologic lesions and the presence of donor-specific antibodies (DSAs). The most characteristic morphologic lesion of chronic active ABMR is transplant glomerulopathy, although it is not specific.8-11
Some retrospective studies have shown that a combined treatment, which may include antibody removal, intravenous immunoglobulin (IVIG), rituximab, eculizumab, or bortezomib, can result in improved renal function in some patients.12 However, reversal of acute renal functional deterioration is limited in long-term follow-up.13
The efficacy of different strategies to treat chronic ABMR is unclear. Some small, uncontrolled studies have suggested a beneficial effect of combined treatment based on IVIG and rituximab.14,15 However, published data on the efficacy and safety of this treatment are scarce. In this study, our aim was to assess the effects of treatment of biopsy-proven chronic active ABMR with combined plasma exchange, IVIG, and rituximab compared with IVIG alone or conservative management on the evolution of renal function in renal transplant recipients.
Materials and Methods
In this retrospective study, we evaluated 65 renal transplant recipients who developed biopsy-proven chronic active ABMR between July 2004 and May 2010 according to the Banff 2005 and 2007 classification.1,7 For detection of C4d in peritubular capillaries, we used a polyclonal anti-C4d antibody (C4dpAb; Biomedica, Vienna, Austria) as described elsewhere in detail.16 Chronic active ABMR was defined as the presence of chronic transplant glomerulopathy (cg score > 0) either with or without C4d deposition in peritubular capillaries and the presence of anti-HLA DSAs determined with immunology tests. All patients were followed up in the Hamed Al-Essa Organ Transplant Center of Kuwait. All patients had been subjected to graft biopsy and C4d staining.
Patients were subcategorized into 2 groups: those who did not receive therapy or received only IVIG represented group 1 (n = 28 patients), and those who received combined rituximab, plasma exchange, and IVIG represented group 2 (n = 37). All patients were older than 18 years, and all provided informed written consent to receive chronic active ABMR treatment. Inclusion required a negative pregnancy test in women of childbearing age. Exclusion criteria included active neoplasia or history of neoplasia during the last 5 years, except for nonmelanoma skin cancer, active bacterial, viral, or fungal infections, and history of hypersensitivity reaction to any of the treatments.
The initial immunosuppressive therapy posttransplant consisted of a triple regimen of cyclosporine micro-emulsion in 39 patients, tacrolimus in 22 patients, in conjunction with mycophenolate mofetil, and prednisolone. Calcineurin inhibitor doses were gradually decreased until the end of year 1, guided by 12-hour trough level. All patients were given cytomegalovirus (CMV) and Pneumocystis jiroveci pneumonia prophylaxis for 6 months in a dose accor-ding to estimated glomerular filtration rate (eGFR).
After diagnosis of chronic active ABMR, patients in group 2 received 1 volume of plasma exchange every other day for 5 sessions using 5% human albumin or fresh frozen plasma as replacement based on daily coagulation profile of the patient. In addition, high-dose IVIG (2 g/kg to maximum of 120 g) was given on 5 consecutive daily doses after plasma exchange. Finally, a single dose of rituximab was added at the end of the last dose of IVIG with dose of 375 mg/m2 adjusted based on the patient’s body surface area. Complete blood counts, blood culture and sensitivity, urine culture and sensitivity, chest radiography, and electrocardiogram were performed before rituximab was given. Group 2 patients also received preme-dication in the form of injection of methyl-prednisolone (250 mg in 100 mL normal saline) over 30 minutes, intravenous injection of hydrocortisone (100 mg), chlorpheniramine maleate (2 mg tablet), and paracetamol (1 g tablet). Infusion was given slowly as follows: 50 mL/hour during the first hour and 75 mL/hour over the next 30 minutes, which was then increased by 25 mL every 30 minutes until the end of infusion.
Patients in group 1 received only IVIG or were maintained on a more potent maintenance immuno-suppression regimen.
Response to antibody-mediated rejection therapy was defined as reduced eGFR by at least 30% in the period 6 months after initiation of therapy compared with the period 6 months before intervention.
