Journal Logo

Original Clinical Science—General

A Randomized Controlled Clinical Trial Comparing Belatacept With Tacrolimus After De Novo Kidney Transplantation

de Graav, Gretchen N. MD1; Baan, Carla C. PhD1; Clahsen-van Groningen, Marian C. MD, PhD2; Kraaijeveld, Rens BSc1; Dieterich, Marjolein BSc1; Verschoor, Wenda BSc1; von der Thusen, Jan H. MD, PhD2; Roelen, Dave L. PhD3; Cadogan, Monique BSc1; van de Wetering, Jacqueline MD, PhD1; van Rosmalen, Joost PhD4; Weimar, Wilem MD, PhD1; Hesselink, Dennis A. MD, PhD1

Author Information
doi: 10.1097/TP.0000000000001755

Belatacept, an inhibitor of the CD28-CD80/86 costimulatory pathway, has the potential to improve long-term outcomes of kidney transplantation.1-5 Seven-year follow-up of the Belatacept Evaluation of Nephroprotection and Efficacy as First-Line Immunosuppression Trial study demonstrated a higher patient and graft survival, as well as better graft function in patients who were treated with belatacept as compared with ciclosporin.1 Nonetheless, the higher incidence and severity grade of acute rejection (AR) that have been observed among belatacept-treated patients remain a concern.6-9 Up until now, belatacept has not been compared head-to-head with tacrolimus in randomized-controlled trials (RCTs) in kidney transplantation without the use of lymphocyte-depleting therapy.10-12 Observations made in uncontrolled studies suggest that the performance of belatacept in terms of preventing AR as compared with tacrolimus may be inferior.13,14

Identification of patients' pretransplantation that will develop AR during belatacept treatment would greatly help to personalize immunosuppressive therapy and maximize the potential of the drug. Experimental studies in rhesus macaques and ex vivo studies using human lymphocytes have demonstrated that antigen-experienced, cytotoxic CD28CD8+ T cells are not dependent on costimulatory signaling via CD80/86 and are therefore less susceptible to the immunosuppressive effects of belatacept.15-17 Recently, Espinosa and colleagues18 suggested that patients with a high frequency of cytotoxic CD57+PD1CD4+ T cells were at increased risk of AR during belatacept treatment. A preliminary study in nonhuman primates suggested another biomarker for AR under belatacept, namely, CD28++ end-stage terminally differentiated memory (EMRA) CD8+ T cells that rapidly downregulate CD28 after kidney transplantation.19 Biomarkers such as these may help in risk stratification and a more rational use of belatacept, but require prospective validation.

Alternatively, therapeutic drug monitoring (TDM) of belatacept therapy may improve outcomes. Because serum belatacept concentrations tend to vary little between individual patients, pharmacokinetic TDM is currently not recommended.5,20 However, pharmacodynamic TDM of belatacept is feasible. Ex vivo flow cytometric measurement of CD86 occupancy on monocytes by belatacept reflects effector T cell function,21 demonstrating the potential of TDM to improve the outcomes of belatacept therapy. However, no data from prospective clinical trials are available to provide guidance in this respect.

Here, the results of a RCT are reported in which 40 patients were randomized to receive either belatacept- or tacrolimus-based immunosuppressive therapy after de novo kidney transplantation. The primary aims of this RCT were to compare the AR rate between belatacept-treated and tacrolimus-treated patients and to identify biomarkers that were predictive of AR.


Refer to Supplemental Digital Content (SDC,, Materials and Methods for additional and detailed information.

Study Design

This was an investigator-initiated, prospective, randomized-controlled, parallel group, open-label, single-center, clinical trial. Adult patients (≥18 years) who were scheduled to receive a single-organ, blood group ABO-compatible kidney from a living donor at the Erasmus MC, Rotterdam, the Netherlands, were eligible for participation. Historical and current crossmatch-dependent cytotoxicity tests were negative. Table 1 lists the inclusion and exclusion criteria in detail. The study was approved by the institutional review board of the Erasmus MC (Medical Ethical Review Board number 2012-421) and was registered in the Dutch national trial registry (; number NTR4242, registered October 2013). Written informed consent was obtained from all patients before inclusion and randomization. The study was carried out in compliance with the Good Clinical Practice guidelines ( and the Declaration of Istanbul.22

Inclusion and exclusion criteria

Randomization Procedure and Intervention

Enrolled patients were randomly assigned on a 1:1 basis by 1 of the coordinating investigators (G.N.G. or D.A.H.) to either receive tacrolimus (Prograf; Astellas Pharma, Leiden, the Netherlands) or belatacept (Nulojix; Bristol Myers-Squibb, New York City, NY). Randomization was performed by use of 40 sealed, opaque, sequentially numbered envelopes containing treatment allocation. The random allocation sequence was generated by an independent biostatistician by use of a random number generator. Before the start of the study, it was determined that 20 patients would be allocated to each treatment arm. Data were collected and monitored by the coordinating investigators in a hospital-based electronic study database.

