Two Hundred Living Donor Kidney Transplantations Under Alemtuzumab Induction and Tacrolimus Monotherapy: 3-Year Follow-Up


* Corresponding author: Henkie P. Tan,


Alemtuzumab has been used in off-label studies of solid organ transplantation. We extend our report of the first 200 consecutive living donor solitary kidney transplantations under alemtuzumab pretreatment with tacrolimus monotherapy and subsequent spaced weaning to 3 years of follow-up. We focused especially on the causes of recipient death and graft loss, and the characteristics of rejection. The actuarial 1-, 2- and 3-year patient and graft survivals were 99.0% and 98.0%, 96.4% and 90.8% and 93.3% and 86.3%, respectively. The cumulative incidence of acute cellular rejection (ACR) at the following months was 2%≤6, 9.0%≤12, 16.5%≤18, 19.5%≤24, 23.5%≤30, 24.0%≤36 and 25%≤42. The mean serum creatinine (mg/dL) and glomerular filtration rate (mL/min/1.73 m2) at 1 and 3 years were 1.4 ± 0.6 and 58.7 ± 21.6 and 1.5 ± 0.7 and 54.9 ± 20.9, respectively. Fifty (25%) recipients had a total of 89 episodes of ACR. About 88.7% of ACR episodes were Banff 1, and of those, 82% were steroid-sensitive. Nine (4.5%) recipients had antibody-mediated rejection (AMR). About 76.5% were weaned but only 46% are currently on spaced dose (qod or less) tacrolimus monotherapy, and 94.4% remained steroid-free from the time of transplantation. Infectious complications were uncommon. This experience suggests the 3-year efficacy of this approach.


Significant gradual improvements in outcomes after kidney transplantation have occurred with the introduction of azathioprine (1), prednisone (2), cyclosporine (3), tacrolimus (4) and antilymphoid globulins (ALGs) (5). The original polyclonal ALGs (6,7) have been replaced by ever more specifically targeted monoclonal antibodies (mabs) (8–12). One of the most promising is alemtuzumab (Campath-1H®), a broadly reactive humanized anti-CD52 mab developed by Waldmann and Hale (11,12) and first used for kidney transplantation by Calne et al. (13) in combination with minimal cyclosporine-based maintenance immunosuppression.

The low need for long-term immunosuppression in the pilot kidney trial of alemtuzumab was associated with reasonable patient and graft survival and a high quality of life, including freedom from opportunistic infection (14). Subsequent reports of alemtuzumab alone (15) or in combination with calcineurin inhibitors (CNI, tacrolimus or cyclosporine), sirolimus, mycophenolate mofetil (MMF) and steroids have generally confirmed Calne's initial observations (16–32). However, the optimal approach to the use of alemtuzumab has not been standardized (33,34).

In this report, we extend our experience with alemtuzumab 30 mg pretreatment and tacrolimus monotherapy in the first 200 consecutive live donor kidney transplants (LDKT), with a mean follow-up of 35.7 ± 9.9 months (range 20.5–41.5) (23). In contrast to other reports, alemtuzumab was given before, rather than after graft revascularization, hence the term alemtuzumab pretreatment or preconditioning rather than induction. In addition, an attempt was made following transplantation to minimize exposure to tacrolimus monotherapy by prolonging the intervals between doses to longer than daily (spaced weaning). The advantages and disadvantages of this strategy were evaluated, with a particular emphasis on an analysis of recipient death, graft loss and acute rejection characteristics.

Materials and Methods

Patient population and outcome measurements

Between March 2003 and December 2005, 200 consecutive solitary LDKT were performed from 200 laparoscopically recovered donor kidney allografts in 198 patients receiving alemtuzumab 30 mg (Campath-1H®, Berlex Laboratories, Montville, NJ) pretreatment and tacrolimus monotherapy. The causes of renal failure were similar to those in other previously reported studies (13–31,34) (Table 1). Glomerular filtration rate (GFR, mL/min/1.73 m2) was estimated with the modification of diet in renal disease (MDRD) and Schwartz formula. Noncompliance was defined as failure to take tacrolimus (weeks to months), failure to obtain laboratory studies (many months to years) and persistent refusal to undergo renal allograft biopsy (1 week to many weeks) as confirmed by at least three health professionals.