All patients were evaluated monthly for kidney graft function using eGFR according to the Modification of Diet in Renal Disease equation at baseline and 1 year after treatment. Patients who had graft loss 12 months after treatment of chronic active ABMR were assigned a 12-month eGFR of 0 mL/min/1.73 m2. Secondary outcomes were intergroup and intragroup differences between baseline and 1 year proteinuria.
Safety was evaluated according to the information of new or worsening adverse events as determined by investigators, with special stress on respiratory, urinary infections, and opportunistic viral infections, including CMV and BK viremia. Death with a functioning graft and graft loss were recorded. All patients were screened at regular intervals for BK viremia and/or viruria according to international guidelines.15 In case of positive results, the kidney allograft biopsy was investigated for BK neph-ropathy by SV40 staining. All patients tested negative for BK viremia and/or viruria at the time of index biopsy.
A descriptive analysis was carried out for all variables at baseline to check homogeneity between groups. Nonnumerical variables were compared by chi-square test or the Fisher exact test. Quantitative variables were compared by t test. P < .05 was considered significant. Statistical analyses were performed with SPSS software (SPSS: An IBM Company, version 22, IBM Corporation, Armonk, NY, USA).
Renal transplant recipients with biopsy-confirmed chronic active ABMR were categorized into 2 groups according to the management protocol given. Group 1 (28 patients) did not receive active management (6 patients) or only received IVIG (22 patients). Group 2 (37 patients) received combined active treatment in the form of rituximab and plasma exchange in addition to IVIG.
The 2 groups were matched regarding mean recipient age (41.9 ± 15.4 vs 37.8 ± 15.5 years in group 1 vs 2) and mean body weight. Most patients in both groups were matched with regard to recipient sex (with most being men in both groups). Moreover, most kidney donors were also males in both groups (Table 1; P > .05). Glomerulonephritis represented the most common cause of end-stage kidney disease (ESKD) in both groups (39.3% vs 37.8%) followed by diabetic nephropathy; both groups were comparable regarding causes of ESKD.
The type of induction and pretransplant comor-bidities were comparable between groups (P > .05) except for a significantly higher number of chronic hepatitis C infections in group 1 (Table 1; P = .007).
The number of patients with posttransplant diabetes was similar between groups until the time of chronic chronic active ABMR development (P > .05). We found that baseline levels of mean serum creatinine and levels at 6 months and 1 year posttransplant were comparable between groups (P > .05). Mean serum creatinine values before and after treatment of chronic active ABMR were also comparable between groups (Table 2; P > .05), with similar findings regarding their quantitative proteinuria (around 1.2 ± 0.5 g /day).
We observed that there was a trend toward a higher incidence of T-cell-mediated rejection in group 2 (14.3% vs 7.1% in group 1) and lower incidence of biopsy-proven moderate and severe interstitial fibrosis and tubular atrophy in group 2; however, this finding was not significant (Table 3; P > .05). As shown in Table 3, C4d staining was significantly higher in biopsy-proven chronic active ABMR of group 2 (60.7% in group 1 vs 78.4% in group 2; P = .04).
We did not find any significant differences between groups regarding DSA levels at different follow-up intervals after chronic active ABMR treatment (Table 4; P > .05). Regarding response to chronic active ABMR treatment, we found that most patients did not respond to therapy (65.5% in group 1 vs 70.3% in group 2) and that graft function, as represented by eGFR, remained nearly unchanged versus that shown before treatment. The number of patients who partially responded to treatment (but their graft function did not return to baseline values) were comparable in both groups (30.8% in group 1 vs 27% in group 2; Table 5; P = .9).
Active treatment with heavier immunosup-pression (rituximab and plasma exchange) was associated with posttreatment viral (CMV and BK virus) and bacterial infections that necessitated hospitalization versus that shown in group 1 (Table 5; P > .05). However, both graft (P = .002) and patient (P = .028) outcomes were significantly better in group 2 (active treatment group) than in group 1 (Table 5).