Tacrolimus was dosed based on bodyweight (a dose of 0.2 mg/kg per day in 2 equally divided doses, rounded off to the nearest 0.5 mg) per the package insert ( Thereafter, the tacrolimus dose was adjusted based on whole-blood predose concentrations. The tacrolimus target predose concentrations were as follows: 10 to 15 ng/mL (weeks 1 and 2), 8 to 12 ng/mL (weeks 3 and 4), and 5 to 10 ng/mL from week 5 onward. Belatacept was dosed per the less-intensive regimen as described previously.6,7 Belatacept was administered intravenously in a dose of 10 mg/kg on the day of transplantation (day 0) and on days 4, 15, 30, 60, and 90 after transplantation. Thereafter, the dose was reduced to 5 mg/kg and given as monthly infusions up until month 12 after transplantation (end of study). Additional treatment is discussed in the SDC,


Refer to SDC, Material and Methods for data collection on (serious) adverse events.

Primary End Points

The overall aim of this trial was to determine the effect of belatacept- and tacrolimus-based immunosuppressive regimens on alloreactivity after kidney transplantation. The primary endpoint of the study presented here was the incidence of biopsy-proven AR (BPAR) within the first year after transplantation. BPAR rates were compared between belatacept- and tacrolimus-treated patients. We postulated that the incidence of BPAR would be higher among belatacept-treated patients7 and that BPAR biomarkers could be identified. All kidney transplant biopsies were obtained for cause, and no protocol biopsies were obtained. Refer to SDC, Materials and Methods for BPAR scoring system.

Pretransplant circulating frequencies of CD8+CD28, CD4+CD57+PD1, and EMRA CD8+CD28++ T cells, as well as their intracellular expression of a granzyme B (GrB) (an important cytotoxic protease during AR) were measured as immunological primary end points.19,23,24 These cell subsets were also measured posttransplantation, during AR before additional antirejection therapy was given or 3 months after transplantation in nonrejecting belatacept-treated patients. Free CD86 expression on circulating CD14+ monocytes were determined pretransplantation as a predictor for rejection; and before every dose of belatacept administered after transplantation as a pharmacodynamic drug monitoring tool. A monoclonal antibody competitive for belatacept was used (clone HA5.2B7, Beckman Coulter, Brea, CA). In patients who were rejected, the free CD86 expression was also assessed before additional antirejection therapy was given. Refer to SDC, and Materials and Methods for detailed information about our laboratory studies, including detection methods for DSA.

No formal statistical power calculation for the present study was performed, because (1) when the study was designed, it was unclear what the difference would be between belatacept- and tacrolimus-treated patients in terms of BPAR, as only data from the BENEFIT and BENEFIT-EXT, in which the comparator was ciclosporin, were available at the time6,7; (2) there were no published data available regarding the studied biomarkers and their association with BPAR that could serve for such a power calculation; and (3) because of financial constraints, we chose to conduct the present randomized controlled clinical trial with a limited number of patients in both arms.25 In our view, the present trial should therefore be regarded as a pilot study. It may serve as the basis for a larger study by providing the data needed to perform a statistical power calculation.

Statistical Analyses

Additional information is depicted in SDC, Materials and Methods. Percentages and counts are given for categorical variables, and medians plus ranges for continuous variables, unless otherwise specified. Continuous variables were compared between the belatacept and the tacrolimus group or between belatacept-treated rejectors and nonrejectors using the Mann-Whitney U test, and categorical variables using the Fisher exact test. Patient- and death-censored graft survivals, as well as death-censored BPAR-free survival, were compared between the belatacept and tacrolimus group using the log-rank test. All included patients were analyzed per the intention-to-treat principle.

To determine if high numbers of cytotoxic CD4+CD57+PD-1+, CD8+CD28, or CD8+CD28++ EMRA T cells, as well as CD86 molecules/monocyte were risk factors for BPAR, univariable Cox regression analyses were performed with death-censored BPAR-free survival as the dependent variable. Independent variables included the cell types after log transformation (to ensure approximately normal distribution of these variables), treatment arm, age, sex, ethnicity, HLA mismatches, HLA-DR mismatches, highest PRA, and cytomegalovirus (CMV) serostatus. Independent variables with a P less than 0.10 in the univariable analyses were intended to be included in a multivariable Cox regression analysis to predict BPAR.

