ABO-incompatible (ABO-I) kidney transplantation (KTx) is currently performed worldwide to expand living donor sources. In the early period of its development, KTx across the blood type barrier indicated that hyperacute rejection is caused by naturally occurring antibodies. Recently, desensitization treatment comprising the anti-CD20 antibody rituximab and pretransplant plasmapheresis has reduced the risk of hyperacute rejection, allowing us to avoid splenectomy and reducing the incidence of infections. Thus, recent studies have revealed comparable graft and patient survival.1-3 However, several studies have suggested that ABO-I KTx is still associated with a higher incidence of infectious complications.1,4,5 Infections in patients who have undergone ABO-I KTx might be associated with current intensive immunosuppressive treatments in combination with desensitization using rituximab (375 mg/m2).4,5 A recent Japanese study suggested that the effect of low-dose rituximab (200 mg/body) on B-cell depletion and suppression of anti-A/B antibody titers was similar to that of the conventional dose.6 However, the safety and efficacy of regimens with low-dose rituximab has not been clearly established.
Several protocol biopsy (PB)–based studies have compared the histopathological findings between ABO-I and ABO-compatible (ABO-C) KTx.4,7–10 However, their study populations were small, desensitization was different, and the results varied. We therefore focused on our desensitization protocol with low-dose rituximab and plasmapheresis and performed a large-scale retrospective study of 327 patients who underwent living-donor KTx (including 101 who underwent ABO-I KTx) to investigate the allograft pathology, incidence of biopsy-proven acute rejection (BPAR), and incidence of infections.
MATERIALS AND METHODS
From July 2008 through May 2014, 436 patients underwent living-donor KTx in Kyushu University Hospital (ABO-C KTx, n = 298; ABO-I KTx, n = 138). We excluded 109 patients (ABO-C KTx, n = 72; ABO-I KTx, n = 37) with preformed donor-specific antibodies (DSA) determined by a FlowPRA Single Antigen test and/or positive flow cytometry crossmatch test to exclude the effect of desensitization treatment in the ABO-C KTx group and antibody-mediated rejection (AMR) caused by preformed anti-HLA DSA in the ABO-I KTx group. Finally, we investigated 226 patients who underwent ABO-C KTx and 101 who underwent ABO-I KTx in this retrospective study. All PBs were obtained according to our clinical follow-up protocol, and no extra biopsy specimens or urine/blood samples were obtained for the purpose of the study. Written informed consent was obtained from eligible patients, and this study was approved by the Institutional Review Board at Kyushu University Hospital (protocol 24-54).
All patients in this study received induction therapy with basiliximab (20 mg/body on days 0 and 4) and a triple-drug regimen comprising tacrolimus (Tac), mycophenolate mofetil (MMF), hand methylprednisolone (mPSL). Patients in the ABO-C KTx group began treatment with Tac, MMF, and mPSL 7 days before KTx. The dose of tacrolimus was adjusted to achieve a trough level of 5 to 8 ng/mL (chemiluminescent enzyme immunoassay) before KTx and during the first 2 weeks after KTx, 6 to 7 ng/mL during the next 2 weeks, 5 to 6 ng/mL during the next month, and 4 to 5 ng/mL thereafter. We administered MMF at a fixed dosage of 2000 mg/d from 7 days before KTx to 2 months after KTx and maintained the dosage at 1000 mg/d thereafter. Oral mPSL was administered at 8 mg/d before KTx; intravenous mPSL at 250 mg on the day of KTx; and oral mPSL at 24 mg/d for 7 days, 16 mg for the next 7 days, 8 mg/d for the next 2 weeks, and 4 mg/d for maintenance.
