Arthur J Matas
Professor of Surgery
Department of Surgery
University of Minnesota
Supported by NIH DK13083
The ultimate goal of kidney transplant recipient care is to achieve 100% long-term patient and graft survival without drug-related morbidity. Studies have clearly demonstrated that a successful transplant provides better longevity and quality of life than maintenance dialysis.(1,2 )But transplant success is limited by drug-related morbidity, either due to too much overall immunosuppression or to drug-specific side-effects. All post-transplant immunosuppressive protocols are designed to limit total immunosuppression. Drug-specific minimisation protocols are designed to limit the side-effects of a specific agent. To date, drug minimisation strategies have tried to address two separate problem areas:
- Calcineurin inhibitor (CNI) toxicity.
- Prednisone-related side-effects.
Of note, most CNI-minimisation trials have continued to use prednisone, and most prednisone-minimisation trials have continued to use CNIs.
CNIs have been responsible for significant improvement in short-term transplant outcome. But there are concerns that these agents may have a detrimental long-term impact. In fact, data suggest that although there have been significant improvements in short-term patient and graft survival in the last decade, there is no associated improvement in long-term graft survival.(3) The two main causes of late graft failure are patient death (with a functioning graft) and chronic allograft nephropathy.(4) CNIs may contribute to the former by increasing the incidence of hyperlipidaemia, hypertension or diabetes, and to the latter by a direct nephrotoxic effect. Nankivel et al recently reported on sequential protocol biopsies in a cohort of kidney–pancreas transplant recipients.(5) At 10 years, severe chronic allograft nephropathy was present in over 50% of recipients. The authors felt that the major cause of the late histological deterioration was CNI-induced injury. It is difficult to generalise these results to the overall transplant population for several reasons. First, patients were transplanted at a time when higher levels of CNIs were used. Secondly, there are now numerous recipients who have taken CNIs over 20 years and have not had any deterioration in graft function. The data, however, illustrate an important concern.
The major goal of CNI minimisation trials is to decrease this drug-specific nephrotoxicity. These trials have taken three separate forms:
- Complete avoidance of CNIs.
- Initial use of CNIs followed by protocol withdrawal.
- Withdrawal of CNIs in recipients with allograft dysfunction.
Complete avoidance of CNIs
A number of two-drug CNI-free immunosuppressive drug combinations have been tried, all including concomitant interleukin-2 (IL-2) inhibitor, mycophenolate mofetil (MMF), polyclonal antibody (Thymoglobulin)/MMF, sirolimus/ azathioprine (AZA) and sirolimus/MMF.(6,7) All have been associated with unacceptably high rates of acute rejection. In contrast, when sirolimus/MMF (and prednisone) have been combined with antibody induction, either IL-2 inhibitor or Thymoglobulin, the acute rejection rates have been low, and other outcome measures (patient and graft survival and renal function) have been excellent.(8) However, because of the high blood level of sirolimus that was targeted in these studies, sirolimus-specific side-effects (eg, hyperlipidaemia, thrombocytopenia) have been common.
CNI withdrawal by protocol
An alternative approach has been to use CNIs for the first few months (usually in combination with sirolimus and prednisone) and then to withdraw the CNI.(9–11) The goal is to provide the immuno‑suppressive benefit of the CNI when most required (ie, early post-transplant) and simultaneously allow lower sirolimus doses to be used (which will lead to fewer sirolimus-specific side-effects).
Long-term follow-up of a prospective randomised trial has shown the potential of this approach.(9) Recipients in a European study were treated with sirolimus/ciclosporin/ prednisone. At three months, the patients were randomised to stay on the triple-drug combination or to have their ciclosporin stopped, and targeted sirolimus blood levels increased. For the first three months after randomisation, there was an increased rate of acute rejection in the sirolimus/prednisone arm. However, at four years, there was significantly better graft survival and death-censored graft survival in the sirolimus/prednisone group. Most importantly, the sirolimus/prednisone arm had significantly better serum creatinine levels, calculated glomerular filtration rate (GFR) and slope of GFR vs time.(9)
One caveat is that these studies were carried out in low-immunological-risk recipients. Additional studies are necessary to see whether these protocols can be used in a higher-risk cohort.
CNI conversion or dose reduction in recipients with reduced graft function
There have been numerous reports on CNI conversion or dose reduction for recipients with graft dysfunction.(12–17) To date, no randomised controlled trials have been carried out. Thus, it remains unclear which patients will benefit from such protocols.
It is clear that, for most recipients taking CNIs, serum creatinine levels temporarily fall when CNIs are stopped. But, in some recipients, stopping the CNIs (or conversion to another immunosuppressive drug) leads to an acute rejection episode and further deterioration of graft function. In other recipients, the CNIs have probably been stopped too late in the course, and thus graft deterioration continues. Igidi et al have noted that recipients are more likely to benefit from conversion to sirolimus if serum creatinine level is >3mg/dl (p=0.024).(16) Diekmann et al found that recipients with low proteinuria (<800mg/day) benefited from conversion, whereas those with more proteinuria did not.(18)
The primary goal of prednisone-minimisation trials is to avoid prednisone-related side-effects, and to do so without increasing the rates of acute rejection, chronic graft dysfunction or graft loss.