Repeated measurements of CD86 occupancy on monocytes over time were compared between the study groups using a linear mixed model. To ensure a normal distribution of the model residuals, the dependent variable in the model was log transformed. Predictors were the values of CD86 molecules/monocyte pretransplantation, timepoint after transplantation (coded as categorical variable), treatment arm (belatacept or tacrolimus), and an interaction effect of time point and treatment arm to account for different trends over time between groups. The dependent variable was the value of CD86 molecules/monocyte after transplantation at a given timepoint. A random intercept was included in the linear mixed model to account for the within-subject correlations.

All tests were 2-tailed and statistical significance was defined as a P value less than 0.05. Bonferroni correction for multiple testing was applied when necessary.26 Statistical analyses were performed using IBM SPSS version 21 (SPSS Inc., Chicago, IL).



Between October 1, 2013 (first patient, first visit), and February 26, 2015 (last patient, first visit), 280 patients were screened, of whom 88 were eligible for participation (Figure 1). Forty-eight patients did not wish to participate. Major reasons were fear of AR and inconvenience of the monthly belatacept infusions. Forty patients were randomized and included in the intention-to-treat analysis. The baseline characteristics of these patients are described in Table 2. Seventeen (85%) patients in the belatacept and 19 (95%) in the tacrolimus group completed the 1-year follow-up period (last patient, last visit occurred on February 19, 2016).

Trial flowchart. All patients who were included in the study were randomized, underwent transplantation and received at least 1 dose of belatacept or tacrolimus. CDC, cytotoxicity-dependent crossmatch; CNI, calcineurin inhibitor; EBV, Epstein-Barr virus; HIV, human immunodeficiency virus; MGUS, monoclonal gammopathy of unknown significance.
Baseline characteristics at time of transplantation

Patient and Graft Survival

Patient survival was 95% in the tacrolimus group and 100% in the belatacept group (P = 0.32). One patient, randomized to the tacrolimus group, died 294 days after transplantation because of traumatic head injury. Three graft losses, all in the belatacept group, occurred on days 12, 59, and 161 after transplantation, resulting in a 1-year death-censored graft survival of 85% in the belatacept group versus 100% in the tacrolimus group (P = 0.08). All 3 graft losses were the result of glucocorticoid-resistant AR (Banff type IIB in 2 cases and type III in the third patient23).


In total, 29 for cause biopsies were performed in the belatacept group and 10 in the tacrolimus group in 14 and 6 patients, respectively (P = 0.015). The incidence of BPAR was higher among the belatacept-treated patients than in the tacrolimus-treated patients (n = 11 (55%) vs n = 2 (10%), respectively; P = 0.006; Table 3). The death-censored BPAR-free survival was significantly lower in the belatacept-treated patients than in the tacrolimus-treated patients (P = 0.002; Figure 2). Median time to rejection of patients who experienced AR was 56 (3-120) days in the belatacept group and 81 (10-152) days in the tacrolimus group. BPAR was of a more severe histological grade in the belatacept than in the tacrolimus group (P = 0.003; Table 3).

Incidence of rejection per the treatment group
BPAR-free survival. The time to first BPAR is depicted for the belatacept (dotted line) and the tacrolimus (solid line) group. In the tacrolimus group 1 patient died 294 days after transplantation due to traumatic head injury.

A detailed overview of the clinical course of the individual patients is depicted in Figure 3. In the belatacept group, n = 10 patients (50%) were treated for BPAR with pulse methylprednisolone therapy. Six patients (30%) received additional treatment with alemtuzumab, which is the preferred T cell depleting antibody in our center.27 In retrospect, and after revision by the second pathologist, 1 more patient in the belatacept group (case no. 13) was diagnosed as suffering from rejection but he was not treated with additional antirejection therapy. This patient had a so-called isolated v lesion, and despite not treating him, his graft function has remained excellent to the present day. After exclusion of this case, the BPAR rate was still significantly higher in the belatacept group than in the tacrolimus group. Nine patients (45%), all suffering from BPAR, were converted from belatacept to tacrolimus.


In the tacrolimus group, n = 2 patients were treated for BPAR: in 1 case with methylprednisolone pulse therapy only, in the other, additional treatment with alemtuzumab was given. Five patients (2 in the belatacept and 3 in the tacrolimus arm) received methylprednisolone for suspected rejection (for details, see Figure 3 legend).


