Immunosuppression with Belatacept-Based, Corticosteroid-Avoiding Regimens in De Novo Kidney Transplant Recipients


Corresponding author: Ronald Ferguson,


Current immunosuppressive regimens in renal transplantation typically include calcineurin inhibitors (CNIs) and corticosteroids, both of which have toxicities that can impair recipient and allograft health. This 1-year, randomized, controlled, open-label, exploratory study assessed two belatacept-based regimens compared to a tacrolimus (TAC)-based, steroid-avoiding regimen. Recipients of living and deceased donor renal allografts were randomized 1:1:1 to receive belatacept-mycophenolate mofetil (MMF), belatacept-sirolimus (SRL), or TAC-MMF. All patients received induction with 4 doses of Thymoglobulin (6 mg/kg maximum) and an associated short course of corticosteroids. Eighty-nine patients were randomized and transplanted. Acute rejection occurred in 4, 1 and 1 patient in the belatacept-MMF, belatacept-SRL and TAC-MMF groups, respectively, by Month 6; most acute rejection occurred in the first 3 months. More than two-thirds of patients in the belatacept groups remained on CNI- and steroid-free regimens at 12 months and the calculated glomerular filtration rate was 8–10 mL/min higher with either belatacept regimen than with TAC-MMF. Overall safety was comparable between groups. In conclusion, primary immunosuppression with belatacept may enable the simultaneous avoidance of both CNIs and corticosteroids in recipients of living and deceased standard criteria donor kidneys, with acceptable rates of acute rejection and improved renal function relative to a TAC-based regimen.


calculated GFR


confidence interval




calcineurin inhibitor


Epstein-Barr virus


Modification of Diet in Renal Disease


mycophenolate mofetil


mammalian target of rapamycin


panel reactive antibody


posttransplant lymphoproliferative disorder


standard deviation






Calcineurin inhibitors (CNIs) and corticosteroids, often used as maintenance immunosuppression in kidney transplant patients, are associated with toxicities and side effects that may also contribute to increased morbidity and mortality. CNIs contribute to renal and nonrenal toxicities that limit allograft function, contribute to late graft failure and contribute to increased cardiovascular risk. Aside from their acute effects on afferent renal arterioles, the CNIs are also associated with chronic nephrotoxicity, leading to interstitial fibrosis and tubular atrophy and ultimately to late graft failure (1–5). In addition, CNIs can contribute to hypertension, dyslipidemia and new-onset diabetes (5–11). Similarly, corticosteroids are associated with adverse metabolic and lipid effects of particular concern in a population already at increased risk for cardiovascular mortality, including weight gain and diabetes (9,10,12–14). In addition, CNIs and corticosteroids are associated with other side effects, such as tremor, hirsutism, gingival hyperplasia and osteoporosis that may limit tolerability (15–17).

Some studies of regimens that either avoid either CNIs (7,18–21) or corticosteroids (22–28) have individually yielded acceptable efficacy and safety. However, regimens that avoid both classes of treatment have been much less successful (29,30). The challenge remains to provide effective immunosuppressive efficacy while optimizing the therapeutic regimen for kidney transplant recipients to avoid nephrotoxicity, nonrenal toxicities and immunosuppression-related side effects.

Belatacept, a selective costimulation blocker that prevents T cell activation, may serve as the basis for future immunosuppressive regimens that reduce reliance on CNIs and corticosteroids. Results from Phase II and III studies of belatacept in kidney transplantation suggest that belatacept-based regimens may provide effective immunosuppression, better allograft function and improved cardiovascular and metabolic risk profiles compared to cyclosporine-based regimens (31–33). Belatacept-based therapy was associated with higher grades of acute rejection in both Phase III studies (32,33) and with a higher incidence of acute rejection in one of these studies (33). Increased risk for posttransplant lymphoproliferative disorder (PTLD) involving the central nervous system was concentrated in patients who were negative for Epstein-Barr virus and who were taking a more-intensive regimen of belatacept (32,33).

The aim of the current Phase II exploratory trial was to determine whether use of belatacept-based maintenance regimens could allow avoidance of CNIs and corticosteroids in renal transplantation in comparison to a tacrolimus (TAC)-based, steroid-avoiding regimen. The study included two belatacept treatment groups, one in combination with sirolimus (SRL) and the other with adjunctive mycophenolate mofetil (MMF). A comparator group received a combination of TAC and MMF. Thymoglobulin, commonly used in steroid-avoiding regimens, was used as induction therapy in each treatment group.


