Long-Term Exposure to Belatacept in Recipients of Extended Criteria Donor Kidneys



Patients in the BENEFIT-EXT study received extended criteria donor kidneys and a more intensive (MI) or less intensive (LI) belatacept immunosuppression regimen, or cyclosporine A (CsA). Patients who remained on assigned therapy through year 3 were eligible to enter a long-term extension (LTE) study. Three hundred four patients entered the LTE (n = 104 MI; n = 113 LI; n = 87 CsA), and 260 continued treatment through year 5 (n = 91 MI; n = 100 LI; n = 69 CsA). Twenty patients died during the LTE (n = 5 MI; n = 9 LI; n = 6 CsA), and eight experienced graft loss (n = 2 MI; n = 1 LI; n = 5 CsA). Three patients experienced an acute rejection episode (n = 2 MI; n = 1 LI). The incidence rate of serious adverse events, viral infections and fungal infections was similar across groups during the LTE. There were four cases of posttransplant lymphoproliferative disorder (PTLD) from the beginning of the LTE to year 5 (n = 3 LI; n = 1 CsA); two of three PTLD cases in the LI group were in patients who were seronegative for Epstein–Barr virus (EBV(−)) at transplantation. Mean ± SD calculated GFR at year 5 was 55.9 ± 17.5 (MI), 59.0 ± 29.1 (LI) and 44.6 ± 16.4 (CsA) mL/min/1.73 m2. Continued treatment with belatacept was associated with a consistent safety profile and sustained improvement in renal function versus CsA over time.


calculated glomerular filtration rate


calcineurin inhibitors


central nervous system


cyclosporine A


Epstein–Barr virus


extended criteria donor


high-density lipoprotein-cholesterol


low-density lipoprotein-cholesterol


less intensive


long-term extension


Modification of Diet in Renal Disease


more intensive


mycophenolate mofetil


posttransplant lymphoproliferative disorder


Kidney transplants utilizing extended criteria donor (ECD) kidneys are becoming more common due to the increased number of patients awaiting transplant and the limited availability of standard criteria donor organs [1]. While the utilization of ECD kidneys helps to meet the need in a growing population of potential recipients, ECD kidneys often include organs from older donors and donors with health issues that increase risk of graft loss or dysfunction [2]. In addition, ECD transplant recipients are challenging, as they tend to be older and are at increased risk for cardiovascular events and mortality compared with younger, non-ECD recipients [3-5].

While calcineurin inhibitors (CNIs) are the most widely used primary immunosuppressants in kidney transplant recipients, they are nephrotoxic and have deleterious effects on cardiovascular and metabolic parameters that may impact long-term graft and patient survival [6-9]. There is a need for immunosuppressive regimens that balance short-term risks against a need to preserve renal function; however, long-term data on CNI-free or CNI-avoiding regimens are scarce.

Belatacept, a selective costimulation blocker that prevents T cell activation, is designed to provide effective immunosuppression while avoiding both the renal and nonrenal toxicities associated with CNIs [10]. The BENEFIT-EXT study (Belatacept Evaluation of Nephroprotection and Efficacy as First-line Immunosuppression Trial-EXTended criteria donors) is the largest Phase III trial to date conducted in de novo recipients of ECD kidneys. Results from the 3-year partially blinded portion of the study found that belatacept-based immunosuppression enabled avoidance of CNIs, resulting in similar rates of patient and graft survival as well as acute rejection, but superior renal function and cardiovascular/metabolic profile [11-14].

The long-term extension (LTE) trial for BENEFIT-EXT commenced at the completion of 36 months of treatment. Patients who completed 36 months on therapy were eligible to enroll in the LTE. A new informed consent form was signed for participants entering the LTE. The primary objective of the LTE was to assess the long-term safety and tolerability of belatacept in the LTE patient cohort.