In total, 205 AEs occurred in the belatacept group (mean, 10.3 per patient) and 238 in the tacrolimus group (mean 11.9 per patient); P = 0.41 (Table S1, SDC, Of these, 22 and 35, respectively, were judged to be serious (means per patient, 1.1 and 1.8, respectively; P = 0.15), excluding BPAR, graft loss, and death.

Estimated glomerular filtration rate, excluding graft losses, was not different between belatacept-treated and tacrolimus-treated patients 12 months after transplantation (Table S2, SDC, 54 (28-89) and 50 (33-84) mL/min per 1.73 m2, respectively; P = 0.57. Median protein/creatinine ratio was 13.2 (5.7-343.8) mg/mmol in the belatacept group and 9.0 (5.3-43.5) mg/mmol in the tacrolimus group (P = 0.44). Additional routine measurements are depicted in Table S2, SDC, (

For the on-therapy analysis on month 12; graft function before, during, and after BPAR in the belatacept group; the incidence of DSA and non-DSA; and pharmacokinetic drug monitoring, refer to Results and Tables S3-5, SDC (

Immunological Primary Endpoints (Biomarkers)

Three potential biomarkers for (belatacept-resistant) rejection were measured pretransplantation, namely, CD8+CD28 T cells, CD4+CD57+ PD1 T cells, and CD8+CD28++ EMRA T cells. There were no significant differences in the numbers or percentages of these cells at baseline between the tacrolimus and belatacept groups (Table 4). The limited number of patients experiencing BPAR in the tacrolimus group (n = 2) precluded a meaningful statistical comparison between rejectors and nonrejectors in this group. Gating strategies, pretransplant numbers, and percentages of the abovementioned cell subsets are depicted for future rejectors and nonrejectors in the belatacept group (see Table S6, SDC,; Figure 4), and no statistically significant differences were observed. Intracellular GrB expression was measured in the cell subsets (Figure 4A). Next, we analyzed whether high numbers or proportions of these cell types increased BPAR risk within the first 12 after transplantation by conducting univariable Cox regression analyses (Table 5):

Absolute numbers and percentages of T cell subsets pretransplantation
CD8+CD28, CD4+CD57+PD1, and CD8+CD28++ EMRA T cells pretransplantation. CD4+ and CD8+ T cells were gated from 7-AAD negative CD3+ lymphocytes (based on forward and sideward scatter) and EMRA T cells were gated as CCR7 and CD45RO T cells. Typical examples are given for nonrejectors and rejectors in the belatacept group for CD8+CD28, CD4+CD57+PD1, and CD8+CD28++ EMRA T cells and their intracellular GrB expressions (A). The absolute numbers and percentages of CD8+CD28, CD4+CD57+PD1, and CD8+CD28++ EMRA T cells are presented for nonrejectors and rejectors (B).
Univariable Cox regression analyses for the risk of BPAR
  • (1) CD8+CD28 T cells
  • CD8+CD28 T cells are mostly effector-memory cytotoxic T cells that produce large amounts of proinflammatory cytokines,15-17 and are not susceptible to costimulation blockade by belatacept. Almost 70% (31-89%) of CD8+CD28 T cells produced GrB. Higher numbers and proportions of pretransplant CD8+CD28 T cells (irrespective of their intra-cellular GrB expression) did not significantly increase BPAR risk in the first 12 months after transplantation (hazard ratio [HR], 1.06; 95% confidence interval [CI], 0.61-1.83 and HR, 1.05; 95% CI, 0.50-2.20, respectively; Table 5).
  • (2) CD4+CD57+PD1T cells
  • Next, pretransplant CD4+CD57+PD1 T cells were compared between rejecting and nonrejecting belatacept-treated patients. These cells were recently described as being cytolytic, CD28, and to be associated with belatacept-resistant rejection.18 The proportion of pretransplant CD4+CD57+PD1 T cells was low (<2% of the CD4+ T cell population in most patients). Approximately 24% (1-74%) of these cells were GrB positive. Neither the absolute number nor the proportion of these cell-predicted BPAR (HR, 0.89; 95% CI, 0.58-1.27, and HR, 0.90; 95% CI, 0.59-1.38, respectively; Table 5).
  • (3) CD8+CD28++EMRA T cells
  • Finally, CD8+CD28++ EMRA T cells were analyzed as high numbers of these cells predicted belatacept-resistant rejection in primates.19 It was postulated that these cells rapidly downregulate their surface CD28 expression after transplantation, making them resistant to costimulatory blockade.19 Circa 3% (0-3%) of these cells expressed intracellular GrB. The absolute numbers or proportions of pretransplant CD28++ cells within the CD8+ EMRA T cell population did not increase BPAR risk (HR, 0.86; 95% CI, 0.58-1.27; and HR, 1.23; 95% CI, 0.64-2.33, respectively; Table 5) Interestingly, from the 5 patients with greater than 35 CD8+CD28++ EMRA T cells/μL, 4 were rejectors and only 1 was a nonrejector (Figure 4B). In the tacrolimus group, the n = 2 rejectors had less than 10 CD8+CD28++ EMRA T cells/μL pretransplantation.