Published data have shown that the number of trials done to treat chronic active ABMR is so far insufficient, especially because it is one of the main reasons for late graft dysfunction. Moreover, many studies have rather small sample sizes, with longer than expected recruitment time. In fact, all trials were investigator determined and faced budget limitations due to the high cost of treatment. These have been major obstacles for the development of efficient treatment for chronic active ABMR.
In this retrospective study, we evaluated the impact of treatment of chronic active ABMR with combined plasma exchange, IVIG, and rituximab (group 2) compared with treatment with only IVIG or just intensification of maintenance immuno-suppression (group 1). The 2 study groups were matched regarding their demographic data (P > .05), with glomerulonephritis and diabetic nephropathy as the most frequent causes of ESKD in both groups. We observed that chronic hepatitis C virus infection was significantly more prevalent in group 1 (P < .05) but did not significantly affect the induction or maintenance immunosuppression (P > .05).
We observed that graft function as represented by eGFR before treatment of chronic active ABMR was comparable in both groups (P > .05), ranging from 39 to 40 mL/min, which was in accordance with results from another study by Stolyarevich and associates17 who reported eGFR around 42 mL/min in their cohort.
At 1 year after treatment of chronic active ABMR, we observed better graft outcomes in group 2 (P = .002), although eGFR was comparable in both groups (P > .05; Table 2). This finding matched that shown by Moreso and associates18 who reported, with caution because of small sample size, that the eGFR decline during the first year was not different between their treatment and placebo groups. This finding suggested that the combined use of IVIG and rituximab did not stabilize renal function in patients with chronic active ABMR who had transplant glo-merulopathy. Choi and associates19 suggested that IVIG and rituximab did not significantly modify the natural history of chronic active ABMR with transplant glomerulopathy. In addition, the absence of any effect on circulating DSAs suggests that this treatment may be also not efficient in patients with chronic active ABMR diagnosed at earlier stages. Fehr and associates15 and Billing and associates14 reported findings similar to ours regarding successful treatment of chronic activeABMR with IVIG and rituximab.
Billing and associates20 also reported significant improvements in eGFR decline 6 months after treatment, especially in patients without transplant glomerulopathy. Thus, we cannot attribute the lack of response among group 1 patients to be due to the selection of a population with too advanced histologic damage, as the groups were comparable regarding their biopsy findings.
In other studies, investigators evaluated different strategies combining steroid boluses, IVIG, rituximab, and thymoglobulin for the treatment of chronic active ABMR in noncontrolled cohorts,14,15,21 which showed some benefits. In our study and after 1-year follow-up, we found that most patients in both groups did not respond to therapy (65.5% in group 1 and 70.3% in group 2; P > .05) and that graft function as represented by eGFR remained nearly unchanged from that shown before treatment. However, the number of patients with functioning grafts was significantly higher in group 2 (P = .002).
In a recent systematic review that evaluated the utility of rituximab with or without IVIG in 6 retrospective controlled cohort studies, this treatment did not appear to reliably improve outcomes in chronic active ABMR.22 High-dose IVIG is considered an immunomodulatory agent that may inhibit T-cell proliferation, cytokine synthesis, complement activation, and anti-idiotypic blockade of alloantibodies.23 Jordan and associates concluded that IVIG was better than placebo in reducing anti-HLA antibody levels and improving transplant outcomes in highly sensitized patients with ESKD.24
Lefaucheur and associates25 compared the outcomes of a plasmapheresis (plasma exchange), IVIG, and rituximab-based protocol versus IVIG alone. The researchers concluded that high-dose IVIG was inferior to combination therapy. Garces and associates26 reported that the efficacy of IVIG as a monotherapy for AMR is likely limited, with better allograft outcomes in patients who received a combination therapy of plasma exchange and rituximab. These reports support our study, which revealed better graft and patient outcomes with combined treatment rather than IVIG as monotherapy.
The efficacy of rituximab in patients with chronic active ABMR remains controversial.27 Two clinical trials that assessed rituximab in patients with chronic ABMR had been recorded in the ClinicalTrials.gov database. One study, which had randomized patients with de novo DSA (ClinicalTrials.gov NCT00307125), was terminated at 9 years due to insufficient recruitment. The RituxiCAN-C4 study, which started in 2007, has not reported results (ClinicalTrials.gov NCT00476164). On the other hand, the RITUX ERAH study28 failed to show any benefit of rituximab added to plasma exchange, IVIG, and corticosteroids for the treatment of acute ABMR. However, the interpretation of this study should be cautious since the study drug was administered to the control group as rescue therapy in a significant proportion of patients (8 of 19 patients).