Study design

Details of the BENEFIT-EXT study design and primary results have been published [11-13, 15, 16]. Briefly, BENEFIT-EXT was a 3-year multicenter, randomized, partially blinded, parallel-group study in de novo adult recipients of ECD renal allografts who were randomly assigned (1:1:1) to receive the more intensive (MI) belatacept regimen, the less intensive (LI) belatacept regimen or cyclosporine A (CsA) (Figure 1). All patients received basiliximab induction (20 mg IV on Days 1 and 5), mycophenolate mofetil (MMF; 2 g/day in divided doses) and corticosteroids. The monthly dose of belatacept in the MI and LI treatment groups was identical after month 6. Patients were seen in clinic on a monthly basis. Most laboratory assessments were conducted every 3 months. Data were recorded at each site on standard paper Case Report Forms provided by the study sponsor, which were then entered into a database and were reviewed for quality assurance by data managers from the contract research organization and study sponsor staff.

Figure 1.

BENEFIT-EXT study treatment regimens.

ECD for this trial was defined as donors aged ≥60 years old; donors aged 50–59 years and who had at least two other risk factors, including cerebrovascular accident, hypertension or serum creatinine >1.5 mg/dL (UNOS definition); plus patients with anticipated cold ischemia time of ≥24 h; or donation after cardiac death.

The LTE population is a subset of the original intent-to-treat population; therefore, the LTE analysis does not reflect outcomes in patients who were discontinued from assigned therapy in the initial 36-month period. Patients in the LTE were required to remain on assigned therapy. Patients who discontinued assigned therapy in LTE were also discontinued from the study. Clinical outcomes (e.g. death and graft loss) following discontinuation of LTE were not collected beyond the 56 days of safety follow-up period after discontinuation.

The study was conducted in accordance with the ethical principles that have their origin in the current Declaration of Helsinki, and was consistent with International Conference on Harmonization Good Clinical Practice and applicable regulatory requirements. The study protocol and any amendments were reviewed and approved by the Institutional Review Board/Independent Ethics Committee for each site prior to initiation of the study. This trial is registered with ClinicalTrials.gov (id: NCT00114777).


For the LTE, the primary objective was to assess the safety and tolerability of belatacept with prolonged exposure. To that end, the frequency of adverse events, serious adverse events and deaths were tabulated, and the reasons for study discontinuation were recorded. Additionally, the incidence rates for adverse events were calculated based on the number of events per 100 patient-years of exposure.

The incidence, characteristics and treatment of clinically suspected, biopsy-proven acute rejection were assessed as previously described [10, 17]. The decision to treat acute rejection was based on a local biopsy reading, and the diagnosis and grade of acute rejection was confirmed by a central pathologist blinded to the patient assignments. Standardized blood pressure measurements, conducted in triplicate, were performed soon after the patient's arrival at the clinic for scheduled study visits, before venipuncture, study medication administration, and before any other scheduled study procedures, using specified procedures and standardized equipment. Additional endpoints included the mean change in serum lipids (non-high-density lipoprotein-cholesterol (non-HDL-C)); total-, low-density lipoprotein- and high-density lipoprotein-cholesterol (total-C, LDL-C and HDL-C); and triglycerides from baseline to year 5. Renal function was assessed by calculated GFR (cGFR), using the Modification of Diet in Renal Disease (MDRD) equation [18, 19].

Statistical methods

The purpose of this study was to assess the long-term safety and tolerability of belatacept in subjects who had received a kidney transplant. All subjects who completed the initial study (through month 36) and who consented to participate in the LTE were eligible to continue to participate in this study and receive study medication and follow-up. The sample size was not re-estimated for this LTE phase.

The mean change in cGFR from month 3 was analyzed by an analysis of covariance (ANCOVA) model with treatment as factor and the month 3 values as covariate to assess the difference between each of the belatacept treatment groups and the CsA group. The mean change from baseline in serum lipids was analyzed in a similar manner. p-Values were generated for these comparisons. Other outcomes were summarized descriptively. Missing values for outcomes were not imputed.