Table 1.  Donor and recipient characteristics of 200 LDKT under alemtuzumab pretreatment and weaning of tacrolimus monotherapy with a 3-year mean follow-up
 Female/male  119 (59.5%)/
 81 (40.5%)
 African American 28 (14%)
 Mean donor age (years) 42.3 ± 10.8
 For adult recipients (n = 183) 42.8 ± 10.9
 For pediatric recipients (n = 17) 35.9 ± 8.2 
 HLA mismatch 3.2 ± 1.6
Donor relationship to recipient n (%)HLA mismatch
 Parent38 (19%)2.7 ± 0.6
 Child33 (16.5%)2.3 ± 0.7
 Sibling45 (22.5%)2.0 ± 1.7
 Other related15 (7.5%)3.9 ± 1.1
 Genetically unrelated69 (34.5%)4.6 ± 1.2
 Hypertension 34 (17%)
 Glomerulonephritis/nephritis 30 (15%)
 Diabetic nephropathy 26 (13%)
 Polycystic 24 (12%)
 FSGS 18 (9%) 
 IgA Nephropathy 16 (8%) 
 HIV+  3 (1.5%)
 Other diagnosis 58 (29%)
 Adult/pediatric  183 (91.5%)/
 17 (8.5%)
 Female/male  80 (40%)/
120 (60%) 
 African American 34 (17%)
 Mean recipient age (years) 43.9 ± 17.9
 Mean adult age (years) 47.2 ± 14.8
 Mean pediatric age (years) 8.7 ± 5.6
 ≥70 years old  9 (4.5%)
 PRA > 20% 16 (8%) 
 Primary graft  175 (87.5%)
 Retransplants  25 (12.5%)
 2nd graft 18 (9%) 
 3rd or 4th graft  7 (3.5%)
Pretransplant diabetes (fasting 56 (28%)
 blood sugar >110 mg/dL)  


In the operating room or in the evening prior to transplantation, a single infusion of alemtuzumab 30 mg (0.4–0.5 mg/kg in pediatric patients) was given after premedication with 1 gm (or 10–15 mg/kg for pediatric) methylprednisolone, with another gram of methylprednisolone repeated prior to graft reperfusion. Tacrolimus monotherapy was given twice daily for the first 4 to 6 months with 12-h target trough levels of 8–10 ng/mL. No other immunosuppressive agent was added unless there was a breakthrough rejection (2% in the first 6 months).

After 3 to 6 months and in the absence of clinical rejection, the twice daily doses were consolidated to a single daily dose. In the continued absence of a rising serum creatinine, graft dysfunction, rising T-cell activation by the Cylex® assay, de novo anti-HLA alloantibodies monitored by the enzymed linked immunoabsorbent assay (ELISA), and identification of donor-specific antibodies (DSA) by Luminex®, the interval between tacrolimus doses was increased after every 3 to 6 months to every other day (qod), three times a week (3×/wk), and, less often, to twice a week (2×/wk), and once a week (1×/wk) (23).

Diagnosis and treatment of acute cellular rejection (ACR) and antibody- mediated rejection (AMR)

Biopsies were obtained if there was a rise in serum creatinine from a stable baseline. Of the 200 living donor transplants, 105 (52.5%) underwent biopsies for a rising serum creatinine, and 50 (25%) had ACR. With all biopsies, the findings were encoded into the Banff scores (35).

Interstitial fibrosis and tubular atrophy, no evidence of any specific etiology (IF/TA NES, previously known as chronic allograft nephropathy [CAN]) (35) scores were recorded as chronic glomerulopathy (cg), chronic interstitial fibrosis (ci), chronic fibrointimal vascular thickening (cv) and chronic tubular atrophy (ct), with a range from 0 to 3. From July 2003, all biopsies were stained for C4d. Protocol biopsies were not performed.

After September 2004, pre- and posttransplantation (every 3 months) sera of recipients were screened by ELISA for the presence of class I and class II IgG anti-HLA antibodies (One Lambda Inc., Canoga Park, CA). The results provided insight into the state of pre- and posttransplant humoral immune reactivity. DSA was identified and characterized by Luminex® if the ELISA class I or II antibodies were ≥10%, or when AMR was suspected. Identification of DSA in the recipients’ serum in the presence of graft dysfunction and supported by histologic findings (including C4d stains) was consistent with the diagnosis of AMR. However, identification of DSA in the context of patients with stable renal function was thought to be a marker of an increased risk for rejection (36), and patients on spaced weaning were taken back to once daily tacrolimus; if DSA did not subsequently disappear or the antibody titer continued to increase, MMF was added (37). This is also (but less often) done using a rise in the Cylex assay (i.e. if the numerical value continued to increase dramatically, spaced weaning might be abandoned). Serial measurements were the key to managing the patients.

Biopsy-proven ACR was treated with intensification of immunosuppression that ranged from a single large dose of methylprednisolone, resumption of daily tacrolimus in patients on spaced weaning and/or addition of other agents (e.g. MMF). Patients with steroid-resistant ACR were typically treated with a single dose of 30 mg alemtuzumab (or less commonly with Thymoglobulin®). Patients with diagnosis of AMR underwent a course of plasmapheresis (PP) and low-dose IVIg (100 mg/kg). PP/IVIg was stopped when the serum creatinine returned to baseline and/or when the DSA titer decreased significantly to a low titer or became undetectable.

Infection prophylaxis

All patients were on valganciclovir for at least 6 months. BK virus (BKV) screening in the plasma and urine was performed every 1 to 3 months beginning in Jan 2006. Assays of cytomegalovirus (CMV) antigenemia/PCR were performed weekly for 3 months, every other week during months 3 to 6, and then monthly.