The abovementioned cell surface biomarkers were also measured in belatacept-treated patients during AR and before additional antirejection therapy was given and were compared with the month 3 samples from patients who remained rejection-free (Figure S1, SDC, No statistically significant differences were observed between rejecting and nonrejecting belatacept-treated patients.

The only significant risk factor for rejection in this study population was the use of a belatacept-based immunosuppressive regimen (HR, 7.2; 95% CI, 1.6-32.6; P = 0.01) compared with tacrolimus-based therapy (Table 5). Because no other variable significantly influenced AR risk and the sample size was small, no multivariable Cox regression analysis was conducted.

Pharmacodynamic Monitoring of Belatacept

The pharmacodynamic effect of belatacept was monitored by measuring free CD86 molecules on circulating monocytes. CD86 was saturated by belatacept at all time points, in both rejectors as nonrejectors. Moreover, pretransplantation CD86 molecules/monocyte were not predictive for BPAR (HR, 0.33; 95% CI, 0.1-2.2). For details about CD86-expression on monocytes in belatacept- and tacrolimus-treated patients, refer to Results and Figure S2, SDC,


In this RCT, a belatacept-based and a tacrolimus-based immunosuppressive regimen without lymphocyte-depleting induction therapy were compared head-to-head for the first time in de novo kidney transplantation. The results of this trial demonstrate that belatacept is not as potent as tacrolimus in preventing rejection.

In comparison to the 1-year results of the BENEFIT-trial where ciclosporin was used as comparator,7 we found a more pronounced difference in both BPAR incidence and severity. Ninety-one percent of BPAR in the belatacept group was classified as type II (or higher),28 whereas in the BENEFIT trial this was 69%. The use of lymphocyte-depleting therapy to treat rejection was comparable: circa 50% of BPAR in the BENEFIT-trial versus 55% in this study. The incidence of graft loss caused by BPAR was higher in this study than in the BENEFIT-trial: 3 of 11 versus 2 of 39 rejecting patients, respectively.

This larger difference in rate and severity of BPAR is not explained by dissimilarities between study groups. In the present study (1) there were no transplantations with deceased donors; (2) there were no patients with a PRA greater than 30%; and (3) the proportion of whites was larger. All 3 characteristics are associated with a lower BPAR risk.29-35 In contrast, the proportion of preemptive transplantations was high in our study (55% of included patients), which may have led to the inclusion of patients with a more potent immune system.36-38 Another explanation for the higher BPAR rate could be that in this study, TDM for MPA was performed, whereas this was not the case in the BENEFIT trial. It is therefore, theoretically possible that belatacept-treated patients in BENEFIT were exposed to higher MPA concentrations.1 However, we feel that this is an unlikely explanation as ciclosporin lowers exposure to MPA, whereas tacrolimus does not have such an effect.39

Our findings are in line with the higher BPAR rates observed in large retrospective studies and a small cohort study comparing belatacept with tacrolimus.13,14,18 Wen et al13 conducted a retrospective cohort study using registry data of a time period of 3 years, and compared 1-year clinical outcomes between belatacept- and tacrolimus-treated adult kidney transplant recipients. Although the incidence of BPAR was not as high as in the present trial, Wen et al also observed significantly higher BPAR rates among belatacept-treated patients as compared with tacrolimus-treated patients who would have been eligible for participation in the BENEFIT study: 15% of patients treated with belatacept and lymphocyte depleting antibody therapy versus 23% of patients treated with belatacept without lymphocyte depleting antibody therapy versus 6% of tacrolimus-treated patients. Nonetheless, it is important to stress that the higher incidence of BPAR in the present study should be interpreted with caution, because the study here included limited numbers of patients, had limited statistical power, and may therefore be a chance finding.

