Journal Logo

Clinical and Translational Research

Early and Late Acute Antibody-Mediated Rejection Differ Immunologically and in Response to Proteasome Inhibition

Walsh, R. Carlin1; Brailey, Paul2; Girnita, Alin2; Alloway, Rita R.3; Shields, Adele Rike1; Wall, Garth E.; Sadaka, Basma H.; Cardi, Michael4; Tevar, Amit1; Govil, Amit3; Mogilishetty, Gautham3; Roy-Chaudhury, Prabir3; Woodle, E. Steve1,5

Author Information
doi: 10.1097/TP.0b013e318218e901
  • Free


It has long been known that acute rejection episodes differ in their effects on renal allograft survival. Over the past three decades, several studies demonstrated that late acute rejection episodes (defined as occurring 3 or 6 months after transplantation) had a substantially greater negative impact on renal allograft survival than those occurring in the early posttransplant period (1–4). In addition, histologic grades of acute rejection have also been demonstrated to influence long-term renal allograft survival (4).

Diagnostic criteria and characterization of antibody-mediated rejection (AMR) is rapidly evolving. Brennan and coworkers (5) have demonstrated that focal C4d has similar implications to diffuse C4d staining in terms of long-term renal allograft survival. Halloran and coworkers (6) have recently provided evidence that pathologic assessment of microcirculatory injury rather than C4d should be a primary criterion for AMR diagnosis. This study, and others (7–9) have indicated that T-cell mediated acute cellular rejection (ACR) does not seem to impose a significant negative effect on long-term renal allograft survival, but rather it is AMR and mixed acute rejection (MAR) episodes that connote significant negative influences on renal allograft survival.

Improved ability to detect humoral components of acute rejection and the demonstration of the substantial effects on long-term renal allograft survival have led to recognition of the need to develop targeted and more effective antihumoral therapies so that long-term outcomes can be improved. Increasing experience is emerging with the use of proteasome inhibitor (PI) therapy in treating AMR (10–20). To date, PI has been demonstrated to provide effective therapy in refractory AMR and MAR in renal transplant recipients (10–14, 18–20). In addition, more recent reports have demonstrated that PI provides effective front line therapy for AMR in renal allograft recipients (12). Improved monitoring of human leukocyte antigen (HLA)-specific antibodies against the donor (donor-specific antibodies [DSAs]) afforded by widespread application of solid phase assays has allowed serial monitoring of DSA levels during AMR therapy. This has allowed reliable, ongoing evaluation of the effects of AMR therapy on DSA levels in transplant recipients with AMR or MAR. As such, these advances are rapidly changing how AMR is diagnosed and treated.

With 2 years of experience with PI therapy for acute AMR and acute MAR, we have empirically observed that the timing of acute AMR episodes after transplantation seems to significantly influence PI therapeutic outcomes. Similar to historical reports that late acute ACR conferred a poor prognosis (21–23), our recent experiences with late AMR have also demonstrated a poor prognosis. Our previous work, however, has shown that ACR, in the absence of AMR, regardless of whether it is early or late, is associated with good long-term renal allograft survival (8). The purpose of this study therefore was to characterize and compare early and late acute AMR, and also to compare responses with PI therapy in early and late acute AMR in renal transplant recipients.


Demographic data, including immunologic risk factors and rejection characteristics, are presented in Table 1. When presented below, results are expressed as early versus late acute AMR.

Demographic data, immunologic risk factors, histology, and rejection characteristics

A total of 30 acute AMR episodes were treated with a PI-based regimen. Of these rejection episodes, 36% (11/30: 3/13 early and 8/17 late) were treatment refractory to plasmapheresis±rituximab±intravenous immune globulin± rabbit anti-thymocyte globulin (Thymoglobulin, Genzyme, Cambridge, MA). Seventy-three percentage of AMR episodes occurred in patients with renal allografts alone (10 early and 12 late), 20% occurred in simultaneous kidney-pancreas transplant recipients (1 early and 5 late), and 7% occurred in pancreas after kidney transplant recipients (2 early). Mean follow-up was similar between groups with 6.6±4.9 months for early acute AMR patients compared with 7.0±6.0 months in late acute AMR (P=0.85).