Patient disposition

Three hundred four patients entered the LTE (Figure 2), and 260 patients continued through year 5 (n = 91 MI; n = 100 LI; n = 69 CsA). Of the patients who were randomized and transplanted, 57–65% of patients in the belatacept groups entered the LTE, and 49–57% completed 5 years of treatment. Forty-seven percent of randomized and transplanted CsA patients entered the LTE, and 38% completed 5 years of treatment. The demographic characteristics of patients entering the LTE were similar to the baseline characteristics of the intent-to-treat population. The mean recipient age was 55 years. Most patients were male (67%) and white (72%), and 43% came from Europe. The median daily steroid dose during months 57–60 was 5 mg across all three dose groups, the median CsA trough level (local reading) was 97 ng/mL at month 60 and the median MMF daily dose of MMF was 1500–2000 mg during months 57–60. More patients treated with belatacept discontinued due to adverse events (n = 7 MI; n = 9 LI), while more patients treated with CsA withdrew consent (n = 4) or died (n = 5).

Figure 2.

Patient disposition. CsA, cyclosporine A; ITT, intent-to-treat; LI, less intensive; LTE, long-term extension; MI, more intensive.


The most common serious adverse events during the LTE are listed in Table 1, with the frequency and incidence rates of viral infections, central nervous system (CNS) infections, fungal infections, malignancies and posttransplant lymphoproliferative disorder (PTLD). As with previous reports, higher rates of fungal infections were observed in the belatacept groups, but most were nonserious and mucocutaneous in nature. Four cases of tuberculosis were reported prior to month 36 in patients who subsequently entered the LTE, and there was one case of pulmonary tuberculosis reported during the LTE in a patient from the belatacept LI group residing in an endemic area for tuberculosis. The infection resolved with treatment, and the patient remained on belatacept.

Table 1. Frequency and incidence rates of serious adverse events (≥3.5% of patients in any treatment group); viral, CNS and fungal infections; malignancies, nonmelanoma skin cancer and PTLD from the beginning of the LTE through year 5
n (%)Belatacept MI (N = 104)Belatacept LI (N = 113)Cyclosporine A (N = 87)
n (%)Incidence raten (%)Incidence raten (%)Incidence rate
  1. CNS, central nervous system; LI, less intensive; LTE, long-term extension; MI, more intensive; PTLD, posttransplant lymphoproliferative disorder.
  2. aExcluding nonmelanoma skin cancers. Serious adverse events listed by frequency in LI group.
Serious adverse events48 (46)34.165 (58)44.744 (51)38.6
Acute renal failure1 (1)0.56 (5)2.94 (5)2.5
Urinary tract infection5 (5)2.65 (4)2.45 (6)3.2
Diarrhea2 (2)1.05 (4)2.46 (7)3.8
Dehydration005 (4)2.41 (1)0.6
Pneumonia3 (3)1.64 (4)1.93 (3)1.9
Sepsis004 (4)1.94 (5)2.5
Atrial fibrillation004 (4)1.91 (1)0.6
Pyelonephritis2 (2)1.13 (3)1.46 (7)3.8
Viral infections22 (21)13.321 (19)11.221 (24)14.5
BK polyomavirus4 (4)2.11 (1)0.51 (1)0.6
Cytomegalovirus4 (4)2.14 (4)1.93 (3)1.9
Herpes (simplex, zoster, other)9 (9)4.99 (8)4.410 (12)6.6
CNS infections2 (2)1.01 (1)0.500
Fungal infections12 (12)6.619 (17)9.712 (14)7.9
All malignanciesa3 (3)1.66 (5)2.93 (3)1.9
Nonmelanoma skin cancer7 (7)3.82 (2)0.96 (7)3.8
PTLD0 3 1 

There were four cases of PTLD identified during the LTE (n = 3 LI; n = 1 CsA). Two of the three PTLD cases in the LI group occurred in patients who were seronegative at baseline for Epstein–Barr virus (EBV), and one case (in the EBV[+] patient) involved the CNS. Two of the three PTLD patients in the belatacept group died. The single case of PTLD in the CsA group during the LTE occurred in a patient who was seropositive for EBV at baseline. The case manifested as a bone marrow B cell lymphoma, and the patient died.