Study design

This was an open-label, randomized, multicenter exploratory study that was conducted from July 2007 through May 2009 at 25 sites in the United States, Spain and Italy. The current report describes outcomes assessed at Month 12; patients were also eligible to enter a long-term study extension. The study was conducted in accordance with the ethical principles that have their origin in the Declaration of Helsinki and was also consistent with International Conference on Harmonization Good Clinical Practice and other regulatory requirements. The trial was sponsored by Bristol-Myers Squibb. Before beginning patient enrollment, the study was registered with (id NCT00455013) and institutional review committees at each center approved the study protocol and patient consent forms. While the study was in progress, a data and safety monitoring committee periodically reviewed reports of safety and efficacy.

Patient population

Adult kidney transplant recipients who had received a renal allograft from a nonhuman leukocyte antigen (HLA)-identical living or standard criteria deceased donor were eligible to enter the study. Principal exclusion criteria included recipients of extended criteria donor kidneys according to an expanded UNOS definition, that is, donors ≥60 years old; donors aged 50–59 years with any combination of at least 2 of the following conditions: death from cerebrovascular accident, hypertension, serum creatinine >1.5 mg/dL; anticipated cold ischemia time ≥24 h, or donation after cardiac death; any prior or concurrent nonrenal solid organ transplant; high immunological risk features including current (pretransplant) panel-reactive antibodies (PRA) ≥50% (>30% if retransplant) or with any prior graft loss due to acute rejection; underlying renal disease that could recur in the allograft; infection with human immunodeficiency virus (HIV), hepatitis B or hepatitis C virus, or latent tuberculosis; history of malignancy in the previous 5 years (other than nonmelanoma skin cancer cured by resection). Patients seronegative for Epstein Barr virus (EBV) or with a body mass index >35 kg/m2 were also excluded. Female patients of childbearing potential were required to use an adequate method of contraception during the study. All patients provided signed informed consent.

Immunosuppression and concomitant therapy

The three treatment regimens compared were belatacept with MMF (CellCept®, Roche Pharmaceuticals, Nutley, NJ, USA), belatacept with SRL (Rapamune®, Wyeth, Madison, NJ, USA) and TAC (Prograf®, Astellas Pharma, Tokyo, Japan) with MMF. Immediately before transplantation, patients were randomized 1:1:1 to treatment groups using an interactive voice response system with centralized randomization.

Thymoglobulin® (Genzyme, Cambridge, MA, USA) was administered to all 3 treatment groups as a 1.5-mg/kg IV infusion beginning on the day of transplantation (Day 1) and again on Days 2, 3 and 4. Dosing could be stopped earlier if absolute lymphocyte count was <100/mm2, or reduced or extended if a patient developed neutropenia or thrombocytopenia. However, dosing was not to extend past Day 10 or cumulatively exceed 6 mg/kg. All patients received 500, 250, 125 and 60 mg IV methylprednisolone on Days 1, 2, 3 and 4, respectively. For Thymoglobulin infusions on Days 5–10, if given, patients were premedicated with 30 mg IV methylprednisolone.

Belatacept was administered 10 mg/kg IV on Days 1 and 5, then once every 2 weeks through Month 3, every 4 weeks through Month 6 and 5 mg/kg every 4 weeks from Month 7 onward. The dose was chosen based on the more intensive belatacept regimen used in ongoing phase III studies; this selection was made while the Phase III studies were ongoing and the more intensive regimen was selected due to the absence of steroids in the regimen (32,33). For patients who discontinued belatacept or TAC, the choice of switch therapy was determined by the investigator.

SRL was initiated at 5 mg/day orally on Day 1 and adjusted to keep predose (C0) levels at 7–12 ng/mL from Day 3 through Month 6 and 5–10 ng/mL thereafter. Dose reduction was permitted if adverse events occurred attributable to SRL. If still not tolerated, patients could be switched to MMF.

The dosage of MMF was 1 g twice daily, which could be reduced and/or split into 4 divided doses. If MMF was still not tolerated, patients could be switched to mycophenolic acid delayed release tablets (Myfortic®) and if that was not tolerated, switched to SRL.

TAC was initiated when serum creatinine had improved to ≤4 mg/dL in the absence of dialysis. TAC was given orally as 0.1 mg/kg/day in two divided doses and adjusted to keep predose (C0) levels at 8–12 ng/mL through Day 30 and 5–10 ng/mL thereafter.

Acute rejection episodes of Banff 97 grade IIA or lower were treated with methylprednisolone. Steroid-resistant acute rejection episodes and those of Banff grade IIB or higher were recommended to be treated with lymphocyte-depleting therapy.

All patients received cytomegalovirus (CMV) prophylaxis (ganciclovir or valganciclovir) for at least 3 months after transplantation and Pneumocystis jiroveci prophylaxis for at least 6 months. Both prophylaxis regimens were repeated if lymphocyte-depleting therapy was administered to treat acute rejection.