In this study, no suitable pretransplant biomarker was found to predict belatacept-resistant BPAR.15,16,18,19 The first potential biomarker, pretransplant CD8+CD28 T cell number, seemed a logical choice because these highly cytotoxic cells lack surface CD28 and are therefore not susceptible to belatacept.15-17 Possible explanations for the observation that these cells were not associated with BPAR may be that (1) even though these cells are highly cytotoxic, they lack proliferative capacity,40 and (2) the CD28-CD80/86 pathway is not the sole mediator of belatacept-resistant rejection. Targeting other costimulatory pathways, like CD40-40L, simultaneously with belatacept, might be more efficient to prevent BPAR.41,42 Preliminary data from Cortes-Cerisuelo et al43 suggest that not the lack of CD28 on these cells before transplantation, but the potential to downregulate CD28 after donor antigen stimulation is associated with BPAR in belatacept-treated patients.

The second biomarker, pretransplant CD4+CD57+PD1 T cell number, was associated to belatacept-resistant rejection in an observational cohort study.18 These findings were not confirmed here. Apart from differences in study design, the dissimilarities in study populations may explain this discrepancy.23-26 Our study population (1) was mostly white, (2) received mostly preemptive transplants, and (3) was shorter on dialysis. These factors have, however, not been associated with CD57 expression, and the proportions of CD4+CD57+PD1 T cells were similar pretransplantation. Age and CMV status, which influence these proportions,40,44-48 were also comparable (data not shown).

The final biomarker, CD8+CD28++ EMRA T cell number, showed potential to predict BPAR under belatacept, even though the group medians did not differ between rejectors and nonrejectors. One of the 9 nonrejectors and 4 of the 11 rejectors had high numbers of these cells pretransplantation (>35 cells/μL). In-depth analysis of the antigen specificity of these CD8+CD28++ EMRA T cells in larger studies seems warranted.49

Pharmacodynamic drug monitoring in the form of measuring free, non–belatacept-bound CD86 molecules on circulating monocytes was not useful to predict BPAR under belatacept therapy, because free molecules were not higher in rejectors pretransplantation, and followed the same dynamics in rejectors as in nonrejectors.

Limitations of this study are the small sample size and the resulting increased chance of type II errors. The increased rejection risk among belatacept-treated patients therefore needs to be confirmed in larger RCTs. Ideally, such trials will also include biomarker studies and analyze pretransplant donor-specific immunity. Also, research on regulatory T cells would be of interest because blockade of CD80/86 leads to anergic T cells,50 which consequently may fail to activate regulatory T cells via CD28. Studies on antigen-specific biomarkers, such as the IFNγ Elispot assay, would also be useful to study in larger, prospective trials.51,52

In conclusion, this small RCT showed that belatacept-based immunosuppressive therapy results in a significantly higher rejection rate and severity compared with standard, tacrolimus-based therapy. The biomarker data were not informative because there were no differences in pretransplant cellular biomarkers between rejectors and nonrejectors. Belatacept adequately blocked the CD28-CD80/86 costimulatory pathway in all patients, making insufficient saturation an unlikely explanation for this higher rejection risk.


The authors would like to thank Dr. M. Kho and Dr. J.I. Roodnat for their help in informing and including patients in the study; Mrs. J. Kal and Mrs. M. Laging for data collection and management; Ms. S.H. Brand for her help in interpreting the Luminex data; Mrs. M.J. Boer-Verschragen, Mrs. N.J. de Leeuw-van Weenen and Mrs. B. Nome for their help in managing the logistics of the trial and the blood withdrawals; and Dr. T. van Gelder for critically revising our article.