Institutional oversight

As summarized elsewhere (38), all patients described in this article were treated according to formal clinical practice guidelines that were reviewed and approved for use in renal transplant patients by the Innovative Practices subcommittee of the Technology and Assessment Committee of the University of Pittsburgh Medical Center (UPMC). The proposed clinical guideline were also reviewed and approved by the UPMC Pharmacy and Therapeutics Committee because of the ‘off label’ use of alemtuzumab (FDA-approved, but not for use in solid organ transplantation). Following these reviews and institutional approvals in 2003, all adults receiving renal transplants at the Starzl Transplantation Institute were treated according to these guidelines, with therapy individualized by their treating physician, according to each patient's clinical course. Prior to the retrospective review of these clinical records for the purpose of this report, approval was obtained from the University of Pittsburgh IRB.

Statistical analysis

Kidney graft failure was defined as death of recipient, removal of the allograft, or loss of function requiring return to dialysis. Actuarial Kaplan–Meier survivals were calculated using the Statistical Package for the Social Science (SPSS) software.


Patient and graft survival

Donors:  Laparoscopic live donor nephrectomies (39–41) were performed without any serious complications; minor complications were seen in less than 2%. The live donors were predominantly female (59.5%), and 14% were African American (Table 1). The relationships of the donors to the recipients are summarized in Table 1, along with the HLA mismatches in each category.

Recipient characteristics and survival:  There were 183 (91.5%) adults (mean age: 47.2 ± 14.8 years) and 60% males (Table 1). Since there were no exclusions, there were many higher-risk recipients including 28% pretransplant diabetics/patients with glucose intolerance, 17% African Americans, 8.5% pediatric patients, 8% with a PRA >20%, 4.5%≥70 years old, 1.5% with HIV and 12.5% with previous failed transplants: 9% recipients were undergoing their 2nd, and 3.5% their 3rd or 4th transplant. The average HLA mismatch was 3.2 ± 1.6.

With a mean follow-up of 35.7 ± 9.9 months, the Kaplan–Meier patient actuarial survival at 1, 2 and 3 years was 99.0%, 96.4% and 93.3% (Figure 1A and Table 2). The best patient survivals were in the patients who remained on spaced weaning tacrolimus monotherapy (Table 2), likely a selection bias; patients who required minimal immunosuppression without rejection did the best. Survival in pediatric recipients exceeded that of adults (1-year patient and graft survival of 100%) but with shorter follow-up (mean 21.5 ± 14.5 months).

Figure 1.

Actuarial (A) patient and (B) graft survival of recipients of 200 live donor kidney transplantation at mean 3-year follow-up.

Table 2.  Survival (patient and graft) of 200 LDKT who received alemtuzumab pretreatment and tacrolimus monotherapy
 Recipient survivalGraft survival
All 200 grafts (198 patients)
 1 year99.0%98.0%
 2 year96.4%90.8%
 3 year93.3%86.3%
92 grafts who remain on spaced weaning tacrolimus monotherapy
 1 year100%100%
 2 year97.8%96.7%
 3 year96.0%94.9%
47 grafts who were never weaned
 1 year95.7%93.6%
 2 year93.5%84.3%
 3 year88.6%79.6%
61 grafts whose spaced weaning was stopped
 1 year100%98.4%
 2 year96.6%86.7%
 3 year92.5%78.2%

There were 12 (6%) recipient deaths; 8 had excellent graft function at the time of their death (Table 3). De novo malignancy contributed to 3 deaths (1, 6 and 11); only one of the 3 patients had ACR. Infection contributed to 5 deaths (Table 3): patients 4, 5, 7 (Cryptococcal meningitis, the only systemic fungal infection in this series), 8 and 9. The remaining 4 deaths were unrelated to infection or rejection (2,3,10 and 12) (Table 3).

Table 3.  Characteristics of the 12 recipient deaths
PtAge at time of Tx (years)Cause of ESRDPatient survival post-Tx (d)Renal allograft function at time of deathCause of deathCause of allograft lossImmunosuppression
  1. Pt = patient; Tx = transplant; ESRD = end stage renal disease; d = days; FSGS = focal segmental glomerulosclerosis; IVDA = intravenous drug abuse; CA = cancer; ACR = acute cellular rejection; FK = tacrolimus; SLE = systemic lupus erythematosis; HTN = hypertension; MI = myocardial infarction; CAD = coronary artery disease; IDDM = insulin dependent diabetes mellitus; CHF = congestive heart failure; GN = glomerulonephritis; adenoCA = adenocarcinoma; PP = plasmapheresis; IVIg = intravenous immunoglobulin; AMR = antibody-mediated rejection; HUS = hemolytic uremic syndrome; ORIF = open reduction internal fixation; fx = fracture; Trach = tracheostomy.