Historically, late prednisone withdrawal in recipients taking ciclosporin/prednisone or ciclosporin/AZA/prednisone has failed. Numerous studies and two meta-analyses have shown that, in selected recipients at low immunological risk, prednisone withdrawal (more than three months posttransplant) is associated with increased rates of acute rejection and graft loss.(19,20) Importantly, a Canadian multicentre randomised study of prednisone withdrawal at three months post-transplant showed no impact up to about 500–600 days following randomisation. However, after that point, there was significantly worse graft survival in the withdrawal arm.(21) More recently, two large prospective randomised multicentre studies of late prednisone withdrawal in recipients taking ciclosporin/MMF/prednisone again showed an increased rate of acute rejection in the prednisone withdrawal arm.(22,23)
In contrast to these late withdrawal protocols, some centres are now utilising protocols incorporating complete avoidance of prednisone or rapid prednisone discontinuation (in the first week).(24–35) Various protocols have been utilised, usually incorporating an antibody (IL-2R inhibitor, Thymoglobulin or Campath), a CNI (ciclosporin or tacrolimus) and either MMF or sirolimus. Many of these combinations have been associated with low rejection rates. However, in general, when Thymoglobulin or Campath have been used, rejection rates have been somewhat lower than with IL-2 inhibitors. Of importance, these protocols are associated with low acute rejection rates in decreased donor recipients and recipients with increased immunological risk (children, African Americans and recipients with high levels of preformed antibodies).(27,28,31)
Recently published randomised studies have now clearly demonstrated that rapid discontinuation of prednisone can be achieved without increasing the rate of acute rejection. Ter Meulin et al randomised 381 first- or second-transplant recipients to tacrolimus/MMF/IL-2R (two doses)/prednisone (two doses) versus tacrolimus/MMF/prednisone.(33) At
12 months, there was no difference in the rates of biopsy-proven acute rejection, in the severity of acute rejection and in patient survival, graft survival or renal function. Rostaing et al randomised 538 adult recipients from 47 transplant centres to tacrolimus/MMF/IL-2R (two doses).(34) At six months, there was no difference in the rate of acute rejection or steroid-resistant rejection and in the time to acute rejection, and no difference in patient survival, graft survival or renal function.
We have now treated over 600 kidney transplant recipients with a protocol incorporating rapid discontinuation of prednisone. We utilised Thymoglobulin induction, five days of prednisone, ciclosporin or tacrolimus, and MMF or sirolimus. Our five-year actuarial data show low acute rejection rates and good long-term patient and graft survival.(35) We have not seen the late deterioration of graft survival that was noted in the Canadian multicentre study of late prednisone withdrawal. Of importance, compared with historical controls at our institution, patients treated with rapid discontinuation of prednisone had significantly less cytomegalovirus (CMV) infection, cataracts, avascular necrosis, post-transplant diabetes, fractures and nonpost-transplant lymphoproliferative disease malignancy. As reported by Sarwal et al, children on prednisone-free maintenance immunosuppression had better linear growth and better compliance with the immunosuppressive protocol.(28)
We are currently doing a prospective randomised trial of ciclosporin/MMF vs tacrolimus/sirolimus in this rapid discontinuation of prednisone protocol. To date, all study arms have equivalent patient and graft survival and acute rejection rates.(29) It will be important to learn whether any one specific maintenance regimen provides superior long-term outcome in recipients not taking prednisone.
CNI- and prednisone-minimisation trials
Each of the above protocols has attempted to minimise the complications of one of the immunosuppressive drugs. If long-term results remain good, future trials may attempt to combine the benefits of CNI minimisation and prednisone minimisation. Shapiro et al have reported successful use of a protocol incorporating antibody induction and CNI (tacrolimus). Patients on tacrolimus monotherapy were then weaned to a dosing of two or three times per week. It will be important to see the long-term results of this approach.(36)
Both CNI- and steroid-minimisation trials have been successful. It will be important to determine whether these protocols can be applied to the entire transplant population or they are effective only in specific subgroups.
- N Engl J Med 1985;312:553-9.
- N Engl J Med 1999;341:1725-30.
- Am J Transplant 2004;4:1289-95.
- Kidney Int 2002;62:704-14.
- N Engl J Med 2003;349:2326-33.
- Am J Kidney Dis 2001;38 Suppl:S16-21.
- Transplantation 2001;71:1282-7.
- Am J Transplant 2004;4:1776-85.
- Transplant Int 2005;18:22-8.
- Transplantation 2003;75:1404-8.
- Transplantation 2002;74:1560-7.
- Transplantation 1997;64:1706-10.
- Transplantation 1998;65:661-7.
- Transplantation 2005;79:304-9.
- Transplantation 2005;79:466-75.
- Transplant Proc 2003; Suppl:131S-7S.
- Kidney Int 2001;59:1567-73.
- Am J Transplant 2004;4:1869-75.
- J Am Soc Nephrol 1993;4:1300-5.
- J Am Soc Nephrol 2000;11:1910-7.
- CMAJ 1992;147:645-57.
- Transplantation 2000;70:1352-9.
- Transplantation 1999;68:1865-74.
- Transplantation 2001;71:1089-90.
- Transplantation 2001;72:845-50.
- Transplant Proc 2001;33:1011-2.
- Transplantation 2002; 73:169-77.
- Transplantation 2003;76:1331-9.
- Kandaswamy R, et al. A prospective randomized trial of steroid-free maintenance regimens in kidney transplant recipient – an interim analysis. Am J Transplant In press 2005.
- Am J Transplant 2004;4:803-10.
- Transplantation 2004;78:1397-9.
- Ann Surg 2004;240:510-6.
- Am J Transplant 2004;4:803-10.
- Transplantation 2005;79:807-14.
- Matas AJ, et al. Prednisone-free maintenance immunosuppression – a 5-year experience. Am J Transplant In press 2005.
- J Am Coll Surg 2005;200:505-15.