|Home | About | Journals | Submit | Contact Us | Français|
Steroid-free immunosuppression in kidney transplantation has been gaining popularity over the past decade, as documented by a continuous and steady rise in the number of kidney transplant patients discharged on steroid-free regimens. This increased interest in steroid-free immunosuppression is fueled by the recognition that half of transplant loss is related to patient death due to cardiovascular disease and/or infectious complications and that the long-term use of steroids contributes to such elevated cardiovascular morbidity and mortality. The availability of newer and more potent immunosuppressive agents has furthered such interest. Many clinical trials over the past two decades have demonstrated the feasibility of steroid-free regimens, at the expense of a slight increase in the rate of acute rejection, which is an important end point in any clinical trial of relatively short duration. The largest epidemiological study to date has reassured the transplant community that the selective use of steroid-free immunosuppression in kidney transplant patients provides no inferior outcome in patient and graft survival at intermediate term. Steroid-free regimens have the potential to improve cardiovascular risk profile. The challenges that remain are to identify the subset of kidney transplant patients who may not benefit from steroid-free immunosuppression and to demonstrate the survival advantage of steroid-free immunosuppresion in suitable kidney transplant candidates.
Steroids have been a part of any immunosuppressive regimen since the beginning of kidney transplantation because of their anti-inflammatory and immunosuppressive property, in addition to their ability to reverse acute rejection.1–4 However, it was soon recognized that the long-term use of steroids was associated with a wide range of adverse effects, even at a relatively low dose.5 These adverse effects include worsening hypertension, worsening dyslipidemia, new-onset diabetes, osteoporosis, avascular necrosis, susceptibility to infection, and so on. Many of these are responsible for a persistently elevated cardiovascular risk burden in kidney transplant patients, and therefore, excessive mortality.6 In fact, the recognition of steroid-associated adverse effects has perpetuated the effort of sparing steroids from immunosuppression regimens over the past three decades.7–9 With the introduction of cyclosporine (CsA) in the early 1980s, tacrolimus (Tac) and mycophenolate mofetile (MMF) in the mid 1990s, and the availability of various induction agents, the interest in sparing steroids has grown more intense. This has led to multiple reports of experiences with randomized and/or no randomized clinical trials.
The very initial systematic attempt to wean kidney transplant patients off steroids came during the late 1980s and early 1990s when CsA became widely used as the main immunosuppressant, either alone or in combination with azathioprine (AZA). Griffin et al.10 conducted one of the very early randomized and controlled clinical trials comparing CsA monotherapy with CsA plus low-dose maintenance steroids and found an equal renal outcome but a higher incidence of CsA nephrotoxicity in the CsA monotherapy group. Patients in the CsA monotherapy group had significantly lower infectious and cosmetic complications. However, more than 50% of patients eventually returned to a steroid-containing regimen. Schulak et al.9 presented their experience with early steroid withdrawal in patients treated with anti-lymphocyte globulin, CsA, and AZA. Although there were no differences in patient and graft survival, patients who had steroids withdrawn did experience more frequent and severe acute rejection episodes. Sinclair11 reported the experience from a multicenter randomized trial comparing low-dose steroids with placebo in 523 kidney transplant patients receiving a CsA maintenance regimen. Although graft survival was similar between the two groups within the first 500 days, after more than 6 years of follow-up, graft survival was significantly worse in patients on no steroids. There was also poorer adherence to the test drug in the placebo group, with more than half of the patients eventually dropping out of the assigned study drug protocols.
Two meta-analyses performed by Hricik et al.8 in the early 1990s and by Kasiske et al.12 in early 2000 summarized the experience of steroid withdrawal, mostly with CsA-based immunosuppression during the late 1980s and early 1990s. The results of thousands of patients from 7 different studies in one case and from 10 different studies in another confirmed the findings of individual studies. Steroid withdrawal was associated with a greater proportion of patients experiencing acute rejection, and the relative risk for graft failure was 40% higher in patients withdrawn from steroids when follow-up extended to 5 years and beyond.
Thus, these early observations clearly showed an increased incidence in acute rejection in patients withdrawn from or not receiving steroids. As high as 50% of patients initially withdrawn from steroids had to return to steroids. This could represent a serious drawback of steroid withdrawal, as the increased long-term risk for graft failure may counterbalance any potential benefit of steroid withdrawal. As a consequence of these findings, the interest in steroid-free immunosuppression was dampened among many transplant physicians and surgeons in the transplant community.