Preformed HLA antibody was a risk factor for early acute AMR as 23.1% of patients had a cytotoxic panel reactive antibody (PRA) greater than 50% at the time of transplant compared with none in the late AMR group (P=0.09). Additionally, early AMR patients were numerically more likely to be repeat transplant recipients (15.4% vs. 0%, P=0.18) or women (69% vs. 35%, P=0.07). Rejection due to noncompliance tended to be more prevalent in the late AMR group (23.1% vs. 47.1%, P=0.09).

More than half of all rejection episodes manifested as pure AMR. Single-antigen bead saturation was evident in 23.5% of late AMR patients but was not apparent in patients with early AMR (P=0.09). The mean pretreatment (untitered) immunodominant DSA (iDSA) level was greater in late AMR episodes (6913±4206 vs. 11,533±6143 mean fluorescence intensity [MFI]; P=0.03) (Table 2). However, the number and classification of total DSA specificities (2.6±1.7 vs. 2.6±1.6; P=0.96), total number of HLA class I DSA specificities (1.2±1.1 vs. 0.8±1.0; P=0.23), and total number of HLA class II DSA specificities (1.4±1.0 vs. 1.8±0.8; P=0.21) were similar between groups. The iDSA specificity was more likely to be an HLA class I specificity in early acute AMR (38.5% vs. 0%; P<0.01). An HLA DQ specificity iDSA was more prevalent in late acute AMR (38.5% vs. 76.5%; P=0.04).

Outcomes and toxicity

Therapeutic Outcomes

Improvement in immunologic, histologic, and allograft function parameters were greater after PI-based therapy in early than late AMR (Table 2). The mean percent reduction in iDSA level at nadir in early AMR was significantly greater than that in late AMR (81.5%±21.1% vs. 51.4%±27.6%; P<0.01). This pattern of iDSA reduction (Fig. 1A) was detectable as early as 7 days posttreatment (38.6%±47.3% vs. 17.6%±27.6%; P=0.15), and persisted at 14 days (60.1%±30.4% vs. 34.8%±35.1%; P=0.05) and 30 days posttreatment (77.0%±20.4% vs. 42.4%±30.3%; P<0.01). A greater percentage of early AMR patients reached a more than 50% reduction in iDSA by treatment day 14 (76.9% vs. 35.3%; P=0.03).

(A) Comparison of percent reduction in immunodominant donor-specific antibody (iDSA) titer posttreatment for early or late rejection at 7, 14, and 30 days after bortezomib-based treatment initiation. (B) Comparison of estimated glomerular filtration rate pre- and posttreatment in early and late posttransplant acute antibody-mediated rejection (AMR).

Renal allograft biopsies which were C4d positive tended to be more prevalent in early AMR (69.2% vs. 52.9%; P=0.54). When a repeat renal allograft biopsy was performed, resolution, or improvement in AMR was noted in 87.5% of early vs. 53.8% of late AMR episodes (P=0.13).

Estimated glomerular filtration rate (eGFR) was lower both before and after AMR therapy in the late group. Before treatment, the eGFR was 40±17 mL/min/1.73 m2 in early AMR vs. 27±12 mL/min/1.73 m2 in late AMR (P=0.02). Renal function improved significantly with treatment in both groups (Fig. 1B); however, it was still significantly better in early AMR after treatment (66±31 mL/min/1.73 m2 vs. 37±25 mL/min/1.73 m2; P<0.01). Twelve-month noncompliance censored allograft survival was 90% in early AMR and 80% in late AMR (P=0.726; log-rank).