Study endpoints

The primary efficacy endpoint was the incidence of acute rejection by Month 6. Acute rejection was defined as (1) biopsy-proven and either (2) clinically suspected for protocol-defined reasons or (3) clinically suspected for other reasons and treated. Protocol-defined reasons for obtaining a biopsy for suspected acute rejection included unexplained rise in serum creatinine ≥25% from baseline and one or more of the following: any unexplained decrease in urine output, fever and graft tenderness, or a persistent elevation in serum creatinine in the 14 days after transplantation with clinical suspicion of rejection. The biopsy was read by a clinical pathologist blinded to group allocation and graded according to the Banff 97 criteria (34).

Secondary endpoints included the incidence and severity of acute rejection; patient survival; graft survival and calculated glomerular filtration rate (cGFR) using the Modification of Diet in Renal Disease (MDRD) formula (35). Other secondary endpoints included steroid-free status and CNI-free status at Month 12 and cardiovascular and metabolic profiles including blood pressure, serum lipids and new onset diabetes.

The presence of anti-donor HLA antibodies was assessed before transplantation, at months 6 and 12 and after any suspected acute rejection episode. Assessment of anti-donor antibodies was performed at Emory University, Atlanta, GA (36–38). Whole blood samples from a subset of 71 patients (n = 28 belatacept-MMF; n = 22 belatacept-SRL; n = 21 TAC-MMF) for lymphocyte immunophenotyping assays were obtained on Day 5 and Weeks 2, 4, 12, 24 and 52. Lymphocyte immunophenotyping was performed with the validated MultiTEST™ IMK assay, the processed samples acquired on a FACSCalibur™ flow cytometer and the data analyzed with Multiset™ v 2.2 analysis software (BD Biosciences, San Jose, CA, USA; at baseline, n = 28 belatacept-MMF; n = 22 belatacept-SRL; n = 21 TAC-MMF). The frequencies of memory and naive cells among CD4 and CD8 T lymphocytes, as well as the frequency of regulatory cells among CD4 T lymphocytes were assessed by flow cytometric assays. The memory and naive T cells were identified by the phenotypes of CD45RO+CD45RA and CD45ROCD45RA+, respectively (at baseline, n = 28 belatacept-MMF; n = 22 belatacept-SRL; n = 23 TAC-MMF). The regulatory T cells were identified by the surface expression of CD3+CD4+CD25hi and intracellular expression of FoxP3 (at baseline, n = 30 belatacept-MMF; n = 24 belatacept-SRL; n = 24 TAC-MMF).

Safety assessments included adverse events as reported by the investigators, laboratory abnormalities and clinically significant changes in vital signs and physical examination findings. Through 12 months following randomization, all patients were monitored for graft and vital status and key safety events, regardless of whether study treatment had been discontinued.

Statistical analysis

All safety and efficacy analyses were conducted according to the intention to treat with the use of data from all patients who underwent randomization and transplantation. No statistical testing was prespecified in the study and only descriptive summaries are provided. The number of patients per group was planned so that the upper bound of a 95% confidence interval (CI) would exclude a clinically unacceptable acute rejection rate of 30%. Rates of acute rejection at 6 and 12 months and patient and graft survival at 12 months were summarized along with 95% CIs. For patient and graft survival by Month 12, two-sided 95% CIs were also generated for the difference between each of the belatacept regimens and TAC.


Patient population

Ninety-three patients were randomized and 89 received a transplant and immunosuppressive therapy (Figure 1). Baseline demographic and clinical characteristics of transplant recipients were similar across treatment groups as shown in Table 1. During the 12 months of the study, discontinuation or switching of regimen occurred in 8 (24%), 12 (46%) and 2 (7%) of the patients in the belatacept-MMF, belatacept-SRL and TAC-MMF groups, respectively. In the belatacept-MMF group, 6 patients discontinued belatacept and 2 patients switched from MMF to SRL. Of the 12 patients who discontinued or switched treatment in the belatacept-SRL group, most (n = 10) switched from SRL to MMF and 5 patients (including 3 who also switched from SRL) discontinued belatacept. Of the 11 patients who discontinued belatacept, 6 were switched to TAC. Two patients in the TAC-MMF group discontinued TAC. Most patients across groups who discontinued or switched did so within the first 60 to 90 days. The most common reasons for discontinuation of belatacept were lack of efficacy (i.e. acute rejection) and adverse events (Figure 1). Mean trough levels of TAC and SRL at Week 2 were 10.6 ng/mL and 17.7 ng/mL, respectively, and were 9.6 ng/mL and 9.0 ng/mL, respectively, at Month 12.

Figure 1.

