Immunosuppression in Cardiac Transplantation
http://www.100md.com
《新英格兰医药杂志》
In this issue of the Journal, Hershberger and colleagues present data from a multicenter trial of immunosuppression in cardiac-transplant recipients showing that daclizumab, a monoclonal antibody against the interleukin-2 receptor, reduces the overall incidence of cellular rejection.1 In discussing these data, I will focus on three areas: the role of and difficulty in carrying out randomized, controlled trials of immunosuppressive regimens in heart-transplant recipients, the end points one is forced to choose in performing such studies, and the clinical relevance of the primary end point of cellular rejection, as well as the possibility of a negative outcome.
Any group that successfully conducts a randomized clinical trial involving heart-transplant recipients should be acknowledged and commended. The first human-to-human cardiac transplantation was performed in 1967. The results of the first randomized, controlled clinical trial of immunosuppression among heart-transplant recipients were published in 1998,2 more than 30 years later. Since 1998, three other moderate-sized randomized, controlled trials of immunosuppression among heart-transplant recipients, including that by Hershberger et al., have been carried out.1,3,4
Why has it been so difficult to conduct what would seem to be logical trials, given the dramatic expansion of our clinical immunosuppression armamentarium? The most obvious reason is the relatively small number of transplantations. In the United States, our average annual heart-transplantation volume is 2500, and this number has not changed appreciably since 1991.5 Also relevant is the fact that approximately 150 heart-transplantation centers perform this small number of operations. Studies are then focused on the larger cardiac-transplantation centers, all of which are sure that their particular immunosuppression cocktail is clearly the best. Hence, the negotiations among centers regarding which therapies are allowed and which are excluded can be complex. A good example of this is that in all three previously reported trials, approximately 50 percent of the centers used early antibody therapy and 50 percent did not, according to their local protocols. Furthermore, in centers using antibodies, some used monoclonal antibodies, and others polyclonal preparations. Finally, any sign of trouble, given the high stakes involved, resulted in removal of the patient from his or her assigned group, leading to an approximate rate of dropout or crossover of 30 percent in all of these studies. What was left was at best 200 and at worst 100 patients per group who could be evaluated, spread among 30 centers, with half the patients given additional and diverse antibody therapy.
The next important issue in trials assessing immunosuppression after heart transplantation is selecting the most appropriate end point. In 1992, I was involved in the early development of the trial by Kobashigawa et al. comparing mycophenolate mofetil with azathioprine.2 We decided that biopsy-proven rejection was an "objective" and frequent end point in heart transplantation and should probably be the primary end point of the study. The reasoning behind this decision was the realization that with one-year survival rates at that time approaching 85 percent, the number of patients required to endow a mortality trial with sufficient statistical power to detect a treatment effect would be prohibitive. Conversely, since the incidence of acute rejection on surveillance biopsy was, depending on the center, 50 percent or higher in the first year, the enrollment of small numbers of patients would yield a sufficient number of end points to have an appropriately powered study. The assumption, of course, was that patients in whom rejection occurred might have a poorer overall outcome than those without rejection. Ultimately, in this study, there was a lower rate of death among patients taking mycophenolate mofetil than among those who received azathioprine (6 percent vs. 11 percent), and there was also a lower rate of treated episodes of rejection (66 percent vs. 74 percent), but clearly, many more patients had an episode of rejection than died. In all the subsequent studies, including the study by Hershberger et al., the rate of the primary end point (usually a composite of histologic rejection, hemodynamic compromise, and death) differed significantly between groups, but what truly differed between groups was the incidence of histologic rejection. In the studies by Eisen et al.3 and Kobashigawa et al.,4 the mortality rates were similar. In the trial by Hershberger et al., the mortality trends are worrisome.
All of us who regularly treat heart-transplant recipients know that most patients with histologic rejection have absolutely no change in allograft function, and the histologic findings are usually easily reversed with an augmentation of immunosuppressive therapy. Conversely, many patients with acute hemodynamic compromise do not have the classic findings of acute rejection. Although rejection is a documented cause of death primarily in the first year after transplantation, it has become exceedingly unusual. In a recent multicenter survey of almost 7300 patients, rejection was a cause of death at one year in only 2 percent.6 Contrast that to the incidence of histologic rejection of 50 to 75 percent in these other studies. Clearly, the results of the multicenter, randomized trials completed to date cannot provide support for a direct relationship between histologic rejection and outcome.