1. Vincenti F, Rostaing L, Grinyo J, et al. Belatacept and long-term outcomes in kidney transplantation. N Engl J Med. 2016;374:333–343.
2. Larsen CP, Pearson TC, Adams AB, et al. Rational development of LEA29Y (belatacept), a high-affinity variant of CTLA4-Ig with potent immunosuppressive properties. Am J Transplant. 2005;5:443–453.
3. Talawila N, Pengel LH. Does belatacept improve outcomes for kidney transplant recipients? A systematic review. Transpl Int. 2015;28:1251–1264.
4. Sharif A, Shabir S, Chand S, et al. Meta-analysis of calcineurin-inhibitor-sparing regimens in kidney transplantation. J Am Soc Nephrol. 2011;22:2107–2118.
5. de Graav GN, Bergan S, Baan CC, et al. Therapeutic drug monitoring of belatacept in kidney transplantation. Ther Drug Monit. 2015;37:560–567.
6. Durrbach A, Pestana JM, Pearson T, et al. A phase III study of belatacept versus cyclosporine in kidney transplants from extended criteria donors (BENEFIT-EXT study). Am J Transplant. 2010;10:547–557.
7. Vincenti F, Charpentier B, Vanrenterghem Y, et al. A phase III study of belatacept-based immunosuppression regimens versus cyclosporine in renal transplant recipients (BENEFIT study). Am J Transplant. 2010;10:535–546.
8. Vincenti F, Larsen CP, Alberu J, et al. Three-year outcomes from BENEFIT, a randomized, active-controlled, parallel-group study in adult kidney transplant recipients. Am J Transplant. 2012;12:210–217.
9. Webster AC, Woodroffe RC, Taylor RS, et al. Tacrolimus versus ciclosporin as primary immunosuppression for kidney transplant recipients: meta-analysis and meta-regression of randomised trial data. BMJ. 2005;331:810.
10. Van Gelder T, Hesselink DA. Belatacept: a game changer? Transplantation. 2016;100:1390–1392.
11. Klintmalm GB, Feng S, Lake JR, et al. Belatacept-based immunosuppression in de novo liver transplant recipients: 1-year experience from a phase II randomized study. Am J Transplant. 2014;14:1817–1827.
12. Ferguson R, Grinyó J, Vincenti F, et al. Immunosuppression with belatacept-based, corticosteroid-avoiding regimens in de novo kidney transplant recipients. Am J Transplant. 2011;11:66–76.
13. Wen X, Casey MJ, Santos AH, et al. Comparison of utilization and clinical outcomes for belatacept- and tacrolimus-based immunosuppression in renal transplant recipients. Am J Transplant. 2016;16:3202–3211.
14. Cohen JB, Eddinger KC, Forde KA, et al. Belatacept compared to tacrolimus for kidney transplantation: a propensity score matched cohort study. Transplantation. 2016 [published online December 8 2016]. doi: 10.1097/TP.0000000000001589.
15. Weaver TA, Charafeddine AH, Agarwal A, et al. Alefacept promotes co-stimulation blockade based allograft survival in nonhuman primates. Nat Med. 2009;15:746–749.
16. Lo DJ, Weaver TA, Stempora L, et al. Selective targeting of human alloresponsive CD8+ effector memory T cells based on CD2 expression. Am J Transplant. 2011;11:22–33.
17. Xu H, Perez SD, Cheeseman J, et al. The allo- and viral-specific immunosuppressive effect of belatacept, but not tacrolimus, attenuates with progressive T cell maturation. Am J Transplant. 2014;14:319–332.
18. Espinosa J, Herr F, Tharp G, et al. CD57(+) CD4 T cells underlie belatacept-resistant allograft rejection. Am J Transplant. 2016;16:1102–1112.
19. Mathews D, Wakwe W, Lowe M, et al. Acute rejection in non-human primates treated with belatacept is mediated by a unique CD8 memory subset. Am J Transplant. 2016;16.
20. Zhou Z, Shen J, Hong Y, et al. Time-varying belatacept exposure and its relationship to efficacy/safety responses in kidney-transplant recipients. Clin Pharmacol Ther. 2012;92:251–257.
21. Latek R, Fleener C, Lamian V, et al. Assessment of belatacept-mediated costimulation blockade through evaluation of CD80/86-receptor saturation. Transplantation. 2009;87:926–933.
22. Participants in the International Summit on Transplant T, Organ Trafficking Convened by the Transplantation S, International Society of Nephrology in Istanbul TAM. The Declaration of Istanbul on organ trafficking and transplant tourism. Transplantation. 2008;86:1013–1018.
23. de Graav GN, Hesselink DA, Dieterich M, et al. An acute cellular rejection with detrimental outcome occurring under belatacept-based immunosuppressive therapy: an immunological analysis. Transplantation. 2016;100:1111–1119.
24. Choy JC. Granzymes and perforin in solid organ transplant rejection. Cell Death Differ. 2010;17:567–576.
25. Abecassis M, Kaplan B. Transplantation: biomarkers in transplantation—the devil is in the detail. Nat Rev Nephrol. 2015;11:204–205.
26. Rice TK, Schork NJ, Rao DC. Methods for handling multiple testing. Adv Genet. 2008;60:293–308.