Patient disposition. (a) Includes 2 patients who discontinued MMF and switched to SRL. (b) includes 7 patients who discontinued SRL and switched to MMF.

Table 1.  Baseline characteristics of transplant recipients
 Belatacept-MMF (n = 33)Belatacept-SRL (n = 26)TAC-MMF (n = 30)
  1. Due to rounding percentages may not total 100%. PRA, panel reactive antibody; MMF, mycophenolate mofetil; SRL, sirolimus; TAC, tacrolimus; SD, standard deviation.

Mean age, years (SD)49.2 (11.1)52.7 (10.8)53.6 (13.2)
Gender, n (%)
 Male25 (76)20 (77)22 (73)
 Female 8 (24) 6 (23) 8 (27)
Race, n (%)
 White24 (73)23 (89)23 (77)
 Black or African–American 8 (24) 3 (12) 5 (17)
 Other1 (3)02 (7)
Geographic region, n (%)
 North America22 (67)16 (62)20 (67)
 Europe11 (33)10 (39)10 (33)
Previous number of transplants, n (%)
 032 (97)26 (100) 30 (100)
 11 (3)00
Reported cause of end-stage renal disease, n (%)
 Glomerulonephritis 7 (21) 5 (19) 4 (13)
 Diabetes 6 (18) 4 (15) 8 (27)
 Polycystic kidneys 4 (12) 4 (15) 5 (17)
 Hypertensive nephrosclerosis 5 (15)2 (8) 3 (10)
 Renovascular and other1 (3)2 (8)2 (7)
 Congenital, familial and metabolic1 (3)00
 Tubular and interstitial diseases3 (9)2 (8)1 (3)
 Other 6 (18) 7 (27) 7 (23)
Categorized PRA, n (%)
 <20% 33 (100)24 (92) 30 (100)
 ≥20%01 (4)0
 Missing data01 (4)0
Type of transplant, n (%)
 Living-related 7 (21) 9 (35) 6 (20)
 Living-unrelated 9 (27) 6 (23) 7 (23)
 Deceased17 (52)11 (42)17 (57)
Mean cold ischemia time for living donor transplant, h (SD)0.9 (0.7)0.7 (0.8)0.4 (0.3)
Mean cold ischemia time for deceased donor transplant, h (SD)16.5 (4.3) 15.7 (6.0) 14.8 (5.0) 

Acute rejection

Four patients in the belatacept-MMF group (12%), 1 in the belatacept-SRL group (4%) and 1 in the TAC-MMF group (3%) had acute rejection by Month 6 (Table 2). One additional patient in the belatacept-MMF group had an episode of acute rejection between Months 6 and 12. All cases of acute rejection were Banff 97 grade IIA or IIB. Rates of biopsy-proven acute rejection were identical to the rates of protocol-defined acute rejection. Six patients tested positive for anti-donor HLA antibodies at any time during the study; 2 experienced an acute rejection episode. Both patients were in the belatacept-MMF group, were not seropositive at baseline and remained seropositive through Month 12.

Table 2.  Summary of key efficacy outcomes at Month 12
 Belatacept-MMF (n = 33)Belatacept-SRL (n = 26)TAC-MMF (n = 30)
  1. Percentages may not total 100% due to rounding. CNI, calcineurin inhibitor; MMF, mycophenolate mofetil; SRL, sirolimus; TAC, tacrolimus; SD, standard deviation.

  2. 1Same patient.

  3. 2Defined as not receiving steroids for >7 consecutive days during Days 337 through 392.

  4. 3Defined as not receiving a CNI during Days 337 through 392.

Acute Rejection at Month 6, n (%)4 (12)1 (4)1 (3)
Difference from TAC (95% CI)8.8 (−6.6, 24.9)0.5 (−14.5, 16.7)
Banff Grade, n (%)
 Mild acute (IA or IB)000
 Moderate acute (IIA)2 (6)01 (3)
 Moderate acute (IIB)2 (6)1 (4)0
 Severe acute (III)000
Acute rejection at Month 12, n (%)5 (15)1 (4)1 (3)
Difference from TAC (95% CI)11.8 (−4.1, 28.7)0.5 (−14.5, 16.7)
Subject and graft survival at Month 12, n (%)30 (91)24 (92)30 (100)
Difference from TAC (95% CI)−9.1 (−23.6, 2.8)−7.7 (−24.1, 4.1)
 Graft loss2 (6)2 (8)0
 Death1 (3)100
 Death with functioning graft1 (3)100
Mean (SD) calculated GFR at Month 12, mL/min/1.73m263.6 (27.27)61.8 (30.66)54.0 (14.95)
Proportion steroid-free at Month 12, n (%)224 (73)20 (77)28 (93)
Proportion steroid-free and 3CNI-free at Month 12, n (%)224 (73)18 (69)1 (3)

Treatment for acute rejection was given based on local biopsy results and was consistent with protocol guidance. Four patients (n = 3 belatacept-MMF; n = 1 belatacept-SRL) with acute rejection were treated successfully with corticosteroids and one (TAC-MMF) with steroid-resistant acute rejection was treated with lymphocyte-depleting therapy. Two others (both belatacept-MMF) were initially treated with lymphocyte-depleting therapy.