Adverse Events

The incidence of adverse events and toxicities associated with PI therapy were similar between groups (Table 3). Hematologic toxicities are the most common adverse events noted with PI therapy. All patients were relatively anemic at the time of treatment initiation with mean hemoglobin values of 9.8±0.9 gm/dL and 10.3±1.5 gm/dL (P=0.35). Moderate anemia (Common Terminology Criteria for Adverse Events [CTCAE] grade 3) defined as a hemoglobin less than 8.0 to 6.5 gm/dL, was observed in 53.8% of early and 35.3% of late acute AMR patients (P=0.26). Moderate thrombocytopenia (CTCAE grade 2) defined as a platelet count less than 75,000 to 50,000 cells/mm3, was observed in 15.4% and 5.9% of early and late AMR patients (P=0.40). More severe thrombocytopenia was not observed in the early AMR patients; however, 23.5% of late AMR patients experienced CTCAE grade 3 thrombocytopenia (<50,000–25,000 cells/mm3), and 5.9% experienced CTCAE grade 4 thrombocytopenia (<25,000 cells/mm3). Induction therapy was analyzed to evaluate a potential effect on thrombocytopenia risk; however, no trends were observed (data not shown). Because all patients received tacrolimus+mycophenolate maintenance therapy, the effects of maintenance immunosuppression on thrombocytopenia could not be evaluated.

Adverse events and toxicities

Baseline peripheral neuropathy was observed in 7.7% of early AMR patients and 17.6% of late AMR patients (P=0.41). Severe peripheral neuropathy (CTCAE grade 3 or 4) was not reported in either treatment group. Moderate (CTCAE grade 2) peripheral neuropathy was reported in 7.7% and 17.6% of early and late AMR patients (P=0.41). Mild peripheral neuropathy was more prevalent in late AMR patients. In all patients with new-onset peripheral neuropathy, symptoms resolved after treatment of rejection.

Mild (CTCAE grade 1) nausea, vomiting, and diarrhea tended to be more common in late AMR patients. A total of 15.4% of early AMR patients reported moderate diarrhea with treatment.

The incidence of opportunistic infection was greater with early AMR treatment (15.4% vs. 0%; P=0.18), with 1 patient each experiencing cytomegalovirus viremia and polyoma virus viremia, both of which resolved with therapy. With more than 6 months follow-up, malignancy has not been observed in either group.


Acute rejection episodes occurring late after renal transplantation have long been known to be associated with poor renal allograft survival (1–4). These earlier studies, however, focused only on the timing of acute rejection, and not on immunologic mechanisms (ACR vs. AMR). The recent application of C4d staining and solid phase assays for anti-HLA antibody detection has greatly enhanced AMR detection and classification of acute rejection episodes as AMR, MAR, or ACR. This has led to recent observations that AMR and MAR significantly reduce renal allograft survival, whereas ACR imposes a minimal effect on allograft survival (6–9).

Logically, subsequent studies should determine whether subclasses of AMR exist that may be responsible for a predominant effect on graft survival. Therefore, the present study compared early and late acute AMR. Early AMR episodes were characterized by iDSA that were of similar frequencies with respect to class I and class II. In contrast, iDSA in late AMR episodes were almost exclusively Class II in general, with the vast majority being DQ specific. Of note, Halloran and coworkers (6) have demonstrated a propensity for class II DSA being present at the time of biopsy for late allograft dysfunction. To our knowledge, the present study provides the first demonstration of propensity for DQ locus as iDSA in late AMR. No late AMR patient in the present study was sensitized to HLA antigens before transplantation, whereas a significant proportion of early AMR patients had preexisting HLA sensitization. These observations indicate that the immunologic nature of early and late AMR are fundamentally distinct, and as such may require therapies targeting differing aspects of the humoral immune response. As we will discuss later, these observations may have important implications for PI therapy and possibly for plasma cell-targeted therapies in general.