 159FSGS, IVDA1158ExcellentNon small cell CA of the lung, pneumonectomyPt death, postwean ACR(1b) at 431 dFK 2×/wk, solumedrol for 1b ACR
 233SLE601FailedIntracranial bleed, coumadin toxicityRecurrent lupus nephritisFK qod
 367HTN482FailedMI, preexisting CADPostwean ACR at 93 d (2b)Campath/solumedrol for ACR
 453FSGS,HTN, IDDM176FailedSepsis from colon perforation during allograft nephrectomyPP for recurrent FSGS, NO ACRFK stopped at time of graft failure
 567HTN582FailedCHF, pneumoniaPostwean ACR (1a) at 235 days, noncompliantFK stopped at time of graft failure
 669Membranous GN1063ExcellentPancreatic adenoCA, WhipplePt death, no ACRFK 2×/wk stopped at time known CA
 747FSGS857ExcellentCryptococcal meningitisPt death, PP/IVIg for AMR, recurrent FSGS, FK HUS, Pre-wean 1a ACRsx2 at 394 dFK qod stopped at time of cryptococcal meningitis
 859Wegener’s597ExcellentSepsis, fasciitisPt death, postwean ACR (1b) at 168 dThymoglobulin for ACR
 967Analgesic abuse471ExcellentPneumoniaPt death, no ACRFK 3×/wk
1056IDDM131ExcellentUnrelated pulmonary emboli post-ORIF of femur fx from fallPt death, no ACRFK bid
1168IDDM, HTN538ExcellentBowel resection for CA with complicationsPt death, no ACRFK qod
1216 monthsRenal tubular interstitial dysplasia547ExcellentTrach disconnected at home with alarm turned offPt death, no ACRFK qod

Graft survival and function:  The mean serum creatinine (mg/dL) and glomerular filtration rate (GFR, mL/min/1.73 m2) of all 200 allografts at 1,2 and 3 years were 1.4 + 0.6 and 58.7 + 21.6, 1.6 + 1.2 and 55.1 + 21.4 and 1.5 + 0.7 and 54.9 + 20.9, respectively. The GFR of the 12 deaths at 1, 2 and 3 years were 51.3 + 16.3, 44.1 + 29.1 and 50.6 + 14.0, respectively. The GFR of the surviving grafts plus those who died at 1, 2 and 3 years were 59.9 + 20.6, 58.2 + 19.3 and 57.2 + 19.2, respectively.

There was no delayed graft function. Actuarial 1-, 2- and 3-year graft survival rates were 98.0%, 90.8% and 86.3%, respectively (Figure 1B and Table 2). The best graft survivals were in the patients who remained on spaced weaning tacrolimus monotherapy (Table 2). There were 15 renal allograft losses in addition to the 12 recipient deaths (Table 4). In 9 (60%) of the 15 patients had developed ACR prior to their graft loss; of these, 5 were noncompliant. In 7 of the 9 patients had postweaning ACRs (3 were noncompliant), and the other 2 had preweaning ACR (both were noncompliant). Of the 7 postweaning ACRs that resulted in allograft loss, 4 (2% of the cohort of 200 transplants) were directly related to postweaning ACR in compliant patients; 3 of these 4 patients also had aggressive AMR with failure to respond to PP/IVIg.

Table 4.  Characteristic of the 15 recipient with renal allograft loss (excluding deaths)
PtAge at time of TxCause of ESRDGraft survival post-Tx (days)Donor age (relationship)Cause of allograft loss, treatmentImmunosuppression, time of ACRCompliance
  1. Pt = patient; Tx = transplant; ACR = acute cellular rejection; HTN = hypertension; FSGS = focal segmental glomerulosclerosis; FK = tacrolimus; Wk = week; PP: plasmapheresis; IVIg = immunoglobulin; PKD = polycystic kidney disease; IDDM = insulin dependent diabetes mellitus; AMR = antibody-mediated rejection; MPGN = membranoproliferative glomerulonephritis; GN = glomerulonephritis; IF/TA NES (CAN) = interstitial fibrosis tubular atrophy not otherwise specified (chronic allograft nephropathy); BKVN = BK viral nephropathy; qod = every other day; SLE = systemic lupus erythematosis.

 163HTN, FSGS69051 (wife)Recurrent FSGSFK 3×/wk prior to recurrent FSGS, no ACRYes
 225IgA (Berger’s)102728 (wife)1a ACR and AMR, solumedrol, PP + IVIgPostwean ACR at 816 daysYes
 358PKD70230 (son)1a ACR, solumedrolPostwean ACR at 289 daysYes
 438MPGN, IgA84436 (brother)1a ACRs (×3), 1b ACR, solumedrolPostwean ACR at 206, 244, 286, 445 daysNO
 535PKD356 (mother)Anesthesia complication, transection of renal vein with repairNo ACRYes
 667Chronic GN89644 (son)Severe IF/TA NES (CAN)No FK, cellcept monotherapy, pleural fluid suspicious for malignancyYes
 755IDDM36563 (husband)1a and 1b ACRs, BKVNPostwean ACR at 175 and 204 daysNO
 815Familial tubulointerstitial nephritis64546 (mother)Vascular torsion of intraabdominal allograft on postop day 645FK qod, no ACRYes
 961IDDM115928 (friend)2a, 1b ×2 ACRs, Campath ×2, PP/IVIg for AMRPostwean ACR at 138NO
1058DM63150 (brother)Severe IF/TA NES (CAN), donor diseaseNo ACRYes
1147Membranous GN46542 (sister)1b and 2a ACRs, AMR, BKVN, PP + IVIg for AMR, decrease FK for BKVNPostwean ACR at 115 and 162 daysYes
1270PKD119561 (friend)No ACR, DSA+, C4d negative, no AMR?, no PP/IVIgCellcept added for DSA+Yes
1367FSGS83941 (son)ACR ×3 (2a, 1b, 1a), Solumedrol ×4Postwean ACR at 179 daysYes
1424SLE72747 (mother)ACR (2a), CampathPrewean ACR at 718 daysNO
1534HTN60333 (wife)ACR (2a), CampathFK qod, prewean ACR at 571 daysNO

Two patients developed BK viral nephropathy (BKVN) with subsequent graft loss despite treatment with cidofivir and reduction of tacrolimus. Six (40%) of the 15 patients had graft loss not related to ACR (1,5,6,8,10,12) (Table 4).