One patient in the belatacept-MMF group had more than one episode of acute rejection and this contributed to graft loss in that patient. All other patients with acute rejection were alive with functioning grafts at Month 12. These included 3 who withdrew and switched from belatacept-MMF to TAC-MMF and 1 each in the belatacept-MMF, belatacept-SRL and TAC-MMF groups who remained in the study on their assigned treatments.

Patient and graft survival

At Month 12, 30/33 (91%) of the belatacept-MMF group, 24/26 (92%) of the belatacept-SRL group and 30/30 (100%) of the TAC-MMF group were alive with functioning grafts. One patient (belatacept-MMF) died of complications of pneumonia on Day 46. Four others lost grafts including 2 from graft thrombosis (belatacept-MMF and belatacept-SRL on Days 2 and 5, respectively), 1 on Day 185 from BK virus nephropathy (belatacept-SRL) and 1 from a second episode of acute rejection after discontinuation from the study (belatacept-MMF).

Renal function

Renal function was better with either belatacept regimen than with TAC-MMF and this difference persisted for the 12 months of treatment (Figure 2). At Month 12, mean cGFR was 8–10 mL/min/1.73 m2 higher in the belatacept groups than in the TAC-MMF group. In a sensitivity analysis, where missing cGFR values were imputed to 10 mL/min/1.73 m2 after death or graft loss, the mean cGFR was 4–6 mL/min/1.73 m2 higher in the belatacept groups. Ten patients had delayed graft function (n = 5 belatacept MMF; n = 3 belatacept-SRL; n = 2 TAC-MMF); only four (n = 2 belatacept-MMF; n = 2 belatacept-SRL) required >1 day of dialysis.

Figure 2.

Mean calculated glomerular filtration rate (cGFR) in patients receiving belatacept-based regimens and TAC-MMF at indicated time points posttransplantation.

CNI-free and steroid-free patients

Seventy-three percent of the belatacept-MMF group and 69% of the belatacept-SRL group remained both CNI- and steroid-free at Month 12. One patient in the TAC-MMF group was both CNI-free and steroid-free at Month 12, having discontinued TAC on Day 2 posttransplant. The percentages of patients remaining steroid-free at Month 12 were 73%, 77% and 93%, respectively, in the belatacept-MMF, belatacept-SRL and TAC-MMF groups. Most patients were put on steroids due to adverse events or acute rejection.

Cardiovascular and metabolic risk factors

No clinically significant changes in blood pressure occurred (Table 3). The majority of subjects were taking one or more antihypertensive medications at baseline and Month 12. Triglycerides decreased from baseline in all three treatment groups. No clinically relevant differences in lipids or antihyperlipidemic medication use were noted between groups. New onset diabetes occurred in 3 patients (n = 2 belatacept-SRL and n = 1 TAC-MMF).

Table 3.  Summary of blood pressure and metabolic changes after 12 months
 Belatacept-MMF (n = 33)Belatacept-SRL (n = 26)TAC-MMF (n = 30)
  1. NODAT, new onset diabetes after transplantation; defined as patient with no history of diabetes with fasting plasma glucose ≥126 mg/dL (7.0 mmol/L) on 2 or more occasions or receives antidiabetic medication for at least 30 days.

Mean blood pressure (SD), mmHg
 Baseline systolic blood pressure133.1 (26.43)126.9 (18.72)141.8 (23.79)
 Month 12 systolic blood pressure129.3 (19.24)131.0 (19.88)138.3 (19.50)
 Baseline diastolic blood pressure 78.6 (12.50) 72.3 (11.27) 75.3 (15.08)
 Month 12 diastolic blood pressure 73.3 (11.96) 75.1 (10.71) 77.6 (10.51)
Antihypertensive medications taken at Month 12, n (%)
 1–2 17 (51.5) 13 (50.0) 17 (56.7)
 ≥ 3  8 (24.2)  7 (26.9)  3 (10.0)
Mean change in lipid values from baseline to Month 12 (SD), mg/dL
 HDL cholesterol 1.5 (12.7)−3.7 (12.9)−0.5 (12.2)
 Non-HDL cholesterol16.0 (45.3)16.1 (45.2)20.5 (42.6)
 LDL cholesterol23.9 (38.2)25.0 (37.2)34.0 (30.2)
 Total cholesterol17.5 (40.7)12.5 (49.0)20.0 (45.8)
 Triglycerides−11.4 (63.3) −1.1 (88.9)−14.2 (94.8) 
Use of ≥1 antihyperlipidemic medication at Month 12, n (%) 12 (36.4) 10 (38.5) 12 (40.0)
Patients without pretransplant diabetes, n212117
Incidence of NODAT, n (%)0 2 (9.5) 1 (5.9)