In our study, benefits were shown in the combined therapy group at 1 year despite its use among patients with chronic active ABMR. Some benefits among the combined treatment group may have been due to a possible trend of relatively less chronic changes and more T-cell-mediated elements detected in the biopsies (despite both groups being matched), which have a better prognosis in the long term.
Regarding the importance of C4d staining, Haririan and associates29 did not show significant differences in the course of rejection depending on type of C4d positivity; others have noticed more favorable prognosis in patients with focal C4d expression. In our study, we observed better graft outcomes in group 2, possibly due to a significantly less number of patients with positive C4d staining.
Regarding the safe use of rituximab among our patients, we found no significant differences between groups regarding posttreatment viral (CMV and BK virus) and bacterial infections necessitating hos-pitalization (Table 3; P > .05). This observation was matched with that reported by Parajuli and associates30 who showed no difference in the rate of BK viremia or other infections (including pyelo-nephritis, cellulitis, and norovirus) between the rituximab group and the standard-of-care group (P = .31 and P = .41, respectively). However, patients who received standard of care (IVIG and steroid pulse) had a higher incidence of CMV infection than the rituximab group (P = .04). This last observation was not matched with our finding, possibly because of our routine antiviral prophylaxis.
Other recent treatments for chronic active ABMR have also been tested. Eculizumab has been evaluated in patients with late-onset chronic active ABMR.23 In one study, patients who were randomized to receive eculizumab for 6 months and followed for an additional 6 months showed a trend for renal function stabilization (P = .09).31 Bortezomib has also been tested in chronic active ABMR; these studies include the BORTEJECT Study32 and the Tribute study (ClinicalTrials.gov NCT02201576), which is so far not published.
The anti-interleukin 6 receptor monoclonal antibody tocilizumab has also been evaluated in an open study as rescue therapy.19 In 36 patients with chronic active ABMR who did not respond to standard of care (plasma exchange, IVIG, and rituximab), patients who received tocilizumab showed reduced mean fluorescence intensity of the immuno-dominant DSA, stabilization of renal function, and amelioration of microcirculatory lesions.19 There is also another ongoing trial with the anti-interleukin 6 receptor monoclonal antibody clazakizumab for chronic active ABMR in kidney transplant recipients (ClinicalTrials.gov NCT03380377).
Combined treatment of chronic active ABMR with plasma exchange, IVIG, and rituximab significantly improved outcomes after renal transplant.
Volume : 17
Issue : 1
Pages : 113 - 119
DOI : 10.6002/ect.MESOT2018.O58
From the 1Kuwait Ministry of Health, Hamed Al-Essa Organ Transplant Center,
Sabah area, Kuwait; and the 2Urology and Nephrology Center, Mansoura University,
Acknowledgements: The authors want to acknowledge sisters working in the outpatient clinic, especially Sherin Farag, Bancy Baby, and Sijy Paul, of Hamed Al-Essa Organ Transplant Center for their intensive collaborative efforts during follow-up and collection of patient data. The authors had no sources of funding for this study and have no conflicts of interest to declare.
Corresponding author: Osama Ashry Ahmed Gheith, Hamed Al-Essa Organ Transplant Center, Sabah area, Kuwait
Phone: +96 566641967
Table 1. Patient Demographics Before Treatment of Chronic Active Antibody-Mediated Rejection
Table 2. Follow-Up of Patients Before Treatment of Chronic Active Antibody- Mediated Rejection
Table 3. Pathologic Parameters of Biopsies Showing Chronic Active Antibody-Mediated Rejection in Both Groups
Table 4. Follow-Up Parameters of Study Patients Before Treatment of Chronic Active Antibody-Mediated Rejection
Table 5. Posttransplant Outcomes of Study Patients