With the introduction of more potent immunosuppressive agents, such as MMF, Tac, sirolimus, and newer induction agents such as anti-thymocyte globulin and anti-IL2 receptor antibodies in the 1990s, the overall incidence of acute rejection has decreased during this time span in patients treated with steroid-containing immunosuppressive regimens.13–16 At the same time, there has been a renewed enthusiasm in steroid-free immunosuppression. This is shown by the continuous increase in the numbers of patients on a steroid-free immunosuppressive regimen at the time of discharge from the hospital after transplant surgery (Figure 1).17
Ahsan et al.18 presented the results of a multicenter, randomized, and double-blind steroid withdrawal trial at 3 months after transplant, in patients receiving a combination of CsA and MMF. Although patient and graft survival were no different at 1 year, there was a significant increase in acute rejection or treatment failure (9.8% with steroids and 30.8% without steroids, P<0.001). In particular, the risk of treatment failure was much higher for blacks (39.6%) than for non-blacks (16.0%). Furthermore, patients withdrawn successfully from steroids had a higher serum creatinine. Vanrenterghem et al.19 conducted a similar trial in Europe involving 500 patients, with half of them having steroids discontinued at 3 months. The rate of biopsy-proven acute rejection was higher in the steroid withdrawal group at 6 months (23 vs 14%, respectively, P=0.008) and 12 months (25 vs 15%, respectively), although the graft loss at 12 months was similar (5 vs 4%, respectively). Neither of these studies involved a universal use of antibody induction.
In the late 1990s, it was increasingly recognized that late posttransplant steroid withdrawal was associated with an increased incidence in acute rejection in some recipients. It was also recognized that the chronic use of steroids may have sensitized lymphocytes by upregulating the expression of cytokine receptors in the surface membrane so that withdrawal in a late stage would have resulted in the stimulation of those lymphocytes.20 Finally, late steroid withdrawal may not prevent steroid-related complications, at least in some aspects. With these considerations in mind, Matas et al. from the University of Minnesota conducted a pilot study of rapid discontinuation of steroids within the first 5 days. Thymoglobulin was the induction agent; CsA and MMF were maintenance medications.21 Compared with the historical controls of CsA/AZA/P and CsA/MMF/P regimens without the use of an induction agent, there was no difference in 1-year patient and graft survival. Only 10% of patients had a biopsy-proven acute rejection that was not different from that of historical controls. However, 25% of patients returned to steroids 5 to 17 months after transplant.
In a multicenter, randomized early steroid withdrawal trial reported by Vincenti et al., anti-IL2 receptor antibody was used as the induction agent, followed by CsA and MMF as the maintenance regimen. It yielded results comparable with those of the steroid-containing regimen regarding the incidence of acute rejection at 12 months (20 vs 16%, non significant).22 Patient and graft survival were similarly good in both groups, but 28% of patients from the steroid withdrawal group returned to steroid-containing regimens at 6 months after transplantation.
Most recently, the Freedom trial compared steroid-free and early steroid withdrawal with a standard chronic steroid-containing regimen using an anti-IL2 receptor antibody as the induction agent and a combination of CsA/MPA as maintenance therapy.23 At 12 months, biopsy-proven acute rejection occurred significantly more often and earlier in steroid-free and steroid withdrawal groups (31.5% in steroid-free, 26.1% in steroid withdrawal and 14.7% in standard steroid groups, P=0.007 and P=0.046 for steroid-free and steroid withdrawal groups vs standard steroid group, respectively). Patient and graft survival were no different among the groups. Renal function was no different.
On the other hand, the experience using a combination of CsA and sirolimus suggested that lower rates of acute rejection could be achieved without the use of steroids. Rajab et al.24 reported their experience with this combination in 301 primary kidney transplant patients. The first year incidence of acute rejection was 4.9% in the steroid-free group compared with 9.4% in a historical cohort of primary kidney transplant patients with a steroid-containing regimen (P<.05). Patient and death censored graft survival at 1 year were 93.1% and 98.1%, respectively.
Since the introduction of Tac into the armamentarium of maintenance immunosuppressive medications, several clinical trials have shown a lower rate of acute rejection compared with CsA regimens.25–28 Rostaing et al.29 reported the outcome of a multicenter European study comparing the regimens of Tac and MMF with and without maintenance steroids. After 6 months of follow-up, the incidence of acute rejection was identical at 16.5% in both groups, involving a total of 538 patients. The severity of acute rejection was also identical. However, the steroid withdrawal group received induction therapy with an anti-IL2 receptor antibody, whereas the steroid-containing group did not. The major drawback of this study was a too short follow-up time and the differential use of induction agent.