Spaced weaning of tacrolimus monotherapy

At follow-up, 153 (76.5%) of the 200 grafts had been weaned to spaced dose monotherapy. 92 (60.1%) of the 153 grafts (or 46% of the cohort of 200 grafts) who were weaned remained on spaced weaning (never stopped) (Figure 2. Weaning was suspended because of postweaning ACR in 36 (18%) patients, and because of the identification, without any evidence of ACR, of de novo DSA, rising nondonor-specific ELISA anti-HLA alloantibodies, and/or patient/physician anxiety in 25 (12.5%) patients. Of the 92 grafts that remained on spaced weaning, 87 (94.6%) continue to have a functioning graft. In the subgroup of patients whose weaning was suspended because of ACR (n = 36), graft failure occurred in 28% (10 patients: 2 patients died with excellent graft function) (Table 3).

Figure 2.

Impact of weaning on 200 living donor renal allografts.

Spaced weaning was not initiated in 47 (23.5%) of the 200 transplantations. The decision not to wean was related either to the development of rejection while on daily therapy (preweaning ACR, n = 14, 7%), identification of de novo DSA, rising ELISA anti-HLA alloantibodies, or patient or physician anxiety. The actuarial 1-, 2- and 3-year patient and graft survivals for these patients in whom tacrolimus was not weaned were 95.7% and 93.6%, 93.5% and 84.3% and 88.6% and 79.6%, respectively, and were worse than in the patients who underwent spaced weaning (Table 2).

At most recent follow-up, of the 173 functioning allografts, 87 (50.3%) were on spaced dose tacrolimus monotherapy (qod: 21.4%, 3×/wk: 21.4%, 2×/wk: 4.6%, and 1×/wk: 2.9%). In 11 (7.2%, including 4 deaths) of the 153 recipients who were weaned had graft failure; 4 were noncompliant. Thirty (17.3%) patients are on multiple immunosuppressive agents because of ACR.

The HLA ABDR mismatch in patients who were on daily (n = 61), spaced dose (n = 87) tacrolimus monotherapy, or multidrug therapy (n = 30) was not statistically different (3.7 ± 1.5 vs. 2.7 ± 1.7 vs. 3.5 ± 1.3 vs. p > 0.05), although there was a clear trend for fewer mismatched antigens in patients who remained on spaced weaning. 94.4% of recipients remained steroid free following transplantation.

Fifteen (7.5%) recipients had zero HLA mismatch or HLA identical kidneys (all siblings), and their tacrolimus dosing is as follows: bid (n = 1), qod (n = 2), qod (n = 3), 3×/wk (n = 2), 2×/wk (n = 4), 1×/wk (n = 3). Only 1 of these 15 patients experienced ACR.


Incidence over time:  The cumulative incidence of ACR posttransplantation was 1.0%≤1 month, 1.0%≤3 months, 2.0%≤6 months, 9.0%≤12 months, 16.5%≤18 months, 19.5%≤24 months, 23.5%≤30 months, 24.0%≤ 36 months and 25%≤ 42 months (Figure 3). Only 3 (1.5%) of the 200 patients had ACR beyond 2.5 years posttransplantation. The highest incidence of ACR appeared to be 6 to 18 months posttransplantation. Patient and graft survival appeared to be worse if ACR occurred between 18 months and 24 months (67% and 50%, Table 5).

Figure 3.

Incidence of cumulative ACR over time.

Table 5.  Characteristics of patients with ACR at 6 months interval posttransplantation
  1. ACR = acute cellular rejection (Banff g, i, v, t scores range from 0 to 3 for each category, with higher scores indicating more severe abnormalities).

  2. HLA MM = human leukocyte antigen mismatch.

  3. *= the same patient may have more than 1 biopsy.