Accepted guidelines for evaluation of therapeutic responses in the setting of AMR do not exist. We have used three criteria for evaluating PI therapy: histologic, serologic (viz., DSA levels), and renal function (estimated/calculated GFR). Experience from the present study indicates that early and late AMR differ with respect to improvement in these three general criteria under PI therapy. PI therapy provided improvements in all three criteria in both early and late AMR; however, PI therapy provided greater improvement in iDSA levels in early AMR as compared with late AMR. Both early and late AMR demonstrated improvements in renal function and allograft histology; however, the late AMR group demonstrated worse renal function and a greater degree of chronic histologic changes before PI therapy. With increasing experience, we have noted that responses to PI therapy for AMR or MAR are not uniform, but rather demonstrate a spectrum of responses. Part of the reason for this variability may be due to differences in early and late AMR.

Previous studies have described results with early and late AMR with conventional AMR therapies (21–23) (Table 4). Glotz and coworkers (22) have recently described factors that influence outcome in renal allograft recipients experiencing AMR. Histologic factors found to be associated with “bad outcomes” included glomerular polymorphonuclear neutrophils (PMNs), peritubular capillary PMNs, peritubular capillary dilatation, and interstitial edema. Pretransplant DSA did not influence outcome, but the persistence of DSA after treatment was associated with a “bad outcome” (22). Results for the present study would suggest that the timing of AMR also substantially affects outcomes.

Studies comparing early and late AMR

Sun and colleagues (21) reported on results for early and late AMR treated with high dose corticosteroids, tacrolimus and mycophenolate mofetil, and immunoglobulin adsorption with protein A columns. Response to therapy was based on serum creatinine (SCr) response at 1 month: complete reversal required return of SCr to “normal range,” partial reversal required decrease in SCr but outside “normal range,” and controlled rejection required that creatinine not decrease and that patient remained off dialysis. Five patients experienced AMR between 15 and 180 days posttransplant: two (40%) experienced partial reversal, one (20%) was controlled, and two (40%) lost their allograft. One year renal allograft survival for early AMR was 60%. Twelve patients experienced late AMR (beyond 180 days): no patient experienced complete or partial reversal, seven (58%) were controlled and five (42%) lost their allograft. One year renal allograft survival for late AMR was 50%.

Pefaur et al. (23) have also reported on early and late AMR treated with intravenous immune globulin-based regimens that included tacrolimus and mycophenolate mofetil therapy. Rejection reversal was observed in all four patients with early AMR; however, two of three patients with late AMR experienced allograft loss.

In this study, a substantial proportion of patients with late AMR were documented to be noncompliant. Of the noncompliant patients, some denied noncompliance when questioned at the time of AMR diagnosis. Later, however, when marked fluctuations were noted in CNI levels, some admitted to noncompliance. Previous studies have documented that late ACR and late AMR are not infrequent in noncompliant patients (24).

It is plausible that ongoing noncompliance may contribute significantly to the poorer therapeutic results for late AMR. This possibility emphasizes the need for noncompliance detection in late AMR, because noncompliance may preclude demonstration of therapeutic superiority of new AMR therapies. If noncompliance can be identified with a high sensitivity, it will therefore allow a better assessment of the efficacy of new regimens for late AMR. Accurate identification with censoring or subset analysis of noncompliant patients may be a critical factor for future AMR studies.

Fundamental differences may exist between early and late AMR with respect to underlying B cell and plasma cell immunobiology that have important implications for PI therapy. In early AMR, the response may be primary or anamnestic, both of which involve B-cell proliferative responses. Actively proliferating cell populations are expected to be sensitive to PI inhibition, as PI inhibition is known to effectively induce cell cycle arrest and resultant programmed cell death in actively proliferating cells, a property that is currently being exploited in clinical trials of several types of hematologic malignancies (25).