Patients in the ABO-I KTx group began treatment with the triple-drug regimen on preoperative day 14, and the drugs were tapered according to the same schedule as in the ABO-C KTx group. In addition, we administered 200 mg/body of rituximab on preoperative day 7 and performed 3 or 4 sessions of plasma exchange before transplantation to remove anti-A/B antibodies until the patients' anti-A IgG/IgM titers and/or anti-B IgG/IgM titers decreased to a level of 1:32 or less. Even in the patients with a low anti-A/B IgG titer, we administered the same dose of rituximab and performed at least 1 session of plasma exchange. Intravenous immunoglobulin was not administered at any time during the study, and splenectomy was not performed at the time of transplantation. If patients developed refractory cytomegalovirus (CMV) infection, sustained polyomavirus BK viremia of >200 viral copies/mL in plasma, or polyomavirus BK nephropathy (BKVN), we switched Tac to cyclosporine or MMF to everolimus. We also switched MMF to everolimus after the 3-month PB in several patients with a history of cancer. Eventually, 7 patients were treated with cyclosporine and 10 were treated with everolimus at 12 months (Table 1).
PB and Pathological Interpretations
Since August 2008, our policy has been to perform PB at 3 and 12 months posttransplantation.11,12 Our preliminary data revealed the usefulness of this protocol in detecting subclinical acute rejection at 3 months under current immunosuppression.11 Allograft biopsy was performed under ultrasound guidance using a Bard Magnum device (Bard Biopsy Systems, Tempe, AZ) and 18-gauge needles. For light microscopy, serial tissue sections were stained with hematoxylin and eosin, periodic acid-Schiff, periodic acid methenamine silver, and Masson trichrome stains. For the immunofluorescence study, we examined the biopsy specimens for immunoglobulin IgG, IgA, IgM, complement (C)3, C1q, fibrinogen, kappa/lambda light chains, and C4d. All biopsy specimens were scored according to the Banff 2009 classification.13 Patients with acute rejection were classified into those with borderline changes, acute T cell–mediated rejection (TCMR) (Banff grade IA or higher), and/or acute AMR. Subclinical acute rejection was defined as rejection diagnosed by PB without a greater than 15% increase in the serum creatinine concentration from baseline levels (determined by the average serum creatinine concentration 3 months before PB) and no previous rejection episodes within 1 month.12
If the patients showed an unexplained rise in the serum creatinine concentration, we performed “indication biopsies” with the same procedure. For the diagnosis of BKVN, we added SV40 large T antigen immunohistochemical staining regardless of the results of plasma polymerase chain reaction for BK viral DNA when the patients showed sustained positive urinary decoy cells before the biopsies.
Prophylaxis, Monitoring, and Diagnosis of Infectious Diseases
Our strategy of prophylaxis, monitoring, and diagnosis of infectious diseases was uniform throughout the study period. All patients received regular screening for CMV antigenemia, and all CMV donor-positive/recipient-negative patients received valganciclovir prophylaxis for 3 months. Screening for polyomavirus infection was based on regular urinary cytology testing followed by plasma polymerase chain reaction for BK viral DNA and allograft biopsy, according to the recommendation of the American Society of Transplantation Infectious Diseases Community of Practice.14 Oral Pneumocystis jirovecii prophylaxis (trimethoprim/sulfamethoxazole 80/400 mg 3 times a week) was also administered to all patients.
Follow-Up at Outpatient Clinics, Graft Function, and Patient and Graft Survival
All patients were managed at outpatient clinics of Kyushu University Hospital or Harasanshin Hospital by transplant surgeons or nephrologists. Although 11 patients were transferred to other centers during follow-up, we collected the clinical information of those patients. No patients were lost to follow-up during this study. The estimated glomerular filtration rate (eGFR) was calculated by the equation established in the Japanese guidelines: eGFR = 194 × serum creatinine−1.094 × age−0.287 × 0.739 (if female)15 for adult patients, and by the Schwarz equation16 for pediatric patients. We compared the eGFR at 3 and 12 months and graft and patient survival at the last visit between the 2 groups.