Hershberger and colleagues found that the risk of their primary composite end point of histologic rejection, graft dysfunction, a second transplantation, or death within six months was reduced by treatment with the monoclonal antibody daclizumab. As in prior studies, only 81 percent of the patients received four or five doses of the assigned study drug. The efficacy benefit of the active drug was solely accounted for by the histologic-rejection component of the composite end point. A worrisome finding was that overall mortality was greater in the daclizumab group than in the placebo group (6.5 percent vs. 3.2 percent) at six months, and this trend continued at one year (P=0.11). Also worrisome was the higher incidence of graft dysfunction in the daclizumab group. The authors did an admirable job in trying to sort out reasons for this apparent increase in mortality in the active-treatment group and discovered that most of the deaths were due to infections and occurred in patients who received both daclizumab and a second monoclonal or polyclonal preparation of antilymphocyte antibody. They conclude that although daclizumab reduces the risk of histologic rejection, "concurrent or anticipated use of cytolytic therapy with daclizumab should be avoided."
How does one integrate the results of this and prior heart-transplantation studies into clinical practice? The vast majority of episodes of histologic rejection are treated effectively with a moderate increase in immunosuppressive therapy. Most certainly, the worst part of having to treat rejection is telling the patient about the diagnosis and dealing with the patient's anxiety. I personally do not know how I can "anticipate" a patient's need for antilymphocyte-antibody therapy in order to identify those I can and those I cannot treat with daclizumab, and I would certainly be reticent to limit my options, should the episode of rejection be more serious and require antilymphocyte-antibody therapy.
Finally, what am I willing to trade for this reduction in the risk of histologic rejection? Everolimus therapy reduces the risk of histologic rejection but increases the risk of renal dysfunction,3 a trade-off that may be acceptable, since the risk of coronary disease after transplantation may be reduced. Tacrolimus reduces the risk of histologic rejection but increases the incidence of diabetes, which may also be considered an acceptable trade-off. In contrast, on the basis of the data from the current trial of daclizumab, I am certain that I would not be willing to trade even a small increase in the risk of death from infection for a reduction in the risk of histologic rejection.
Source Information
From St. Luke's Medical Center, Milwaukee.
References
Hershberger RE, Starling RC, Eisen HJ, et al. Daclizumab to prevent rejection after cardiac transplantation. N Engl J Med 2005;352:2705-2713.
Kobashigawa J, Miller L, Renlund D, et al. A randomized active-controlled trial of mycophenolate mofetil in heart transplant recipients. Transplantation 1998;66:507-515.
Eisen HJ, Tuzcu EM, Dorent R, et al. Everolimus for the prevention of allograft rejection and vasculopathy in cardiac transplant recipients. N Engl J Med 2003;349:847-858.(Jeffrey D. Hosenpud, M.D.)
Any group that successfully conducts a randomized clinical trial involving heart-transplant recipients should be acknowledged and commended. The first human-to-human cardiac transplantation was performed in 1967. The results of the first randomized, controlled clinical trial of immunosuppression among heart-transplant recipients were published in 1998,2 more than 30 years later. Since 1998, three other moderate-sized randomized, controlled trials of immunosuppression among heart-transplant recipients, including that by Hershberger et al., have been carried out.1,3,4
Why has it been so difficult to conduct what would seem to be logical trials, given the dramatic expansion of our clinical immunosuppression armamentarium? The most obvious reason is the relatively small number of transplantations. In the United States, our average annual heart-transplantation volume is 2500, and this number has not changed appreciably since 1991.5 Also relevant is the fact that approximately 150 heart-transplantation centers perform this small number of operations. Studies are then focused on the larger cardiac-transplantation centers, all of which are sure that their particular immunosuppression cocktail is clearly the best. Hence, the negotiations among centers regarding which therapies are allowed and which are excluded can be complex. A good example of this is that in all three previously reported trials, approximately 50 percent of the centers used early antibody therapy and 50 percent did not, according to their local protocols. Furthermore, in centers using antibodies, some used monoclonal antibodies, and others polyclonal preparations. Finally, any sign of trouble, given the high stakes involved, resulted in removal of the patient from his or her assigned group, leading to an approximate rate of dropout or crossover of 30 percent in all of these studies. What was left was at best 200 and at worst 100 patients per group who could be evaluated, spread among 30 centers, with half the patients given additional and diverse antibody therapy.