Six months interval posttransplantation0 to ≤ 6>6 to ≤12>12 to ≤18>18 to ≤24>24 to ≤30>30 to ≤36>36 to ≤42>42
Cumulative% incidence of ACR (n)2% (4)9% (18)16.50% (33)19.50% (39)23.50% (47)24% (48)25% (50)25% (50)
Days (mean) to ACR72 ± 64305 ± 41 444 ± 56 637 ± 56 827 ± 64 10751126 ± 37   
Post-ACR follow-up days1073 ± 235 563 ± 337633 ± 341363 ± 376300 ± 207505268 ± 18  
Patient/graft survival (with ACR)100%/100%93%/43%87%/80%67%/50%100%/88%100%/100%100%/100% 
HLA MM3.5 ± 1.33.7 ± 1.63.8 ± 0.92.3 ± 1.64.0 ± 2.343.5 ± 0.7 
Patients who were weaning at ACR08125812 
ACR total score (g,i,v,t: max 12)2.8 ± 1.94.8 ± 2.53.8 ± 1.75.3 ± 2.23.5 ± 1.153.0 ± 0.7 
 ACR g (glomerulitis) score 0.5 ± 0.580.96 ± 1.120.43 ± 0.68 0.5 ± 0.840.25 ± 0.38 20 
 ACR i (interstitial inflammation) score0.88 ± 0.631.64 ± 0.911.63 ± 0.772.08 ± 0.801.69 ± 0.6511.25 ± 0.35 
 ACR v (vasculitis) score0.25 ± 0.50.23 ± 0.600.07 ± 0.260.17 ± 0.41000 
 ACR t (tubulitis) score1.13 ± 0.851.96 ± 0.991.67 ± 0.90 2.5 ± 0.841.56 ± 0.5021.75 ± 0.35 

Severity:  There were 38 biopsies that were called Banff borderline (i.e. there was evidence of immune activity but insufficient grounds for the diagnosis of rejection); they were generally seen posttreatment for ACR, and were usually not treated or at most were treated with a small bolus of methylprednisolone. There were 89 episodes of ACR in 50 patients (Table 7). Of the 89 episodes of ACR, 79 (88.7%) had Banff 1 (1a: 43 episodes or 48.3% of patients; 1b: 36 episodes or 40.4% of patients). Ten (11.2%) episodes were Banff 2: 2a: 9%; 2b: 2.2%; none were Banff 3. One (50%) patient with Banff 2b and 5 (62%) with 2a ACR eventually lost their grafts.

Table 7.  Severity of 89 ACR episodes (Banff ≥ 1a) in 50 recipients and characteristics of pre- and postweaning ACRs
Banff score89 ACR (Banff ≥ 1a) episodes in 50 recipients
  1. Some patients have 1 episode of rejection with >1 biopsies showing resolution of ACR, e.g. 2b → 1b or 1b → borderline. Patients with 2b → 1b ACR on 2 biopsies were considered as having an initial 2b ACR while receiving treatment with improvement in rejection.

Banff 1a43 (48.3%)
Banff 1b36 (40.4%)
Banff 2a8 (9.0%)
Banff 2b2 (2.2%)
Banff 30 (0%) 
Characteristics of pre- and postweaning ACRs
 Total preweaning ACR (usually <1-year posttransplant)14 (7.0%) 
 —Total graft loss (2 + 1 death) 3 (21%) 
 —Banff 2a ACR 2 (14%) 
 —AMR1 (7%) 
 —Banff ≥ 2b ACR0
 —Death + AMR1 (7%) 
 —AMR1 (7%) 
 Total postweaning ACR36 (23.5%)
 —Total graft loss (7 + 4 deaths)11 (30.5%)
 —Secondary to noncompliant (+1 death) 5 (13.9%)
 —Banff 2a ACR 6 (16.7%)
 —Loss of graft4 (11%) 
 —BKVN1 (2.7%)
 —Banff 2b ACR2 (5.5%)
 —Death and loss of graft1 (2.7%)
 —Banff 3 ACR0

There were 15 graft losses: (a) 10 from ACR (including 5 deaths), (b) 3 ACRs secondary to noncompliance and (c) 2 BKVN (1 patient who developed BKVN was also noncompliant).

Of the 14 (7%) patients with preweaning ACR, 4 (28.6%) were noncompliant and 2 (14.3%) of the noncompliant patients with preweaning ACR lost their graft. There were 3 (21%) graft losses in the 14 recipients with preweaning ACR. In the 36 (23.5%) patients with postweaning ACR, there were 11 (30.6%, including 4 deaths) graft losses and 4 of these patients were noncompliant (Table 7). Eleven (30.6%) of the 36 postweaning ACRs were a result of noncompliance. Of the 50 patients with ACR, 15 (30%) were noncompliant. About 33% (5 of 15) of the graft losses attributable to ACR were related to noncompliance. This relatively simple protocol is actually easy with which to comply, but may have negative consequences when a patient on maintenance monotherapy immunosuppression develops rejection after spaced weaning or as a result of noncompliance.

It appeared that the severity of ACRs was higher in patients with postweaning ACR, but the number of patients with Banff 2 ACR was clearly low (2 Banff 2b ACRs, 8 Banff 2a ACRs and no Banff 3 ACR). There were 2 postweaning 2b ACR. Of the 8 2a ACR, there were 2 preweaning ACR. In the 6 postweaning Banff 2a ACR, 2 were noncompliant.

To define further the contribution of the subcomponents of the Banff ACR score, the degree of glomerulitis (g), interstitial inflammation (i), intimal arteritis (v) and tubilitis (t) scores, which range from 0 to 3, with higher scores indicating more severe abnormalities or ACR, was determined at 6-month intervals in all the patients who had biopsies with ACR (Table 5). The most severe degree of ACR by total ACR score was at 18 to 24 months (ACR score of 5.3), and was associated with poor patient and graft survival (67% and 50%). The degree of (i) and (t) was worst in patients with ACR from 18 to 24 months, respectively. The degree of (v) was worse in patients with ACR from 0 to 6 months (0.25 ± 0.5) and was zero beyond 30 months posttransplantation.

Treatment of rejection

Seventy-three (82%) of the 89 ACR episodes responded to a bolus of methylprednisolone. Fourteen (28%) of the 50 patients (or 7% of the cohort of 200 cases) with rejection who had been on daily tacrolimus never were taken to spaced weaning, while 36 (72% or 18% of the whole group) of those who already were on spaced weaning were permanently returned to once daily tacrolimus. Twenty-one (42%) of the 50 patients who had ACR had recurrent ACR, including second (n = 10), third (n = 7), fourth (n = 2), fifth (n = 1) or sixth (n = 1) rejections, and required additional treatment and initiation of multiple immunosuppressive agent therapy.

We previously demonstrated that alemtuzumab might also be effective in the treatment of steroid-resistant ACR (42). Fourteen (28%) of the 50 patients or 16 (18.0%) of the 89 ACR episodes were steroid-resistant and were treated with alemtuzumab 30 mg intravenously for treatment of their rejections (Banff 1b ACR:n = 7, 2a ACR:n = 5, 2b ACR:n = 2). Five of the 14 patients whose ACR was treated with alemtuzumab lost their graft – 2 had AMR, 3 were noncompliant, and 1 patient died. The remaining 64% had improvement or resolution of their ACR.

Nine (4.5%) patients with AMR were treated with PP/IVIg. Four of the 9 AMR patients eventually lost their grafts (1 developed BKVN and 2 patients were noncompliant) (Tables 3 and 4).


To define the degree of IF/TA NES (CAN) in our 200 living donor transplant recipients, all 105 patients who had biopsies were reviewed (50 patients with ACR/AMR and 55 patients who had biopsies but did not have ACR). The degree of IF/TA NES (CAN) subcomponents of cg, ci, cv and ct scores were determined (Table 6). The IF/TA NES (CAN) total score (cg, ci, cv, ct, maximum score of 12 with 3 each) in these patients with ACR/AMR increased generally over time (Table 6). The degree of ci score appeared to be worst at 1.25 at 24 to 30 months and at 36 to 42 months. The degree of ct (1.5 ± 0.53) appeared to be worst at 24 to 30 months posttransplantation.

Table 6.  Quantitative assessment of histologic chronicity in biopsies of patients who developed ACR
  1. *Based on 50 patients with 89 biopsies.

  2. IF/TA NES (CAN) = interstitial fibrosis and tubular atrophy, no evidence of specific etiology (chronic allograft nephropathy).

Six months interval posttransplantation0 to ≤ 6>6 to ≤12>12 to ≤18>18 to ≤24>24 to ≤30>30 to ≤36>36 to ≤42>42
Number of patients*4141568120
IF/TA NES (CAN) total score (cg,ci,cv,ct: max 12)1.9 ± 1.73.8 ± 2.42.8 ± 1.83.0 ± 1.74.1 ± 1.364.8 ± 0.4 
 IF/TA NES (CAN) cg (chronic glomerulopathy) score (max of 3)00.43 ± 0.6500.17 ± 0.410.56 ± 0.6220.75 ± 1.06 
 IF/TA NES (CAN) ci (chronic interstitial fibrosis) score (max of 3) 0.5 ± 0.581.00 ± 0.761.00± 0.760.92 ± 0.801.25 ± 0.4611.25 ± 0.35 
 IF/TA NES (CAN) cv (chronic fibrointimal vascular thickening) score (max of 3)0.63 ± 0.751.08 ± 0.860.8 ± 0.680.67 ± 0.820.81 ± 0.652 1.5 ± 0.71 
 IF/TA NES (CAN) ct (chronic tubular atrophy) score (max of 3)0.75 ± 0.5 1.36 ± 0.721.03 ± 0.671.25 ± 0.761.50 ± 0.5311.25 ± 0.35 

In the patients who had biopsies but did not have ACR (55 patients), we did not have a sufficient number of biopsies greater than 6 months posttransplant to comment on the incidence and progression of IF/TA NES (CAN) (43). The biopsies of 45 of the 55 patients in this group were performed within 6 months posttransplantation.


There were no cases of CMV disease. One (0.5%) patient developed posttransplant lymphoproliferative disorder 17 months posttransplantation, on every other day tacrolimus. The incidence of recurrent FSGS was 16.7% (3/18).

Fifty-six (28%) recipients had diabetes/glucose intolerance (FBS >110 mg/dL) prior to transplantation, but only 26 (13%) were transplanted for ESRD secondary to diabetic nephropathy. Only 1 (0.7%) of the 144 patients without pretransplant diabetes developed new onset insulin dependent diabetes mellitus posttransplant. We attribute this to the combination of steroid avoidance and low-dose tacrolimus monotherapy.

One vascular thrombosis occurred in this series (patient 5, Table 4). The incidence of readmission to the hospital for any reason was <10% within the first 3 months posttransplantation (23,44).


Kidney alloengraftment has been suggested as a form of partial tolerance explained by mechanisms that are optimally expressed under light immunosuppression but are variably eroded by conventional heavy multiple drug therapy (45,46). In our own recent experience (21–27,46–51), light immunosuppression was guided by two immunologically derived therapeutic principles (45): lymphoid depletion prior to graft revascularization that is pretreatment, and second, the minimal use of posttransplant tacrolimus monotherapy, including extension of the intervals between tacrolimus doses to longer than once daily (spaced weaning). We report here an assessment of this strategy, examining outcomes, with a particular emphasis on the causes of patient death, graft failure, and rejection characteristics.

Alemtuzumab, the most potent currently used lymphoid depleting antibody (33,34,52,53) is associated with profound and prolonged depletion of T cells, and to a lesser extent B cells, NK cells and monocytes (11,15,18,53,54). While alemtuzumab has not been approved by the Food and Drug Administration for use in organ transplantation (55), the experiences with this agent in renal transplantation have been promising (13–16,18,20–32,54,56).

In this report, we present a mean 3-year follow-up of the longest and largest series to date of 200 living donor kidney transplantations performed under alemtuzumab pretreatment and tacrolimus monotherapy, with subsequent weaning when possible. The outcomes confirm the 3-year safety and efficacy of this novel approach. Despite the complexity of the recipient population, the actuarial 1-, 2- and 3-year patient and graft survivals were 99.0 and 98.0%, 96.4% and 90.8% and 93.3% and 86.3%, respectively, and are equal to those of living donor kidney transplants recipients in the Scientific Registry of Transplant Recipients data base (57).

The cumulative incidence of ACR posttransplantation was 1.0%≤3 months, 2.0%≤6 months, 9.0%≤12 months, 16.5%≤18 months, 19.5%≤24 months, 23.5%≤ 30 months, 24.0%≤ 36 months and 25.0%≤ 42 months. Only 3 (1.5%) of the 200 patients had ACR between 2.5 years and 3.5 years posttransplantation. Weaning was attempted in 76.5% of patients. The 94.4% of recipients remained steroid-free after transplantation.

Current maintenance immunosuppression for kidney transplantation includes CNI-based triple drug therapy. Despite induction and strong maintenance immunosuppression, the incidence of ACR in the first year in a 3-arm prospective randomized study was as high as 17% (28). CAN leading to allograft dysfunction and loss may be related, at least in part, to chronic CNI use (58). In addition, heavy chronic immunosuppression is frequently associated with renal, cardiovascular, metabolic and infection-related mortality and morbidity (59). Thus, to improve transplantation outcomes, new immunosuppressive strategies that allow minimization of chronic heavy immunosuppression, while achieving long-term graft function, must be explored.

Our findings support and extend those of Calne and colleagues in 1997, who first employed alemtuzumab as induction therapy followed by low-dose cyclosporine monotherapy in 33 renal transplantation patients (13,14,16). The 5-year follow-up study demonstrated that the incidence of ACR, patient and graft survivals, and incidence of infectious complications was similar in the alemtuzumab and the cyclosporine-based triple immunosuppression control groups (16).

We are seeing an increasing number of plasma cells on histologic biopsies with no evidence of ACR or AMR (data not shown). The significance of plasma cells in these biopsy specimens in the absence of AMR or ACR is unclear at this time. Recent reports have described severe, early acute humoral rejection (60). Since September 2004, we routinely identify DSA and since July 2003, we routinely perform histologic C4d stains on all our recipients. Of the 50 recipients with ACR, 9 (4.5%) patients also had AMR.

The findings in this report have several important limitations. Continued follow-up will clearly be required to assess longer-term graft survival and function, progression of IF/TA NES (CAN), and the incidence of infections and malignancies. The delayed incidence of ACR and possibly higher incidence of AMR that occurs as lymphocytes return to baseline will need to be monitored closely, as will the rates of late IF/TA NES (CAN) in patients with or without acute rejection. It is also hoped that with a lower but effective starting dose of tacrolimus progressively weaned over time, a decreased incidence of CNI nephrotoxicity will be observed compared to traditional triple immunosuppression. We cannot make a conclusion in regard to the progression of IF/TA NES (CAN) in our patients, as we do not have a sufficient number of biopsies to compare with Nankivell's series (43).

The long-term benefits of alemtuzumab pretreatment with tacrolimus monotherapy clearly warrant further investigation. The immunologic mechanisms that permit the ability to wean immunosuppression optimally in these patients have yet to be elucidated. Additional studies directed at elucidating these mechanisms are currently under way. A prospective, randomized study of spaced weaning using this protocol, with protocol biopsies, is being planned and will be initiated shortly. Since our incidence of postweaning ACR is higher than that of preweaning ACR, we now routinely search for DSA, perform histologic C4d stains, ELISA class I and II HLA alloantibodies and T-cell activation by the Cylex® assay.