Death and graft loss

Among patients entering the LTE, the proportion of patients surviving with a functioning graft by year 5 was 93%, 91% and 88% in the MI, LI and CsA groups, respectively. Eight patients experienced death-censored graft loss (n = 2 MI; n = 1 LI; n = 5 CsA), and 20 died (n = 5 MI; n = 9 LI; n = 6 CsA; Table 2). Of the nine deaths in the LI group, six were related to malignancies, one to an infection (meningitis) and two to cardiovascular causes.

Table 2. Causes of death and graft loss during the LTE
Belatacept MI (n = 5)Belatacept LI (n = 9)CsA (n = 6)
1. Hemorrhagic shock1. Metastatic lung cancer1. Ischemic bowel
2. Respiratory arrest2. Pulmonary adenocarcinoma2. Pneumonia
3. Encephalomyelomeningitis3. Digestive lymphoma3. Septic shock
4. Esophageal cancer4. Lung cancer4. Refractory septic shock (pulmonary focus)
5. Worsening CHF5. Respiratory failure5. Heart attack
 6. Septic shock (secondary to bronchopneumonia)6. Heart failure
 7. Meningitis complication 
 8. Cardiac arrhythmia 
 9. Cardiovascular-related death 
Belatacept MI (n = 2)Belatacept LI (n = 1)CsA (n = 5)
  1. CAN, chronic allograft nephropathy; CHF, congestive heart failure; CsA, cyclosporine A; LI, less intensive; LTE, long-term extension; MI, more intensive.
Graft losses
1. Infection (n = 1)1. Noncompliance (n = 1)1. CAN (n = 2)
2. Acute rejection (n = 1) 2. Infection (n = 1)
  3. Other (n = 2)

Acute rejection

There were three cases of acute rejection reported during the LTE (n = 2 MI; n = 1 LI). The cases occurred on days 1176, 1471 (MI) and 1542 (LI). The acute rejection episode in the LI group was grade IB, while the cases in the MI group were grades IIA.

Renal function

The mean cGFR at year 5 was 55.9 mL/min/1.73 m2 (MI), 59.0 mL/min/1.73 m2 (LI) and 44.6 mL/min/1.73 m2 (CsA). The mean cGFR over time is depicted in Figure 3. The mean change in renal function from month 3 to year 5 was significantly different in the MI (+5.7 ± 2.29 mL/min/1.73 m2) and LI (+10.2 ± 2.121 mL/min/1.73 m2) groups versus the CsA group (−3.0 ± 2.487 mL/min/1.73 m2) (p = 0.0108 MI vs. CsA; p < 0.0001 LI vs. CsA).

Figure 3.

Mean calculated GFR over time. cGFR, calculated glomerular filtration rate; CsA, cyclosporine A; LI, less intensive; MI, more intensive.

Among LTE patients in the belatacept LI and CsA groups who had a cGFR of ≥30 mL/min and <60 mL/min at month 12, 85% (63/74) in the LI group and 76% (38/50) in the CsA group at least sustained their cGFR within that range through month 60. Thirty-one percent (23/74) of patients in the LI group and 10% (5/50) of patients in the CsA group improved their cGFR to ≥60 mL/min by month 60.

Blood pressure

Mean systolic blood pressure at year 5 was higher in the CsA group (141.8 ± 22.3 mmHg) than in the belatacept MI (133.7 ± 16.6 mmHg) or LI (132.0 ± 16.9 mmHg) group. Similarly, diastolic blood pressure was 78.3 ± 11.5 mmHg in the CsA group at year 5, compared with 73.9 ± 12.8 mmHg in the MI group and 75.3 ± 10.2 mmHg in the LI group.