Data are expressed as mean ± standard deviation, number (%), or median with range as appropriate. JMP Pro 11.0.0 (SAS Institute, Cary, NC, USA) was used for all statistical analyses. Student t test and the Mann-Whitney U test were used to assess differences in numerical variables between 2 groups. The χ2 or Fisher exact probability tests were used for categorical data as appropriate. The cumulative incidence of BPAR, graft survival, and patient survival was assessed by Kaplan-Meier estimation followed by the log-rank test. Differences in data among 4 groups were tested by one-way analysis of variance followed by Tukey honestly significant difference test for multiple comparisons. A P value less than 0.05 was considered statistically significant.
The demographic and clinical characteristics of the studied patients are summarized in Table 1. The mean age of the recipients and the HLA-A, -B, -DR mismatch count were higher in the ABO-I KTx than ABO-C KTx group, but other clinical parameters in the donors and recipients were not significantly different between the 2 groups. All patients first underwent KTx; patients who were presensitized with a positive FlowPRA Single Antigen test and/or flow cytometry crossmatch test were excluded from this study. The median (range) of the anti-A/B IgG titer in patients who underwent ABO-I KTx was 1:64 (1:1-1:2048), and we performed 2.6 ± 1.5 sessions of plasma exchange.
Protocol biopsy at 3 months was performed in 187 (82.7%) patients in the ABO-C KTx group and 91 (90.0%) in the ABO-I KTx group. Protocol biopsy was also performed in 170 (75.2%) patients in the ABO-C KTx group and 79 (78.2%) in the ABO-I KTx group at 1 year. Acute inflammatory changes, including subclinical borderline changes, acute TCMR, and mixed acute rejection (acute TCMR + AMR), between the ABO-C KTx and ABO-I KTx groups at the 3-month and 1-year PBs are shown in Figure 1A. Subclinical acute rejection occurred in 6.9% of patients in the ABO-C KTx group and 9.9% of patients in the ABO-I KTx group at 3 months, and in 12.4% and 10.1% at 12 months, respectively, without a significant difference. At the 12-month PB, we found only 1 patient with mixed acute rejection among all patients in the ABO-I KTx group. In that patient, the presence of de novo anti-HLA DSAs was confirmed by the FlowPRA Single Antigen test. We also found subclinical BKVN in 1 patient who underwent ABO-I KTx at 3 months. The histopathological grades of subclinical TCMR at 3 and 12 months are shown in Table 2. Although no patients in the ABO-I KTx group developed vascular rejection (Banff grade IIA/IIB) at either the 3-month or 1-year PB, the difference in the distribution did not reach statistical significance. Further investigation of microvascular inflammation revealed comparable distributions of glomerulitis plus peritubular capillaritis scores (g + ptc scores) between patients in the ABO-C KTx and ABO-I KTx groups (Figure 1B). The degrees of interstitial fibrosis and tubular atrophy were not also different between the 2 groups (Figure 1C). Peritubular capillary C4d deposition was common in the ABO-I KTx group, and positive C4d staining (≥C4d1) was found in 82.0% of patients at the 3-month PB and in 77.6% at the 12-month PB (Figure 1D).
BPAR Diagnosed by PBs and Indication Biopsies
During the observation period, 17 patients (7.5%) in the ABO-C KTx group and 8 (7.9%) in the ABO-I KTx group developed acute rejection diagnosed by indication biopsy. All rejection episodes were diagnosed as acute TCMR; no patients fulfilled the criteria for mixed AMR. No significant difference was found in the pathological grades of acute TCMR on indication biopsy, but the grade tended to be mild in the ABO-I KTx group (Table 2). The cumulative incidence of BPAR diagnosed by both PBs and indication biopsies was 20.5% and 19.6%, respectively, and comparable between the 2 groups (Figure 2).
Peritubular Capillary C4d Deposition, Tubulointerstitial Inflammation, and Graft Function
We classified the 249 patients who underwent the 12-month PB into 4 subgroups according to their peritubular capillary C4d deposition and tubulointerstitial inflammation (TIN): (a) the C4d−/TIN− group: peritubular capillary C4d deposition–negative (C4d score of 0) and no evidence of rejection or nonspecific inflammation only (n=127), (b C4d−/TIN+ group: C4d score of 0 and ≥borderline changes (n=51), (c) C4d+/TIN− group: C4d score of ≥1 and no evidence of rejection or nonspecific inflammation only (n=50), and (d) C4d+/TIN+ group: C4d score of ≥1 and ≥borderline changes (n = 21). Comparison of the eGFR at the 12-month PB and the last visit among the 4 groups revealed a significantly lower eGFR in the C4d+/TIN+ group than in the C4d−/TIN− and C4d+/TIN− groups at the last visit (Figure 3).
Infectious complications during the observation period are summarized in Table 3. The incidence of CMV antigenemia tended to be high in the ABO-C KTx group; this might have occurred because young recipients who received CMV donor-positive/recipient-negative transplants were more common in the ABO-C KTx group. The incidence of CMV disease was not different. With regard to BK viral infection, 6 patients (2.7%) in the ABO-C KTx group and 3 (3.0%) in the ABO-I KTx group developed biopsy-proven BKVN (P = 1.0); the incidence of BK viremia of more than 200 copies/mL by plasma PCR was also comparable. The incidences of other infectious diseases shown in Table 3 were not different between the 2 groups.
Graft Function at 12-Month and 5-year Graft and Patient Survival Rates
There was no significant difference in kidney graft function at 3 and 12 months posttransplantation between the ABO-C KTx and ABO-I KTx groups (Table 1). The 5-year graft and patient survival rates were also not different between the 2 groups (Figures 3 and 4).
In this study, we investigated the 3- and 12-month PB findings between patients who underwent ABO-C KTx and ABO-I KTx. Our findings demonstrated no significant differences in subclinical rejection or other findings between the 2 groups. This is a larger study population than in all previous PB-based studies.4,7-10 Because the study took place over a recent 6-year period, patients were treated with a uniform initial immunosuppressive regimen and modern desensitization protocol comprising rituximab and plasmapheresis without splenectomy. The PB protocol was also fixed. We strictly excluded patients with preformed DSA from the analysis to purely observe the effect of ABO-I KTx on the allograft pathology, cumulative incidence of BPAR, and infectious complications.
Because Alexandre et al17 introduced the concept that desensitization must include both plasmapheresis and splenectomy to prevent hyperacute rejection across the blood type barrier, ABO-I KTx has been well studied. However, in the era of desensitization involving plasmapheresis and splenectomy, the incidence of AMR has remained high in patients undergoing ABO-I KTx. Gloor et al7 investigated 1-year PBs and found a higher incidence of AMR in patients who underwent ABO-I KTx than in those who underwent ABO-C KTx (73% vs 13%, respectively). Setoguchi et al8 and Ushigome et al9 also showed a higher incidence of AMR in patients who underwent ABO-I KTx than in those who underwent ABO-C KTx (27.0% vs 5.3% and 17.9% vs 223231.1%, respectively). In 2002, Sawada et al18 reported a successful case of ABO-I KTx using rituximab (375 mg/m2, 3 doses), splenectomy, and plasmapheresis. Later, in 2004, Sawada et al19 described 4 consecutive cases involving the same protocol; no patient developed AMR, but 1 showed TCMR. Tydén et al20 described 11 patients who received single-dose rituximab without splenectomy, and no patient developed AMR or TCMR. These protocols have since become common worldwide. As a result, the incidence of AMR has been obviously reduced, and more recent studies have shown no difference compared with ABO-C KTx.4,10 Anti-HLA DSA rather than antiblood type antibodies are now suggested to be more important pathogenic factors in acute AMR after ABO-I KTx.21
On the other hand, several studies have suggested no difference in the incidence of TCMR between ABO-C and ABO-I KTx.4,7-10 In the present study, the Banff grades and cumulative incidence of BPAR at the 3- and 12-month PBs were comparable between the 2 groups, but acute TCMR as determined by indication biopsy tended to be mild in the ABO-I KTx group (Table 2). Although we found no clear explanation for that trend, we speculate that some causative factors, such as older recipient age and earlier initiation of immunosuppression, were present in patients who underwent ABO-I KTx, suggesting that relatively strong immunosuppression affected the results. A recent study performed by Couzi et al22 suggested that peritubular capillary C4d deposition with TIN was associated with decreased graft function in patients who underwent ABO-I KTx. Thus, we evaluated the 12-month PB findings, classified the patients into subgroups according to C4d deposition and TIN, and compared graft function. C4d+/TIN+ patients showed lower graft function at the last visit than did C4d−/TIN− and C4d+/TIN− patients, which is a trend similar to that reported in the abovementioned study by Couzi et al22 Because we had no information on follow-up biopsies performed after 1 year in the present series, it is difficult to determine the mechanisms of the accelerated graft dysfunction induced by superimposition of TIN on C4d deposition in patients undergoing ABO-I KTx. Further investigations with long-term follow-up and repeated PBs later than 1 year are necessary.
Another focus of this study was the incidence of infectious complications including biopsy-proven BKVN, an important disease that directly injures the kidney allograft. A recent large-scale prospective study from Europe demonstrated that ABO-I KTx was associated with comparable graft and patient survival, but it had a higher incidence of death because of infection.1 Other studies also suggested increased infectious complications in patients who underwent ABO-I KTx,5,23 and ABO-I KTx was reported to be a risk factor for the development of BKVN.24,25 This is in conflict with our results. In the present study, all patients who underwent ABO-I KTx were treated with low-dose (200 mg/body) rituximab; this differs from the studies performed in Europe and the United States, in which a dose of 375 mg/m2 was commonly used. A Japanese large-scale retrospective cohort study demonstrated that ABO-I KTx performed in 2004 or earlier was associated with a higher incidence of CMV and adenovirus infections, higher mean serum creatinine levels, and lower graft survival, whereas such clinical outcomes were comparable in the more recent era with low-dose rituximab administration.3 These results and our data suggest that advances and modifications in immunosuppressive treatment, such as low-dose rituximab and low-dose Tac, as well as close monitoring of patients, could reduce the incidence of infections in patients undergoing ABO-I KTx.
There are several limitations in this study. First, this was a single-center, retrospective cohort study with a short-term observation period. Although the study population was relatively large, the clinical backgrounds were partially mismatched between the ABO-C KTx and ABO-I KTx groups. Second, we used the Banff 2009 classification, which lacks the criteria for C4d-negative AMR. We added FlowPRA testing only when we found microvascular inflammation with peritubular capillary C4d deposition. Thus, we might have missed the diagnosis of some cases of subclinical AMR or C4d-negative AMR. Third, we do not routinely perform electron microscopic observation, and thus might have missed early ultrastructural changes, such as endothelial injuries in the peritubular or glomerular capillaries.
In summary, histological analysis using 3- and 12-month PBs suggested comparable allograft pathology between ABO-C KTx and ABO-I KTx under our current desensitization protocol of low-dose rituximab and plasmapheresis. The incidence of both acute rejection and infectious complications was not increased in ABO-I KTx under the current desensitization protocol with induction and maintenance of immunosuppression.
The authors thank Ms. H. Noguchi for providing excellent technical assistance.
1. Opelz G, Morath C, Süsal C, et al. Three-year outcomes following 1420 ABO-incompatible living-donor kidney transplants performed after ABO antibody reduction: results from 101 centers. Transplantation
2. Shishido S, Hyodo YY, Aoki Y, et al. Outcomes of pediatric ABO-incompatible kidney transplantations are equivalent to ABO-compatible controls. Transplant Proc
3. Okumi M, Toki D, Nozaki T, et al. ABO-incompatible living kidney transplants: evolution of outcomes and immunosuppressive management. Am J Transplant
4. Habicht A, Bröker V, Blume C, et al. Increase of infectious complications in ABO-incompatible kidney transplant recipients-a single centre experience. Nephrol Dial Transplant
5. Lentine KL, Axelrod D, Klein C, et al. Early clinical complications after ABO incompatible live donor kidney transplantation: a national study of Medicare-insured patients. Transplantation
6. Shirakawa H, Ishida H, Shimizu T, et al. The low dose of rituximab in ABO-incompatible kidney transplantation without a splenectomy: a single-center experience. Clin Transplant
7. Gloor JM, Cosio FG, Rea DJ, et al. Histologic findings one year after positive crossmatch or ABO blood group incompatible living donor kidney transplantation. Am J Transplant
8. Setoguchi K, Ishida H, Shimmura H, et al. Analysis of renal transplant protocol biopsies in ABO-incompatible kidney transplantation. Am J Transplant
9. Ushigome H, Okamoto M, Koshino K, et al. Findings of graft biopsy specimens within 90 days after ABO blood group incompatible living donor kidney transplantation compared with ABO-identical and non-identical transplantation. Clin Transplant
. 2010;24(Suppl 22):16–21.
10. Dörge C, Mjøen G, Strøm HJ, et al. One-year protocol biopsies from ABO-incompatible renal allografts compared with a matched cohort of ABO-compatible allografts. Clin Transplant
11. Masutani K, Kitada H, Tsuchimoto A, et al. Usefulness of 3-month protocol biopsy of kidney allograft to detect subclinical rejection under triple immunosuppression with basiliximab: a single center experience. Clin Exp Nephrol
12. Masutani K, Tsuchimoto A, Haruyama N, et al. Protocol biopsy findings in living donor kidney transplant patients treated with once-daily or twice-daily tacrolimus formulation. Transplant Proc
13. Sis B, Mengel M, Haas M, et al. Banff ’09 meeting report: antibody mediated graft deterioration and implementation of Banff working groups. Am J Transplant
14. Hirsch HH, Randhawa P. and the AST infectious disease community of practice. BK virus in solid organ transplant recipients. Am J Transplant
. 2009;9(Suppl 4):S136–S146.
15. Matsuo S, Imai E, Horio M, et al. Collaborators developing the Japanese equation for estimated GFR: revised equations for estimated GFR from serum creatinine in Japan. Am J Kidney Dis
16. Schwarz GJ, Muñoz A, Schneider MF, et al. New equations to estimate GFR in children with CKD. J Am Soc Nephrol
17. Alexandre GP, Squifflet JP, De Bruyere M, et al. Present experiences in a series of 26 ABO-incompatible living donor renal allografts. Transplant Proc
18. Sawada T, Fuchinoue S, Teraoka S. Successful A1-to-O, ABO-incompatible kidney transplantation after a preconditioning regimen consisting of anti-CD20 monoclonal antibody infusions, splenectomy, and double filtration plasmapheresis. Transplantation
19. Sawada T, Fuchinoue S, Kawase T, et al. Preconditioning regimen consisting of anti-CD20 monoclonal antibody infusions, splenectomy and DFPP-enabled non-responders to undergo ABO-incompatible kidney transplantation. Clin Transplant
20. Tydén G, Kumlien G, Genberg H, et al. ABO incompatible kidney transplantations without splenectomy, using antigen-specific immunoadsorption and rituximab. Am J Transplant
21. Toki D, Ishida H, Setoguchi K, et al. Acute antibody-mediated rejection in living ABO-incompatible kidney transplantation: long-term impact and risk factors. Am J Transplant
22. Couzi L, Perera R, Manook M, et al. Incidence and outcome of C4d staining with tubulointerstitial inflammation in blood group-incompatible kidney transplantation. Transplantation
23. Chung BH, Yun JT, Ha SE, et al. Combined use of rituximab and plasmapheresis pre-transplant increases post-transplant infections in renal transplant recipients with basiliximab induction therapy. Transpl Infect Dis
24. Sharif A, Alachkar N, Bagnasco S, et al. Incidence and outcomes of BK virus allograft nephropathy among ABO- and HLA-incompatible kidney transplant recipients. Clin J Am Soc Nephrol
Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved.
25. Becker LE, Siebert D, Süsal C, et al. Outcomes following ABO-incompatible kidney transplantation performed after desensitization by nonantigen-specific immunoadsorption. Transplantation