The most common adverse events by Month 12, which included anemia, pyrexia, leukopenia, diarrhea and constipation, occurred with a similar incidence across groups (data not shown). Aphthous stomatitis (n = 4 belatacept-MMF; n = 8 belatacept-SRL) and proteinuria (n = 4 belatacept-MMF; n = 3 belatacept-SRL) were observed in the belatacept groups but not in the TAC-MMF group. The mean urine protein/creatinine ratios were <1 across groups at Month 12: 0.3, 0.5, and 0.1 in the belatacept-MMF, belatacept-SRL and TAC-MMF groups, respectively. Tremor (n = 1 belatacept-MMF; n = 7 TAC-MMF) and wound dehiscence (n = 3 TAC-MMF) were observed more often in the TAC-MMF group. There were no reports of hypersensitivity or anaphylaxis related to infusion of belatacept in any patient. Serious adverse events occurred in more than half of the patients in the study (Table 4). The most common were hydronephrosis and pyrexia. Only five serious adverse events led to treatment discontinuation (belatacept-MMF: increased serum creatinine, graft loss and pyrexia; belatacept-SRL: renal artery thrombosis, proteinuria).

Table 4.  Most common serious adverse events (≥2%) and other events of clinical interest by Month 12
Event, n1 (%)Belatacept-MMF (n = 33)Belatacept-SRL (n = 26)TAC-MMF (n = 30)
  1. 1Total patients with an event.

  2. 2Reported in ≥2% of total patients.

  3. 3Pyelonephritis acute in 3 patients; all other serious infections 1 patient each.

Any serious adverse event19 (58)16 (62)16 (53)
Most frequent serious adverse events2
 Hydronephrosis1 (3)2 (8)1 (3)
 Pyrexia2 (6)1 (4)1 (3)
 Dehydration01 (4)2 (7)
 Pyelonephritis acute2 (6)1 (4)0
 Graft dysfunction01 (4)2 (7)
 Diarrhea02 (8)1 (3)
 Blood creatinine increased2 (6)01 (3)
 Renal tubular necrosis1 (3)01 (3)
 Urinary fistula01 (4)1 (3)
 Graft loss1 (3)1 (4)0
 Postprocedural hemorrhage1 (3)1 (4)0
 Nausea1 (3)01 (3)
 Hyponatremia01 (4)1 (3)
 Deep vein thrombosis1 (3)01 (3)
 Lymphocele1 (3)1 (4)0
 Neutropenia2 (6)00
 Pulmonary embolism1 (3)01 (3)
Events of clinical interest
 Any infection26 (79)20 (77)20 (67)
 Any serious infection3 7 (21) 4 (15) 5 (17)
 Fungal infection 5 (15)1 (4)2 (7)
 Viral infection 4 (12)2 (8) 6 (20)
 Any malignancy01 (4)1 (3)

Viral infections occurred in 12 patients through Month 12 (n = 4 belatacept-MMF; n = 2 belatacept-SRL; n = 6 TAC-MMF), although most were not serious and no viral infections resulted in discontinuation from the study. BK virus (n = 2; 1 each in belatacept-MMF and TAC-MMF) and CMV infection (n = 4 total; n = 1 each belatacept-MMF and belatacept-SRL, n = 2 TAC-MMF) occurred infrequently across groups. Fungal infections occurred more often in the belatacept-MMF group (n = 5), but were generally mild skin infections that responded to treatment, with the exception of one serious event of cryptococcal meningitis in the belatacept-MMF group that resolved with treatment and did not lead to treatment discontinuation. Malignancies were recorded in 2 patients: 1 in the belatacept-SRL group (basal cell carcinoma) and 1 in the TAC-MMF group (squamous cell carcinoma). There were no cases of PTLD or progressive multifocal leukoencephalopathy.