After initially reporting their pilot experience with rapid steroid withdrawal in CsA- and MMF-treated patients, Matas et al.30,31 further extended their experience into a large cohort of patients to include Tac and sirolimus. In spite of not being randomized and uncontrolled, their experience clearly showed that, with the use of induction therapy, early steroid withdrawal produced acceptable short- and intermediate-term outcomes with regard to both patient and graft survival, as well as the incidence of acute rejection during the 5-year study period. Furthermore, compared with historical controls, steroid withdrawal patients experienced significantly fewer complications related to the long-term use of steroids such as cataracts, new-onset diabetes, avascular necrosis, and fractures.
Kumar et al.32,33 reported their experience with steroid withdrawal 2 days after transplant, using an anti-IL2 receptor antibody as the induction agent and Tac, combined either with MMF or sirolimus, as the maintenance regimen. More than half of their patients were African-American. A surveillance biopsy was performed on all their patients at 1, 6, 12, 24, and 36 months after transplant. Patient and graft survival were similar at 3 years. Clinical acute rejection was diagnosed in 14% of steroid-treated and in 16% of steroid withdrawal patients at 1 year (P=NS). Subclinical acute rejection and chronic allograft nephropathy, diagnosed by surveillance biopsies, were higher in both groups. However, no difference existed between steroid-treated and steroid withdrawal patients over 3 years. Weight gain and incidence of new-onset diabetes were significantly lower in steroid withdrawal patients (change in body mass index 0.3 vs 4.4, respectively, over 3 years, P=0.04, and new-onset diabetes (NODAT) 4 vs 21%, respectively, at 3 years, P<.01). Similarly, a serious infection requiring hospitalization was lower in the steroid withdrawal group (18 vs 35%, P=0.05).
African-American kidney transplant patients are traditionally considered to be at high risk for immunological complications. Steroid withdrawal in this group of patients was typically complicated by a higher rejection rate and poorer graft survival.18,34 However, African-American patients may benefit more from steroid withdrawal-related improvement in cardiovascular risk reduction, such as a reduced incidence of new-onset diabetes and weight gain. By using the combination of Tac and sirolimus, Hricik et al.35 have shown that steroids could be successfully withdrawn with a low incidence of acute rejection. Gruber et al. achieved a similar outcome using a combination of Tac and MMF.36
More recently, Woodle et al.37 published their experience of early steroid withdrawal in patients receiving Tac and MMF. Their study involved 386 patients from multiple centers, randomized to rapid steroid withdrawal (191 patients) and to chronic low-dose steroids (195 patients) with a 5-year follow-up. They excluded highly sensitized patients (pPRA≥50% and/or cPRA≥25%). Steroid withdrawal was completed within 7 days after transplant. Induction was used but the choice of agents, an anti-IL2 receptor antibody or thymoglobulin, was determined by individual centers. After a 5-year follow-up, they found no difference between the two study groups in the composite primary end point of death, graft loss, and moderate–severe acute rejection. Biopsy-proven acute rejection, however, was higher in the steroid withdrawal group (17.8 vs 10.8%, respectively, over 5 years, Kaplan–Meier analysis, log-rank P=0.042). When sub-analysis was performed to determine the effects of the induction agents used, the magnitude in the difference of acute rejection was higher in patients who received an anti-IL2 receptor antibody (24.2% in steroid withdrawal and 11.9% in chronic steroids use group, P=0.105) than in patients who received thymoglobulin (14.4% in steroid withdrawal and 10.3% in chronic steroid group, P=NS). Renal function was no different between the groups after 5 years. Serious infectious events were more common in chronic steroid users (16.4 vs 9.4%, P=0.04). Post hoc analysis did reveal a 5.8% increase in chronic allograft nephropathy in the steroid withdrawal group over 5 years (9.9 vs 4.1%, P=0.028). However, this last observation needs to be interpreted cautiously, as neither baseline nor protocol biopsies were part of the study protocol. Thus, a possibility of preexisting donor disease and/or overdetection bias could not be excluded, as a greater proportion of patients from the steroid withdrawal group underwent biopsy (46.1 vs 32.8%, respectively, P=0.009). The impact of chronic allograft nephropathy on the outcome remains unknown from this study. A longer follow-up may help further delineate this aspect.
Finally, Tan et al.38 reported their 3-year experience with steroid-free Tac monotherapy under alemtuzumab induction. They reported 3-year patient and graft survival at 93.3 and 86.3%, respectively. The incidence of acute cellular rejection was low within the first year (9.0%). However, the cumulative incidence of acute cellular rejection increased to as high as 24% within 3 years after transplantation. Late occurrence of acute cellular rejection, particularly within the second year, was especially detrimental to patient and graft survival. It is important to point out that, in addition to being steroid free, their protocol also incorporated a spaced weaning of Tac.