The next important issue in trials assessing immunosuppression after heart transplantation is selecting the most appropriate end point. In 1992, I was involved in the early development of the trial by Kobashigawa et al. comparing mycophenolate mofetil with azathioprine.2 We decided that biopsy-proven rejection was an "objective" and frequent end point in heart transplantation and should probably be the primary end point of the study. The reasoning behind this decision was the realization that with one-year survival rates at that time approaching 85 percent, the number of patients required to endow a mortality trial with sufficient statistical power to detect a treatment effect would be prohibitive. Conversely, since the incidence of acute rejection on surveillance biopsy was, depending on the center, 50 percent or higher in the first year, the enrollment of small numbers of patients would yield a sufficient number of end points to have an appropriately powered study. The assumption, of course, was that patients in whom rejection occurred might have a poorer overall outcome than those without rejection. Ultimately, in this study, there was a lower rate of death among patients taking mycophenolate mofetil than among those who received azathioprine (6 percent vs. 11 percent), and there was also a lower rate of treated episodes of rejection (66 percent vs. 74 percent), but clearly, many more patients had an episode of rejection than died. In all the subsequent studies, including the study by Hershberger et al., the rate of the primary end point (usually a composite of histologic rejection, hemodynamic compromise, and death) differed significantly between groups, but what truly differed between groups was the incidence of histologic rejection. In the studies by Eisen et al.3 and Kobashigawa et al.,4 the mortality rates were similar. In the trial by Hershberger et al., the mortality trends are worrisome.
All of us who regularly treat heart-transplant recipients know that most patients with histologic rejection have absolutely no change in allograft function, and the histologic findings are usually easily reversed with an augmentation of immunosuppressive therapy. Conversely, many patients with acute hemodynamic compromise do not have the classic findings of acute rejection. Although rejection is a documented cause of death primarily in the first year after transplantation, it has become exceedingly unusual. In a recent multicenter survey of almost 7300 patients, rejection was a cause of death at one year in only 2 percent.6 Contrast that to the incidence of histologic rejection of 50 to 75 percent in these other studies. Clearly, the results of the multicenter, randomized trials completed to date cannot provide support for a direct relationship between histologic rejection and outcome.
Hershberger and colleagues found that the risk of their primary composite end point of histologic rejection, graft dysfunction, a second transplantation, or death within six months was reduced by treatment with the monoclonal antibody daclizumab. As in prior studies, only 81 percent of the patients received four or five doses of the assigned study drug. The efficacy benefit of the active drug was solely accounted for by the histologic-rejection component of the composite end point. A worrisome finding was that overall mortality was greater in the daclizumab group than in the placebo group (6.5 percent vs. 3.2 percent) at six months, and this trend continued at one year (P=0.11). Also worrisome was the higher incidence of graft dysfunction in the daclizumab group. The authors did an admirable job in trying to sort out reasons for this apparent increase in mortality in the active-treatment group and discovered that most of the deaths were due to infections and occurred in patients who received both daclizumab and a second monoclonal or polyclonal preparation of antilymphocyte antibody. They conclude that although daclizumab reduces the risk of histologic rejection, "concurrent or anticipated use of cytolytic therapy with daclizumab should be avoided."
How does one integrate the results of this and prior heart-transplantation studies into clinical practice? The vast majority of episodes of histologic rejection are treated effectively with a moderate increase in immunosuppressive therapy. Most certainly, the worst part of having to treat rejection is telling the patient about the diagnosis and dealing with the patient's anxiety. I personally do not know how I can "anticipate" a patient's need for antilymphocyte-antibody therapy in order to identify those I can and those I cannot treat with daclizumab, and I would certainly be reticent to limit my options, should the episode of rejection be more serious and require antilymphocyte-antibody therapy.
Finally, what am I willing to trade for this reduction in the risk of histologic rejection? Everolimus therapy reduces the risk of histologic rejection but increases the risk of renal dysfunction,3 a trade-off that may be acceptable, since the risk of coronary disease after transplantation may be reduced. Tacrolimus reduces the risk of histologic rejection but increases the incidence of diabetes, which may also be considered an acceptable trade-off. In contrast, on the basis of the data from the current trial of daclizumab, I am certain that I would not be willing to trade even a small increase in the risk of death from infection for a reduction in the risk of histologic rejection.
Source Information
From St. Luke's Medical Center, Milwaukee.
References
Hershberger RE, Starling RC, Eisen HJ, et al. Daclizumab to prevent rejection after cardiac transplantation. N Engl J Med 2005;352:2705-2713.
Kobashigawa J, Miller L, Renlund D, et al. A randomized active-controlled trial of mycophenolate mofetil in heart transplant recipients. Transplantation 1998;66:507-515.
Eisen HJ, Tuzcu EM, Dorent R, et al. Everolimus for the prevention of allograft rejection and vasculopathy in cardiac transplant recipients. N Engl J Med 2003;349:847-858.(Jeffrey D. Hosenpud, M.D.)