Peripheral blood lymphocyte profile

Peripheral whole blood CD4+ and CD8+ T cells as well as NK cells showed significant declines following lymphodepletion with Thymoglobulin in all treatment groups. These lymphocyte populations recovered over the 12 months of the study to the same extent regardless of treatment group (Figure 3). In contrast, B cell numbers were not dramatically impacted by Thymoglobulin treatment (data not shown). In addition to measuring the decrease in the total number of T cells following lymphodepletion, the proportion of regulatory T cells remaining in the peripheral blood was also assessed at baseline and on Days 84, 168 and 364 in 21–30 patients from each group and was found to be similar and constant over time in all treatment groups (approximately 3–5% of the CD4+ T cells; Figure 3). Similarly, the percentage of memory CD4+ T cells in the peripheral blood was also assessed and was also found to be similar across all treatment groups (Figure 3). Similar results were obtained for CD8+ memory T cells (data not shown).

Figure 3.

Changes in the number of peripheral blood CD4+, CD8+, memory and regulatory T cells over time following transplantation and Thymoglobulin administration.


This exploratory, Phase 2, open-label, randomized study is the first to assess the safety and efficacy of belatacept as a component of an immunosuppressive regimen that avoids both CNIs and corticosteroids. The belatacept-based regimens were associated with low rates of acute rejection and there was improved allograft function with the belatacept-based regimens compared to a commonly used TAC-based regimen with similar induction therapy. In addition, the study provides the first belatacept clinical study experience with Thymoglobulin induction and combination with a mammalian target of rapamycin (mTOR) inhibitor.

The 6-month acute rejection rate was 12% in the belatacept-MMF group, 4% in the belatacept-SRL group and 3% in the TAC-MMF group. The 12% acute rejection rate in the belatacept-MMF group is comparable to or better than the acute rejection rates reported in other studies with regimens that avoid either CNIs or corticosteroids and better than those that avoid both. Vincenti et al. reported rates at 6 months of 48% in a CNI-avoiding regimen of daclizumab/MMF/steroids (21). In the CARMEN study, the 6-month rates were 16.5% with a steroid-avoiding regimen of daclizumab/TAC/MMF (23). In the FREEDOM study, where patients received cyclosporine, mycophenolate sodium and basiliximab, the acute rejection rate was 36% in a corticosteroid-free group and 29.6% in a group in which corticosteroids were withdrawn at Day 7 (22). A 5-year study of early corticosteroid cessation noted an acute rejection rate of 14–24%, depending on the induction agent used (28). Flechner et al. reported a pilot study using a CNI-free and steroid-avoiding regimen of alemtuzumab/MMF/SRL where the 12-month acute rejection rate was 36% (29). Ours appears to be the first regimen to show acceptable immunosuppression while avoiding CNIs from the time of transplant and eliminating corticosteroids after 5 days.

The current study also showed evidence of improved renal function with belatacept compared with TAC. Given the associations between allograft function and graft loss (39) and allograft function and cardiovascular events (40–42), the data suggest the potential for a belatacept-based regimen to be associated with improved long-term patient and graft survival. The degree of improvement in renal function warrants confirmation in a larger, controlled study.

Overall, there was a higher rate of withdrawal and switching from the SRL-containing regimen compared with other regimens, a finding consistent with those of other SRL studies (43,44). However, efficacy was favorable with SRL, with only 1 episode of acute rejection. It is not clear if using a less intensive regimen of belatacept would improve outcomes; 2 Phase III studies suggested that a more intensive regimen of belatacept (combined with basiliximab and MMF) did not provide additional efficacy compared to a less intensive belatacept regimen and that overall safety may have been better with the less intensive belatacept regimen (32,33). Sirolimus trough levels were higher than specified posttransplant and remained on the high end of the specified range by Month 12, which may have contributed to the low incidence of acute rejection and poorer tolerability in that group. It is possible that a lower SRL dose in combination with belatacept may be better tolerated while still providing sufficient immunosuppression.

The overall safety profile of belatacept was generally consistent with the profile reported in the 2 Phase III studies of de novo belatacept treatment in kidney transplant recipients (32,33). Some reported adverse events (e.g. oral ulcers, tremor) were events likely associated with SRL and TAC. Importantly, there were no cases of PTLD reported with either belatacept regimen. Data from the Phase II and III studies in kidney transplant recipients showed an increased risk of PTLD, including CNS PTLD, in patients treated with belatacept-based regimens. Although the risk of PTLD with belatacept was increased for patients who were either EBV(−) or EBV (+) at baseline, the highest risk was in patients who were EBV (−) and/or were receiving the more intense regimen.

Belatacept did not appear to impair the recovery of B cell, T cell and NK lymphocyte populations after initial depletion as a result of Thymoglobulin induction, although observations were made on only a subset of patients. In addition, deleterious effects of belatacept treatment on regulatory and memory T cells were not evident in this study, consistent with previous work suggesting that belatacept did not have adverse effects on regulatory T cells (45).