Two important questions remain with regard to all these clinical trials, randomized or not. First, what is the impact of the increased incidence of acute rejection in steroid withdrawal patients on long-term graft and patient survival? Second, are patients who experienced acute rejection systematically different from patients who did not experience acute rejection after steroid withdrawal? Many centers have established certain criteria for selecting patients for steroid-free immunosuppression. The criteria became more inclusive in some centers as they gained experience.23,31,33,39 The risk factors for acute rejection seemed similar for patients in steroid-free regimens as for those in steroid-containing regimens. Certain patients started on steroid-free regimens would have benefited from steroids once they experienced an episode of acute rejection.40,41
As of 2006, more than 32% of new kidney transplant patients in the United States were discharged from initial transplant without a steroid preparation in their recorded immunosuppressive medications. We performed an analysis using national registry data on the outcome of kidney transplant performed between 2000 and 2006.17 The study provided a big picture at the national level on the use of steroid-free immunosuppression and its impact on outcome. A total of 16,491 patients were discharged without steroids in their immunosuppressive regimens after transplant surgery. These patients seemed systematically different from those patients discharged with steroids in their immunosuppressive regimens, suggesting a selection process that governed the decision of placing a patient on steroid-free immunosuppression at individual centers. With the presence of such a selection bias, after a 4-year follow-up, graft survival and patient survival were comparable between patients discharged with and without steroids (Figures 2a and b). Adjusting for multiple confounders, Cox regression analyses showed a reduced risk for graft failure and patient death for patients discharged on steroid-free immunosuppression at 1 and 4 years. Furthermore, there was a wide variation in the adoption of steroid-free regimens across centers in the United States and such a variation seemed to influence the outcome. In fact, when transplant centers were grouped according to the percentage of patients discharged with steroids and used as a covariate replacing the indicator of individual steroid usage in the model, the risk reduction for graft failure is greatest in centers in which 35 to 80% of patients were discharged on steroid-free regimens. This again suggests the importance of patient selection for steroid-free regimens to achieve the desired outcome.
The most compelling rationale for steroid-free immunosuppression has been reducing cardiovascular disease risks to which kidney transplant patients are chronically exposed. 6,42 Several clinical trials have shown the potential for cardiovascular risk reduction, whereas others failed to show such a benefit. The mean plasma concentration of total cholesterol and triglyceride improved in some studies but not in others. This is possibly related to the choice of maintenance regimens other than steroids.18,19,23,24,29,37 Similarly, blood pressure control was improved and/or the amounts of antihypertensive medication were reduced in some but not all studies.18,19,23,43
New-onset diabetes, one of the common complications after kidney transplantation, is associated with increased major cardiovascular events and reduced patient survival.44–46 The diabetogenic effect of steroids has been well recognized since the early era of kidney transplantation.47–49 The use of calcineurin inhibitors, CsA and Tac, also contributes to the risk for new-onset diabetes, with Tac being more diabetogenic than CsA.44 As Tac is the most commonly used calcineurin inhibitor in a steroid-free regimen, the potential benefit of a steroid-free regimen on new-onset diabetes may be counterbalanced by the stronger diabetogenic effect of Tac. The majority of recent steroid-free clinical trials have shown some improvement in the incidence of new-onset diabetes and/or glucose control, although a small number of patients were involved in most of the studies.29,31,33,37 On a large scale, we have observed a reduced incidence of new-onset diabetes among kidney transplant patients discharged on steroid-free immunosuppression within the United States between 2004 and 2006, compared with those discharged with steroid-containing immunosuppressive regimens (15.5 vs 10.2%, respectively, Luan et al., unpublished data). Two recent clinical studies also suggested that steroid withdrawal resulted in a reduction in cardiovascular events and metabolic syndrome.50,51
Despite the progress made over the past three decades with steroid-free immunosuppression in kidney transplantation, there are still lingering questions and concerns. The fact that concerns many of us is that the incidence of acute rejection remains, in general, higher in patients receiving steroid-free immunosuppression than in those receiving steroid-containing regimens. Long-term graft survival in those who experienced acute rejection on a steroid-free immunosuppressive regimen remains unknown. The real impact of improvement in conventional cardiovascular disease risk factors on patient survival has yet to be shown. The fact that more than 50% of all kidney grafts were lost and more than 30% of all kidney transplant patients died within 10 years from the time of transplant in the United States underscores the importance of our effort in defining the best immunosuppressive regimens.52 The major challenge remains the identification of individual patients who may not benefit from a steroid-free immunosuppressive regimen. Until then, steroid-free immunosuppression should be considered as a standard of care only for a carefully selected group of patients.
The authors thank Marsha Salley for carefully reviewing the paper.
DISCLOSURE The authors declared no competing interests.