Serum lipids

Mean changes from baseline to year 5 are depicted in Figure 4. Mean total, LDL- and non-HDL-C concentrations increased more in the CsA group versus the belatacept groups, but the difference was significant only in the LI group versus CsA for total- and non-HDL-C (p < 0.05). Serum HDL-C concentrations changed to a lesser degree and did not differ between groups. Serum triglyceride concentrations decreased in all groups, and the change from baseline was significantly greater for the MI group versus CsA (p < 0.05).

Figure 4.

Mean changes from baseline in serum lipids. n = 79 (MI), 89 (LI) and 62 (CsA) for total-, HDL- and non-HDL-cholesterol; n = 63 (MI), 74 (LI) and 52 (CsA) for LDL-cholesterol; n = 64 (MI), 74 (LI) and 52 (CsA) for triglycerides. CsA, cyclosporine A; HDL, high-density lipoprotein; LDL, low-density lipoprotein; LI, less intensive; MI, more intensive.


Long-term immunosuppression with a belatacept-based regimen was associated with better renal function compared to a CsA-based regimen, and exhibited a consistent safety profile in the LTE of the BENEFIT-EXT study. There were no new safety signals, and the greatest risk for PTLD in patients treated with a belatacept-based regimen remains with transplant recipients who were EBV(−) at baseline.

More patients in the belatacept groups (57–65%) entered the LTE, compared with 47% of patients in the CsA group. As of year 5, 88% of the belatacept-treated patients who entered the LTE were continuing therapy, compared to 79% of CsA patients. Approximately 90% of belatacept patients did not miss any infusion, 5% missed only one, 1.8% missed two and 1.8% missed three or more infusions. Taken together, the degree of retention in the belatacept groups in the BENEFIT-EXT LTE and on therapy suggests that utilization of intravenous therapy is not an impediment to persistence on therapy or to tolerability, at least when the medication is administered free of charge, as in the study setting.

ECD kidneys are associated with increased risk for primary nonfunction, delayed graft function and acute rejection [20, 21]. Furthermore, ECD kidneys have a lower GFR posttransplant than standard criteria donor kidneys, and a shorter half-life [22]. The renal function advantage associated with belatacept was maintained over time and continued to increase with respect to the CsA group, reaching ∼11–14 mL/min/1.73 m2 at year 5. cGFR in both the belatacept and CsA groups at year 5 was 13–15 mL/min/1.73 m2 greater than results observed in the intent-to-treat population at the end of year 3. This is likely due to survival bias for patients (and grafts) who completed the initial 36 months and were performing well on their assigned therapy, as patients were only eligible to enter LTE if they were receiving assigned therapy. Since a higher proportion of patients in the CsA group discontinued through 3 years and during the LTE compared to the belatacept groups, it is possible that the as-observed analysis of cGFR over time is biased in favor of CsA. Approximately 87% of patients in the LI group had a cGFR ≥30 mL/min at year 5 versus 69% of patients in the CsA group. The difference in renal function is important, as a GFR <30 mL/min is associated with increased chances of a return to dialysis, increased medical expenses, graft loss and death [23, 24].

The cardiovascular and metabolic risk profile at year 5 among patients entering the LTE was consistent with earlier results in the intent-to-treat population [14]. Hypertension is difficult to control in kidney transplant recipients, and is associated with increased risk for cardiovascular disease and diminished renal function [25-27]. At year 5 in LTE patients, both diastolic pressure (3–4 mmHg) and systolic pressure (8–9 mmHg) were lower in belatacept-treated patients compared to CsA-treated patients, likely due in part to CsA's vasoconstrictive effects [28, 29]. Similarly, the changes in serum lipids from baseline had a more atherogenic profile in the CsA group compared to the belatacept groups, which are consistent with known effects of CNI-based regimens [30-32]. Since cardiovascular disease is common in patients with chronic kidney disease and remains the leading cause of death with a functioning graft posttransplant, the less-atherogenic profile associated with the belatacept-based regimen may impact long-term outcomes [1, 33].