27. van den Hoogen MW, Hesselink DA, van Son WJ, et al. Treatment of steroid-resistant acute renal allograft rejection with alemtuzumab. Am J Transplant. 2013;13:192–196.
28. Haas M, Sis B, Racusen LC, et al. Banff 2013 meeting report: inclusion of c4d-negative antibody-mediated rejection and antibody-associated arterial lesions. Am J Transplant. 2014;14:272–283.
29. Ojo AO, Hanson JA, Meier-Kriesche H, et al. Survival in recipients of marginal cadaveric donor kidneys compared with other recipients and wait-listed transplant candidates. J Am Soc Nephrol. 2001;12:589–597.
30. Narayanan M, Pankewycz O, El-Ghoroury M, et al. Outcomes in African American kidney transplant patients receiving tacrolimus and mycophenolic acid immunosuppression. Transplantation. 2013;95:566–572.
31. Cosio FG, Dillon JJ, Falkenhain ME, et al. Racial differences in renal allograft survival: the role of systemic hypertension. Kidney Int. 1995;47:1136–1141.
32. Schweitzer EJ, Yoon S, Fink J, et al. Mycophenolate mofetil reduces the risk of acute rejection less in African-American than in Caucasian kidney recipients. Transplantation. 1998;65:242–248.
33. Nijboer WN, Schuurs TA, van der Hoeven JA, et al. Effect of brain death on gene expression and tissue activation in human donor kidneys. Transplantation. 2004;78:978–986.
34. Bleyer AJ, Burkart JM, Russell GB, et al. Dialysis modality and delayed graft function after cadaveric renal transplantation. J Am Soc Nephrol. 1999;10:154–159.
35. Huber L, Lachmann N, Niemann M, et al. Pretransplant virtual PRA and long-term outcomes of kidney transplant recipients. Transpl Int. 2015;28:710–719.
36. Betjes MG. Immune cell dysfunction and inflammation in end-stage renal disease. Nat Rev Nephrol. 2013;9:255–265.
37. Betjes MG, Langerak AW, van der Spek A, et al. Premature aging of circulating T cells in patients with end-stage renal disease. Kidney Int. 2011;80:208–217.
38. Litjens NH, Huisman M, van den Dorpel M, et al. Impaired immune responses and antigen-specific memory CD4+ T cells in hemodialysis patients. J Am Soc Nephrol. 2008;19:1483–1490.
39. Hesselink DA, van Hest RM, Mathot RA, et al. Cyclosporine interacts with mycophenolic acid by inhibiting the multidrug resistance-associated protein 2. Am J Transplant. 2005;5:987–994.
40. Mou D, Espinosa J, Lo DJ, et al. CD28 negative T cells: is their loss our gain? Am J Transplant. 2014;14:2460–2466.
41. Larsen CP, Elwood ET, Alexander DZ, et al. Long-term acceptance of skin and cardiac allografts after blocking CD40 and CD28 pathways. Nature. 1996;381:434–438.
42. Elwood ET, Larsen CP, Cho HR, et al. Prolonged acceptance of concordant and discordant xenografts with combined CD40 and CD28 pathway blockade. Transplantation. 1998;65:1422–1428.
43. Cortes M, Laurie S, Sayed B, et al. Rapid emergence of CD28null cells following ex vivo restimulation is associated with belatacept resistant rejection. Am J Transplant. 2016;16.
44. Vallejo AN. CD28 extinction in human T cells: altered functions and the program of T-cell senescence. Immunol Rev. 2005;205:158–169.
45. Pera A, Vasudev A, Tan C, et al. CMV induces expansion of highly polyfunctional CD4+ T cell subset coexpressing CD57 and CD154. J Leukoc Biol. 2017;101:555–566.
46. Pawelec G, Derhovanessian E. Role of CMV in immune senescence. Virus Res. 2011;157:175–179.
47. Wikby A, Maxson P, Olsson J, et al. Changes in CD8 and CD4 lymphocyte subsets, T cell proliferation responses and non-survival in the very old: the Swedish longitudinal OCTO-immune study. Mech Ageing Dev. 1998;102:187–198.
48. Olsson J, Wikby A, Johansson B, et al. Age-related change in peripheral blood T-lymphocyte subpopulations and cytomegalovirus infection in the very old: the Swedish longitudinal OCTO immune study. Mech Ageing Dev. 2000;121:187–201.
49. Bestard O, Cruzado JM, Lucia M, et al. Prospective assessment of antidonor cellular alloreactivity is a tool for guidance of immunosuppression in kidney transplantation. Kidney Int. 2013;84:1226–1236.
50. Schwartz RH. T cell anergy. Annu Rev Immunol. 2003;21:305–334.
51. Hricik DE, Augustine J, Nickerson P, et al. Interferon gamma ELISPOT testing as a risk-stratifying biomarker for kidney transplant injury: results from the CTOT-01 multicenter study. Am J Transplant. 2015;15:3166–3173.
52. Crespo E, Roedder S, Sigdel T, et al. Molecular and functional noninvasive immune monitoring in the ESCAPE study for prediction of subclinical renal allograft rejection. Transplantation. 2016 [published online June 29 2016]. doi: 10.1097/TP.0000000000001287.

Supplemental Digital Content

Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved.