This site is intended for health professionals only
Assistant Professor of Surgery
Professor of Surgery
Kidney, Pancreas, and Islet Cell Transplantation
Thomas E Starzl Transplantation Institute
University of Pittsburgh
Long-term steroid administration after solid organ transplantation is associated with a number of well-known side-effects (see Table 1). The goal of withdrawing or avoiding steroids is to minimise these side-effects, the cost of which has been calculated to be US$5,300 ($1=e. 1) per year per patient.(1)
Reducing steroid use can be achieved either by withdrawal or avoidance. In steroid withdrawal, doses of steroids may be tapered early (less than three months) or late (more than three months) after transplantation. Steroid avoidance has been used in trials using antilymphocyte globulin induction therapy or anti-IL2 receptor antagonists and maintenance with ciclosporin and mycophenolate mofetil. T-lymphocyte depletional strategies with rabbit antithymocyte globulin or alemtuzumab followed by tacrolimus or sirolimus monotherapy are also examples of steroid avoidance.
This article looks at the results of some of the main trials of steroid withdrawal and avoidance following kidney transplantation. Reducing steroid use following other types of organ transplantation will be discussed in the next issue.
Steroid withdrawal in adult patients
Ciclosporin-based immunosuppression protocols
The Canadian Multicenter Transplant Study group published the largest randomised controlled trial of steroid withdrawal.(2) Five-year graft survival was 73% in the steroid withdrawal group versus 85% in the ciclosporin and the prednisolone groups (p<0.03). The authors concluded that continued administration of prednisolone was advisable.
In a recent European multicentre study, kidney transplant recipients receiving ciclosporin and mycophenolate mofetil were randomised to receive either a standard steroid regimen or a low-dose steroid regimen (50% of the dose of the control group), followed by steroid withdrawal three months post-transplantation.(3) The biopsy-proven acute rejection rate at one year was 25% in the low-dose/steroid withdrawal group, versus 15% in the control group (p<-0.01). At one year post-transplantation, patient and graft survival rates were 98% and 94% in the low-dose/steroid withdrawal group versus 97% and 93% in the control group. The low-dose/withdrawal group experienced a reduction in cardiovascular risk factors (eg, hypertension, cholesterol and triglyceride levels) and an improvement in bone density compared with the control group.
The North American multicentre, prospective, randomised, double-blind trial of prednisolone withdrawal recruited primary adult transplant recipients with no acute rejection by 90 days, who were taking mycophenolate mofetil >-2g/day, ciclosporin 5–15mg/kg/day, and prednisolone 10–15mg/day. Patients were randomised to receive prednisolone tapered over eight weeks to zero, or to continue taking 10mg/day. After 266 patients were enrolled, recruitment was stopped because of a significantly increased incidence of rejection in the prednisolone withdrawal group.(4) Despite this difference in acute rejection, there was no statistical difference in patient or graft survival, or in six-month serum creatinine level. African-American patients in the steroid withdrawal arm had a higher risk of acute rejection than non-African-Americans. The authors concluded that the majority of post-transplant patients maintained on ciclosporin and mycophenolate mofetil could be withdrawn from steroids without experiencing acute rejection.
There are no randomised, controlled studies of steroid withdrawal in kidney transplant patients receiving tacrolimus as baseline therapy. The Pittsburgh group discontinued prednisolone in 531 (71%) of 795 renal transplant patients, at a median of 9.7 months after transplantation; 398 (75%) remained off prednisolone (OFF), 133 (25%) were restarted on prednisolone a median of 14.5 months after discontinuing steroids (OFF/ON), and 217 patients (29%) were never taken off prednisolone (ON). The one-, three- and five-year graft survival rates in the OFF group were 99.5%, 96% and 89%, respectively; in the OFF/ON group they were 98.5%, 89% and 71%, and in the ON group they were 74%, 52% and 42%.(5) Acute rejection-free survival was 65% in the OFF group, compared with 20% and 24% in the OFF/ON and ON groups, respectively. Patients were not randomly assigned to steroid withdrawal, and those who never came off steroids (ON) had a number of risk factors, including retransplantation, sensitisation and acute rejection. Patients who discontinued and then resumed steroids had comparable survival, but lower graft survival and worse renal function, than those who came off steroids. Patients who never discontinued prednisolone had the worst outcome.
Twenty-seven patients were entered into a study of corticosteroid cessation one week after renal transplantation using tacrolimus–mycophenolate mofetil-based immunosuppression.(6) Acute rejection occurred in seven patients (26%), with three of these (43%) occurring before corticosteroid withdrawal. If these three patients were excluded from analysis, 17% of patients experienced rejection after withdrawal of steroids.
Kaufman et al carried out a study of 54 consecutive patients with renal transplant who underwent rapid corticosteroid elimination after three days.(7) The control group had 40 patients and received a standard prednisolone taper. All patients received induction with an anti-IL2-receptor antibody (except HLA-ID grafts) and three doses of methylprednisolone (500mg, 250mg and 125mg); maintenance immunosuppression was with tacrolimus and mycophenolate mofetil only. The incidence of rejection at six months post-transplant in the prednisolone-free group was slightly higher than the controls (13.2% versus 8.3%). This difference disappeared at 12 months.(7)
In an extension of the previous study, Stuart et al administered alemtuzumab intraoperatively followed by intravenous methylprednisolone taken on four consecutive days in a stepwise reduction (500mg, 250mg, 125mg, stop) to 60 consecutive renal transplant patients.(8) Forty-five patients received tacrolimus and mycophenolate mofetil, starting the day after transplantation, and 15 patients received tacrolimus 7–10 days after transplantation. There was only one episode of rejection, which responded to a short course of methylprednisolone, in the patients treated with tacrolimus the day after transplantation. In the patients who were started on tacrolimus 7–10 days after surgery, three patients sustained acute cellular and vascular rejection, which was reversed with methylprednisolone, plasmapheresis and antilymphocyte globulin.
Between November 1999 and October 2000, 51 adult living donor first-transplant recipients were immunosuppressed with thymoglobulin (1.25mg/kg intraoperatively and then four times daily for four days); methylprednisolone (500mg intraoperatively, 1mg/kg for one day, 0.5mg/kg for two days, 0.25mg/kg for two days, then stopped); mycophenolate mofetil (1g twice daily); and ciclosporin (4mg/kg twice daily).(9) There was no significant difference in six- and 12- month patient survival, graft survival and rejection-free graft survival between recipients on the rapid discontinuation of steroid (RDS) protocol versus historical controls (prednisolone, mycophenolate mofetil and either azathioprine or ciclosporin, both without antibody induction). Of the 51 RDS patients, five (10%) had acute rejection (all within 3.5 months of transplantation). After treatment, all five were maintained on 5mg prednisolone; there have been no second acute rejection episodes.
One hundred and twenty-four patients underwent steroid withdrawal at various time points ranging from one week to more than two years post-transplantation. Ninety-five patients (76.6%) remain off steroids.(10) There was a 4% incidence of acute rejection. In addition, 24 patients required reinstitution of steroids because of chronic rejection, hypoadrenalism, sirolimus discontinuation, graft loss, recurrent disease or patient preference.
Meta-analysis of steroid withdrawal trials(11)
In a meta-analysis of nine prednisolone withdrawal trials (n=1,461), the proportion of patients with acute rejection was increased by 0.14 (95% confidence interval 0.10–0.17, p<-0.001). In nine trials (n=1,899) analysed for graft failure, the relative risk (RR) of graft failure after withdrawal was also increased (RR=1.40; range 1.09–1.70, p=0.012). There was no evidence of between-study heterogeneity for either acute rejection or graft failure.
There was no difference in acute rejection between the steroid withdrawal and control groups in a comparison of trials that used mycophenolate mofetil with those that did not. The mean follow-up of all prednisolone withdrawal trials was 28 months.(11)
Steroid withdrawal in paediatric transplantation
Ellis studied the effect of steroid withdrawal in 68 children undergoing renal transplantation using tacrolimus-based immunosuppression, 87% of whom remained off steroids during the subsequent two years.(12) The proportion of children who lost their graft or died after steroid withdrawal (8/68) was significantly lower compared with those who remained on steroids at one year after transplantation (5/8) (p<-0.01). Assessment of growth in 52 children who underwent steroid withdrawal in the first post-transplant year showed normalisation of growth.(12)
In another report from Pittsburgh,(13) 74 paediatric patients were taken off prednisolone a median of 5.7 months after transplantation. Fifty-six (70%) patients remained off prednisolone (OFF), while 18 (22.5%) were restarted on prednisolone a median of 14.8 months after discontinuing steroids (OFF/ON); six (7.5%) were never taken off prednisolone. The one-, three- and five-year graft survival rates in the OFF group were 100%, 95%, and 82%, respectively. In the OFF/ON group, they were 100%, 89%, and 83%; and in the ON group, they were 100%, 50%, and 33%. Serum creatinine at most recent follow-up was 1.2±0.5, 1.8±0.9 and 2.0mg/day, respectively. The incidence of rejection (prior to steroid withdrawal) in the OFF, OFF/ON, and ON groups was 39%, 77% and 100%, respectively. Patients were not randomly assigned to steroid withdrawal, and patients who never came off steroids had a number of risk factors, including retransplantation, sensitisation and acute rejection.
In a retrospective, consecutive clinical study with historical controls, 68 renal transplant patients (aged 8–67 years, median 39.5 years) received a steroid-free immunosuppressive regimen comprising 10 days of antilymphocyte globulin induction and maintenance therapy with ciclosporin and mycophenolate mofetil.(14) After an observation of up to 2.5 years, 66 (97%) patients were alive and 64 (94%) grafts were functioning. The incidence of acute rejection was 15%, a rate significantly less than the 37.4% observed in 190 previous consecutive transplants treated without mycophenolate mofetil but otherwise receiving the same protocol.(14)
In a prospective, nonrandomised study, 100 consecutive renal transplant patients in Denmark received steroid-free immunosuppression with an initial 10-day course of induction with antithymocyte globulin and maintenance therapy with ciclosporin and mycophenolate mofetil.(15) After an observation time of 4.5 years, one-, two-, three- and four-year graft survival rates of 97%, 96%, 90% and 82% were observed. The incidence of acute rejection was 13%; 10 were early (<three months) and three were late (6–24 months). The seemingly paradoxical result that a lower rate of rejection was observed in recipients with complete or near-complete avoidance of corticosteroids than in recipients in whom steroids were slowly withdrawn over months could be explained, in theory, by the absence of potential host immune activating effects of chronic corticosteroids.(16) Glucocorticoids decrease cytokine production, but the effect may be offset by upregulation of proinflammatory cytokine receptor expression on T-cells.(17)
In steroid withdrawal protocols, the risk of acute rejection is increased in patients who discontinue steroids within three months of transplantation. Late withdrawal of corticosteroids (12 months post-transplantation) seems to be associated with a lower risk of rejection and may reduce lipid levels and blood pressure. However, steroid-related side-effects, such as infection, osteopenia, cataracts and diabetes, may not improve after late withdrawal of corticosteroids.(18) Avoidance of steroids from the beginning may offer better results than withdrawal. Patients initially assigned to steroid-free immunosuppression who required steroids had long-term outcomes similar to those in patients who received steroids immediately after transplantation.(19)
A steroid-free regimen of daclizumab, mycophenolate mofetil and ciclosporin was given to 57 patients undergoing renal transplantation.(20) At one year, patient and graft survival was 95% and 89%, respectively. Fourteen patients (25%) experienced rejection, of which 13 were readily reversed with steroids. The mean serum creatinine at 12 months in patients not experiencing rejection was 1.7±0.7mg/dl, compared with 1.8±1.2mg/dl for those experiencing rejection. A reduction in antihypertensive medication requirements, a low incidence of post-transplant diabetes mellitus and stability of bone density, all noted in the study, may be attributable to the avoidance of steroids.
An open-labelled, prospective study of complete steroid avoidance was carried out in 10 unsensitised paediatric recipients (mean age 14.4 years) of first renal allografts.(21) Steroids were substituted by the extended use of daclizumab, in combination with tacrolimus and mycophenolate mofetil. In the steroid-free group, protocol biopsies were performed at one, three, six and 12 months post-transplantation. Comparison was made to a steroid-based group of 37 matched historical controls. Graft and patient survival was 100% in both groups. Clinical acute rejection was absent in the steroid-free group, with a mean follow-up of nine months. Protocol biopsies in the steroid-free group revealed only two instances of mild subclinical rejection and no chronic rejection. Complete steroid-free immunosuppression is efficacious and safe in this selected group of children, with no early clinical acute rejection episodes. This protocol avoids the morbid side-effects of steroids without increasing infection.
Steroid avoidance using T-lymphocyte depletion strategies
In the absence of an aggressive T-cell response, an engagement of the immune system can lead to tolerance.(22) T-lymphocyte depletional strategies using alemtuzumab(23–25) or thymoglobulin(26) administered as preconditioning agents are based on this principle.
Calne et al used alemtuzumab preconditioning and post-transplant monotherapy with ciclosporin in 31 kidney transplant recipients.(23) At 15–28-month follow-up, 29 (94%) patients had sustained graft function. Twenty-seven patients were on low-dose ciclosporin, while three patients were also on prednisolone 10mg/day and azathioprine 100mg/day after rejection episodes. All patients had immediate and profound depletion of circulating lymphocytes, which recovered within 2–6 months.
The efficacy and safety of alemtuzumab preconditioning with sirolimus post-transplantation was assessed in 24 kidney transplant patients.(24) Follow-up was from 3–19 months. Biopsy-proven rejection occurred at 1–4 weeks in four patients, one resulting in graft loss. Protocol biopsies at six and 12 months showed no acute or chronic rejection. T-cells were depleted by two logs at two weeks. CD4 and CD8 cells remained depleted at six and 12 months, while monocytes substantially repopulated by one month and B-lymphocytes repopulated to 50% of baseline by three months and fully by 12 months.
Kirk et al expanded upon earlier work defining the effects of alemtuzumab, exploring higher doses and the use of the TNFa sequestrant infliximab.(25) Patients received three (n=8) or four doses (n=3) of alemtuzumab (0.3mg/kg/dose) or four doses plus infliximab (3mg/kg for three days, n=4). Sirolimus was added based on clinical grounds or protocol biopsy. No graft loss or rejection has been seen based on 6–22 months of follow-up (mean creatinine 1.4mg/dl, range 0.9–2.0). One patient died of pre-existing cardiac disease with a functioning graft. T-cell depletion was prompt and complete. The marrow was normocellular. Monocyte depletion was marked but short-lived peripherally and minimal centrally. A total of 13 of 14 patients are on monotherapy sirolimus. Eleven required a steroid taper for early lymphopenic rejection. There was one severe vascular rejection. Sirolimus was added from day one through day 21 in patients not receiving infliximab, and its timing did not influence onset of rejection.(25)
Shapiro et al reported a strategy designed to minimise the potentially antitolerogenic effect of immunosuppression by applying two therapeutic principles:(26)
Fifty adult patients received a kidney transplant over a period of five months; 10 (20%) of the 50 patients also had a pancreas transplant, and 21 (42%) had an infusion of donor-specific bone marrow cells (3–8¥10(8)/kg). The follow-up ranged from three to six months. One patient died peri-operatively of an anaesthetic error. A Schwartzman reaction was seen in one patient, who later developed cortical necrosis. Ninety-four per cent of the patients had functioning kidneys, with mean serum creatinine of 1.8±1.2mg/dl. Rejection was commonly observed but responded well to one or two steroid boluses and/or a short course of a second agent. Seventy per cent of the patients had been weaned to doses ranging from every other day to once-weekly tacrolimus.
There are a number of successful strategies for reducing steroid usage in solid organ transplant recipients. While steroid withdrawal has been achieved successfully, steroid avoidance is the goal of more recent studies. With the successful introduction of several new immunosuppressive agents in the last decade, steroid avoidance may well be routinely achievable in the majority of patients. T-cell depletion strategies followed by monotherapy may be particularly attractive. This approach has the added advantage of near-tolerance induction by clonal exhaustion–deletion, although clinical experiences are still being evaluated, and the results may vary depending on the organ involved.