Acute AMR that results from a primary antibody response generates significant numbers of short-lived plasma cells, which may have significant susceptibility to proteasome inhibition, because these plasma cells are likely not preferentially located in an environment where survival signals are prominent. Because short-lived plasma cells are destined for early death, it is likely that they are more susceptible to proteasome inhibition than niche-resident long-lived plasma cells that are continuously in receipt of multiple survival signals.

Both primary and anamnestic responses are characterized by marked proliferations of B cell populations. In anamnestic responses, memory B cells undergo a rapid onsent of vigorous proliferation with production of large numbers of plasmablasts that give rise to mature plasma cells that produce large amounts of high affinity allospecific antibody, many of which are destined to become long-lived, niche- resident plasma cells. Therefore, in anamnestic humoral responses, the rapid production of large numbers of plasma cells producing high levels of class switched, somatically hypermutated, high affinity antibodies has the potential to induce a greater degree of allograft injury within a shorter period of time.

Late AMR is likely to be fundamentally different than early AMR for several reasons. First, in our experience, a substantial proportion of late AMR episodes occur in patients who have been documented to be noncompliant. In addition, late AMR episodes occur in patients whose clinic visits and SCr determinations tend to be substantially farther apart, and thus creating the likelihood that late AMR episodes may have been in existence for a few to several weeks (or longer) before diagnosis. In contrast, early AMR occurs during a time when renal function is being frequently evaluated. Thus, early AMR is likely to be detected early in its course, whereas in late AMR, the AMR process may have had weeks to develop, during which a substantial population of bone marrow niche- resident plasma cells will have been generated. Finally, the higher DSA levels with late AMR suggest that late AMR may be associated with a greater plasma cell mass that is producing the DSA. If true, an overall greater PI exposure may be required for adequate therapy for late AMR, as compared with early AMR.

Bone marrow niche-resident plasma cells, as a result of evolutionary pressure, provide a means for sustaining immunologic memory for decades or more in humans (26). Two primary mechanisms have been proposed and debated regarding the mechanism by which long-term plasma cell memory is maintained: the “tick-over” theory and the “long-term survivor” theory. The “tick-over” theory holds that long term plasma cell memory is maintained by constant, ongoing production and replacement of bone marrow niche-resident plasma cells from the memory B-cell population. The “long-term survivor” theory holds that once plasma cells home to bone marrow niches, they survive for many years—decades or longer. It is now known that the bone marrow niches occupied by plasma cells provide the basic mechanisms for long-term plasma cell survival by the continuous delivery of multiple survival signals for plasma cells. This concept of long-lived plasma cell populations is currently thought to be the predominant mechanism by which long-term humoral memory is maintained in humans (26). These basic aspects of plasma cell and B-cell biology have important implications for PI therapy, particularly when combination therapies with memory B-cell depleting agents are being contemplated.

In contrast to early AMR, late AMR patients had significant residual iDSA levels after PI therapy, which is a plausible explanation for the worse long-term outcomes experienced with late AMR. The relatively high residual iDSA levels after bortezomib therapy likely result from residual (i.e., surviving) plasma cell clonal populations and if residual DSA production is an important pathophysiologic factor in the reduced long-term renal allograft survival observed with late AMR, repeated cycles of bortezomib to further deplete the residual plasma cell clonal population (with a goal to eliminate DSA) may be a reasonable therapeutic approach.

This study has limitations that are important for consideration in future studies. One of these limitations is the current diagnostic criterion for AMR, which sometimes are inadequate for all AMR cases, primarily due to the insensitivity of C4d staining. Although current standard criteria were used (Banff and AMR Consensus Conference criteria), a recent study by Halloran and coworkers (6, 7) has proposed new criteria for AMR, which proposes to use the presence of DSA and microcirculatory changes on light microscopy (glomerulitis, peritubular capillaritis, peritubular capillary multilayering, and glomerulopathy), and disregards C4d staining. This concept provides the advantage of negating the problem of C4d insensitivity in diagnosing AMR and importantly maintains that a negative C4d stain, or a biopsy that does not meet Banff criteria for AMR or ACR (or both), should not preclude a diagnosis of AMR. Therefore, the presence of a normal appearing biopsy in a patient with rising DSA and worsening renal function should warrant a diligent search for evidence of endothelial injury on renal allograft biopsy by both light microscopy and electron microscopy.