Limitations to the study include its relatively small size and lack of statistical power. There was also the potential for observer bias as with any open-label study; however, adjudication of the primary endpoint was performed in blinded fashion. The number of patients who discontinued or switched part of their regimen limits the conclusions that can be drawn regarding treatment groups. For example, 10 patients in the belatacept-SRL group switched from SRL to MMF, often due to poor tolerability associated with SRL. However, approximately 81% of patients in the belatacept treatment groups remained on belatacept through Month 12, suggesting belatacept itself was well tolerated. Chronic allograft nephropathy (IF/TA) was not assessed in this study, which may have provided more information about allograft health. Increased fibrosis has been reported with steroid-free regimens (46), as has progressive chronic allograft nephropathy with CNI-based regimens (5). Belatacept-based regimens were associated with less chronic allograft nephropathy in 2 Phase III studies in kidney transplant recipients compared to a cyclosporine-based regimen (32,33).

In conclusion, this exploratory study demonstrates that a belatacept-based, steroid-avoiding regimen appears to provide effective immunosuppression with better renal function in de novo kidney transplant recipients compared to a TAC-based regimen. The study also suggests that belatacept has potential to be used in combination with Thymoglobulin induction in EBV positive patients. Larger studies that build on this exploratory study, optimizing the potential advantages of a CNI- and steroid-avoiding regimen, would further define the most favorable steroid-avoiding belatacept regimen.


With many thanks to the Belatacept 034 Study Group as follows:


Marwan Abouljoud & Ravi Parasuraman, Henry Ford Hospital, Detroit, MI; Adam Bingaman, The Texas Transplant Institute at Methodist Specialty and Transplant Hospital, San Antonio, TX; Laurence Chan, University of Colorado Health Science Center, Denver, CO; David Conti, Albany Medical College, Albany, NY; Sander Scott Florman, Tulane University Hospital, New Orleans, LA; David Gerber, University of North Carolina, Chapel Hill, NC; Paul Gores, Carolinas Medical Center, Charlotte, NC; Harold J Helderman, Vanderbilt University Medical Center, Nashville, TN; Thomas Johnson, University of Kentucky Medical Center, Lexington, KY; Dixon B Kaufman, Northwestern University School of Medicine, Chicago, IL; Bradley A Marder, Denver Nephrology PC, Denver, CO; William H Marks, Swedish Medical Center, Seattle, WA; Laura Mulloy, Medical College of Georgia, Augusta, GA; Uday S Nori & Ronald Ferguson, Ohio State University Medical Center, Columbus, OH; Surenda Shenoy, Washington University School of Medicine, St Louis, MO; Flavio Vincenti, University of California, San Francisco; E Steve Woodle, University Hospital & Christ Hospital, Cincinnati, OH.


Virginia Cabello, Hospital Universitario Virgen Del Rocio, Sevilla; Josep M. Grinyo, University Hospital of Bellvitge, Barcelona; Manuel Antonio Arias Rodriguez, University Hospital Marques De Valdecilla, Santander.


Franco Citterio & Marco Castagneto, Catholic University, Rome; Paolo Rigotti, Azienda Ospedale Universita Di Padova-Ospedale Giustinianeo, Padova; Silvio Sandrini, A.O. Spedali Civili, Brescia; Sergio Stefoni, St. Orsola University Hospital, Bologna.

The authors would also like to thank Dr. Kim Solez of the Department of Laboratory Medicine and Pathology, University of Alberta, Edmonton, Canada for central evaluation of renal biopsy specimens for acute rejection and Dr. Ping Zhan of Bristol-Myers Squibb for analysis of lymphocyte count data.

Funding source: The study was supported by Bristol-Myers Squibb.


The authors of this manuscript have conflicts of interest to disclose as described by the American Journal of Transplantation:

Drs. Ferguson and Citterio have nothing to disclose.

Dr. Grinyo has participated as an Advisory Board member for Bristol-Myers Squibb.

Dr. Vincenti has received research support from Bristol-Myers Squibb, Pfizer, Novartis, Astellas, Genzyme, Genentech and Roche.

Dr. Kaufman has participated as an Advisory Board member for Novartis and has received support for publication-related activities from Wyeth.

Dr. Woodle has received grant support or research contracts from Genzyme, Astellas, Bristol-Myers Squibb, Roche, Novartis and Wyeth.

Dr. Marder is an employee and stockholder in Denver Nephrology, PC.

Dr. Marks has received research funding from Genzyme.

Drs. Agarwal, Wu, Dong and Garg are employees of Bristol-Myers Squibb.

Drs. Brian Atkinson and David Hartree of Bristol-Myers Squibb provided professional writing and editorial assistance.

This study was supported by Bristol-Myers Squibb.