The most prominent risk factor for PTLD with this belatacept-based regimen is the recipient EBV serostatus pretransplant. An analysis of pooled data from a Phase II study and the two Phase III studies of belatacept found that the risk for PTLD, and specifically CNS PTLD, was higher in EBV(−) recipients treated with this belatacept-based regimen compared to EBV(+) recipients treated with either belatacept or CsA [34]. Of the three PTLD cases occurring in the BENEFIT-EXT LTE in the belatacept LI group, two occurred in EBV(−) patients out of a population of nine who entered the LTE. The one PTLD patient with CNS involvement died. The frequency of PTLD from year 3 to 5 in the belatacept-treated EBV(+) recipients was one case/217 patients who entered the LTE.

There are limitations to this analysis of the BENEFIT-EXT LTE data. The first is that the LTE patient population is inherently biased toward those patients doing well on assigned therapy. This certainly influences outcomes, and may minimize the frequency and impact of certain adverse events that occurred prior to the LTE. Patients who experienced serious events (e.g. PTLD, acute rejection episodes and other serious adverse events) were more likely to discontinue prior to year 3 of the study and not enter the LTE. None of the patients who had PTLD remained on assigned study therapy and, therefore, were not eligible for the LTE. Similarly, albeit to a lesser extent, many patients who experienced acute rejection discontinued assigned study therapy and were not eligible for the LTE. However, a few patients with prior acute rejection were eligible and entered the LTE (n = 10 MI; n = 13 LI; n = 12 CsA). Patients who discontinued during the LTE were discontinued from the study, which also biased results in a positive fashion. There were, for example, more patients in the CsA group (n = 18) who discontinued during the LTE compared to the belatacept groups (n = 13 each in MI and LI). However, the relative clinical profiles of each regimen appear to be consistent with earlier results, and this approach has the capacity to highlight safety signals that arise only with long-term drug exposure. Second, the ability of the BENEFIT-EXT study to detect infrequent events is limited by the sample size and duration of treatment. Since some adverse events occurred with low frequency, additional cases could change the nature of conclusions drawn from the data. However, this is the largest study of ECD kidney recipients, and the extent of exposure permits identification of certain risk factors, such as the heightened PTLD risk in EBV(−) recipients. Last, the use of CsA as the comparator arm carries some limitations. Tacrolimus is more often used than CsA at many transplant centers, and it is uncertain if the differences in GFR- or metabolic-related outcomes would be of the same magnitude if tacrolimus was used as the active comparator.

In summary, this study demonstrates that long-term treatment and exposure to a belatacept-based regimen is feasible, with a comparable proportion of patients surviving with a functioning graft and durable improvements in renal function versus CsA, and a consistent safety profile in ECD kidney transplant recipients. There were no unusual safety signals, and therapy was generally well tolerated in this population.


The BENEFIT-EXT study was supported by Bristol-Myers Squibb. The authors would like to acknowledge Brian Atkinson, PhD, Bristol-Myers Squibb, for writing and editorial assistance.


The authors of this manuscript have conflicts of interest to disclose as described by the American Journal of Transplantation: L. Pupim is an employee of Bristol-Myers Squibb. M. del C. Rial has received fees for expert testimony from Astellas, Bristol-Myers Squibb, Novartis, and Pfizer/Wyeth, and Roche, and has received payment for development of educational presentations from Novartis and Pfizer/Wyeth. S. Florman and J. O. Medina Pestana have received research grants from Bristol-Myers Squibb. Y. Vanrenterghem has been an advisory board member for Astellas. A. Matas has been an advisory board member for Bristol-Myers Squibb and has received research funding from Pfizer, Genentech, Astellas and the National Institutes of Health. B. Charpentier, L. Rostaing, F. Mühlbacher, R. Zhang and J. Grinyó have nothing to disclose.