In conclusion, this experience demonstrates that early and late AMR are fundamentally distinct immunologic entities and that early AMR episodes demonstrate better results with PI therapy than late AMR episodes. A plausible explanation for these differences is that early and late AMR are associated with distinct differences in memory B-cell and plasma cell populations. Observations from this study have important implications for future development of AMR therapies as they highlight the need for improved results for the late AMR group.

It is now reasonable to hypothesize that suboptimal results with late AMR therapy occur in part due to a failure to deplete resistant plasma cell populations. Strategic development of therapies that target the distinct memory B-cell and plasma cell populations in early and late AMR may provide means for improving the poor long-term results with AMR and MAR in renal transplant recipients.


Institutional Review Board approval was obtained from the University of Cincinnati and Christ Hospital (Cincinnati, OH) allowing for the use of patient data for publication.

Subjects and AMR Treatment

Thirteen early posttransplant and 17 late posttransplant AMR episodes in the presence or absence of concomitant ACR were treated with combination therapy consisting of bortezomib (Velcade, Millennium, Cambridge, MA), rituximab (Rituxan, Genentech, San Francisco, CA), and plasmapheresis. Bortezomib dosing was per labeled dose (1.3 mg/m2×four doses over 11 days). Rituximab dosing was also per labeled dose (375 mg/m2) and given as a single dose immediately before the first bortezomib dose, if the patient had not previously received rituximab for rejection therapy. Plasmapheresis was performed using anticoagulant citrate dextrose with 1.5 plasma volumes per session, with replacement preferentially with 5% albumin. Plasmapheresis was performed immediately before each bortezomib dose, with three daily sessions performed beginning 72 hr after the final bortezomib dose. Maintenance immunosuppression was not altered during bortezomib therapy. Beginning with initiation of bortezomib therapy, the following prophylaxis was given: (1) cytomegalovirus: valgancyclovir (225 mg orally daily for 90 days, adjusted based on renal function), (2) nystatin (5 mL orally three times per day for 90 days), and trimothoprim/sulfamethoxazole (or pentamidine or dapsone) for 365 days. Early posttransplant rejection was defined as occurring within the first 6 months after kidney transplant, and late posttransplant rejection was defined as occurring more than 6 months posttransplant. Noncompliance was diagnosed by detection of nonmeasurable tacrolimus levels, and by patient admission to coordinators or physicians.

Rejection Diagnosis

Acute rejection was defined as an increase in SCr at least 20% above baseline SCr with histologic evidence of acute rejection defined on renal allograft biopsy by Banff 1997 criteria (update 2007) (27). Per Banff criteria, renal allograft biopsies were considered consistent with AMR if the two of the three following characteristics were present: (1) donor-specific anti-HLA antibody (DSA); (2) histologic changes consistent with AMR; and (3) positive C4d staining in peritubular capillaries (PTCs)±other structures are present. Per previously published consensus conference criteria (28), AMR was also considered present if three of four criteria were present: (1) light microscopic features of AMR on hematoxylin-eosin (H&E) stained preparation (capillaritis, glomerulitis, interstitial PMN infiltrate, and interstitial edema), (2) C4d staining in PTCs, (3) presence of DSA, and (4) renal allograft dysfunction. MAR was defined as ACR with AMR. C4d staining was performed by immunohistochemistry on paraffin sections using a rabbit polyclonal antibody specific for human C4d (C4dpAb; Alpco Diagnostics, Windham, NH) using a Ventana Benchmark automated stainer (Ventana Medical Systems, Inc., Tucson, AZ) and horseradish peroxidase diaminobenzidine detection. Hematoxylin was used as counterstain. C4d staining in PTCs was considered to be positive if linear circumferential capillary staining was identified in multiple PTCs, excluding scarred or necrotic areas. C4d staining of glomerular capillaries was considered positive if multiple glomeruli were noted to have multiple capillary loops with linear C4d staining. Resolving rejection was defined as improvement of Banff score and resolution of C4d positivity. All evaluations performed by a single pathologist (L.J.A.).

Renal Function Evaluation

Renal function was assessed by eGFR as determined by the modification of diet in renal disease formula (29).

DSA Testing

Donor-specific anti-HLA antibodies (DSA) were identified using antigen bead panels by Luminex assay (LABScreen, One Lambda, Canoga Park, CA). Donor and recipient HLA typing were performed by molecular methods (rSSO, SSp, and SBT). Pre- and posttransplant screening for HLA-specific antibody was performed both by cytoxicity-PRA and solid phase methods (flow PRA and Luminex). Results were reported as MFI. iDSA was defined as the highest level DSA specificity at the time of rejection diagnosis. Titers were preformed in patients with iDSA levels greater than 15,000 MFI to evaluate for single antigen bead saturation. Percent reduction in iDSA over time was evaluated by MFI for DSA without evidence of bead saturation.

Toxicity Evaluation

Hematologic, gastrointestinal, and neuropathic toxicity of bortezomib treatment were graded with the CTCAE version 3.0 (30).

Statistical Analysis

All analyses were performed according to the intention-to-treat principle. Bivariate distributions were evaluated for normality using standard statistical methods and by examining stratified distribution plots. Nonparametric measures of comparison were used for variables evaluated as not normally distributed. Difference testing between groups was performed using the two-tailed student t test. Nominal data was evaluated using the one-sided Fisher's exact test. Allograft survival rates were calculated according to the Kaplan-Meier method; univariate survival distributions were compared with the log-rank test. P values less than 0.05 were considered significant. Statistical analyses were performed using the statistical package STATA/SE™ version 10.1 (StataCorp, College Station, TX).


1. Massy ZA, Guijarro C, Wiederkehr MR, et al. Chronic renal allograft rejection: Immunologic and nonimmunologic risk factors. Kidney Int 1996; 49: 518.
2. Humar A, Kerr S, Gillingham KJ, et al. Features of acute rejection that increase risk for chronic rejection. Transplantation 1999; 68: 1200.
3. Basadonna GP, Matas AJ, Gillingham KJ, et al. Early versus late acute renal allograft rejection: Impact on chronic rejection. Transplantation 1993; 55: 993.
4. Sipkens YW, Doxidis II, Mallat MJK, et al. Early versus late acute rejection episodes in renal transplantation. Transplantation 2004; 75: 204.
5. Kedainis RL, Koch MJ, Brennan DC, et al. Focal C4d+ in renal allografts is associated with the presence of donor-specific antibodies and decreased allograft survival. Am J Transplant 2009; 9: 812.
6. Hidalgo LG, Campbell PM, Sis B, et al. De novo donor-specific antibody at the time of kidney transplant biopsy associates with microvascular pathology and late graft failure. Am J Transplant 2009; 9: 2532.
7. Einecke G, Sis B, Reeve J, et al. Antibody-mediated microcirculation injury is the major cause of late kidney transplant failure. Am J Transplantation 2009; 9: 2520.
8. Everly MJ, Everly JJ, Arend LJ, et al. Reducing de novo donor-specific antibody levels during acute rejection diminishes renal allograft loss. Am J Transplant 2009; 9: 1063.
9. Haririan A, Kiangkitiwan B, Kukuruga D, et al. The impact of c4d pattern and donor-specific antibody on graft survival in recipients requiring indication renal allograft biopsy. Am J Transplant 2009; 9: 2758.
10. Everly MJ, Everly JJ, Susskind B, et al. Bortezomib provides effective therapy for antibody- and cell-mediated acute rejection. Transplantation 2008; 86: 1754.
11. Everly JJ, Walsh RC, Alloway RR, et al. Proteasome inhibition for antibody-mediated rejection. Curr Opin Organ Transplant 2009; 14: 662.
12. Walsh RC, Everly JJ, Brailey P, et al. Proteasome inhibitor-based primary therapy for antibody-mediated renal allograft rejection. Transplantation 2010; 89: 277.
13. Walsh RC, Brailey P, Girnita A, et al. Woodle proteasome inhibitor-based antihumoral therapy: Differential response in early versus late antibody mediated rejection. Am J Transplant 2010; 10(suppl 4): 164.
14. Woodle ES, Light J, Rubin M, et al. Proteasome inhibitor therapy for antibody mediated rejection: Initial report from a multicenter collaborative. Am J Transplant 2010; 10(suppl 4): 83.
15. Eckman PM, Thorsgard M, Maurer D, et al. Bortezomib for refractory antibody-mediated cardiac allograft rejection. Clin Transpl 2009: 475.
16. Govil A, Walsh RC, Tevar A, et al. Bortezomib-based treatment of antibody mediated rejection in pancreas allograft recipients. Clinical Tranpslants 2009: 443.
17. Perry DK, Burns JM, Pollinger HS, et al. Proteasome inhibition causes apoptosis of normal human plasma cells preventing alloantibody production. Am J Transplant 2009; 9: 201.
18. Shapiro R, Basu A, Zeevi A, et al. Bortezomib rescue in refractory acute humoral rejection—Report of a case. Clin Transpl 2009: 431.
19. Hardinger KL, Alford K, Murillo D. Bortezomib for acute humoral rejection in two repeat transplant recipients. Clin Transpl 2009; 479.
20. Djamali A, Buth BL, Torrealba J, et al. Bortezomib as a rescue therapy for hyperacute and multi-drug resistant mixed acute rejection after kidney transplantation. Clin Transpl 2009; 485.
21. Sun Q, Liu ZH, Ji S, et al. Late and early C4d-positive acute rejection: Different clinico-histopathological subentities in renal transplantation. Kidney Int 2006; 70: 377.
22. Lefaucheur C, Nochy D, Hill GS, et al. Determinants of poor graft outcome in patients with antibody-mediated rejection. Am J Transplant 2007; 7: 832.
23. Pefaur J, Diaz P, Panace R, et al. Early and late humoral rejection: A clinicopathologic entity in two times. Transplant Proc 2008; 40: 3229.
24. Lerut E, Kuypers DR, Verbaken E, et al. Acute rejection in noncompliant renal allograft recipients: A distinct morphology. Clin Transplant 2007; 21: 344.
25. Bonvini P, Zorzi E, Basso G, et al. Bortezomib-mediated 26S proteasome inhibition causes cell-cycle arrest and induces apoptosis in CD-30+ anaplastic large cell lymphoma. Leukemia 2007; 21: 838.
26. Amanna IJ, Carlson NE, Slitkin MK. Duration of humoral immunity to common viral and vaccine antigens. N Engl J Med 2007; 357: 1903.
27. Solez K, Colvin RB, Racusen LC, et al. Banff 07 classification of renal allograft pathology: Updates and future directions. Am J Transplant 2008; 8: 753.
28. Takemoto SK, Zeevi A, Feng S, et al. National conference to assess antibody-mediated rejection in solid organ transplantation. Am J Transplant 2004; 4: 1033.
29. Levey AS, Coresh J, Greene T, et al. Using standardized serum creatinine values in the modification of diet in renal disease study equation for estimating glomerular filtration rate. Ann Intern Med 2006; 145: 247.
30. Accessed March 15, 2011.

Proteasome inhibition; Bortezomib; Rituximab; Donor-specific antibody; Antibody-mediated rejection; Mixed acute rejection

© 2011 Lippincott Williams & Wilkins, Inc.