点击显示 收起
2005美国移植会议
American Transplant Congress 2005: 6th Annual Joint Meeting of the American Society of Transplant Surgeons and the American Society of Transplantation
2005年5月21－25日
美国华盛顿西雅图
May 21 - 25, 2005, Seattle, Washington
Induction Therapies for Kidney Transplantation
Daniel Brennan, MD, FACP
Induction therapy (short-term antibody therapy) continues to be the focus of intense research efforts in the quest for minimization or avoidance of steroids and calcineurin inhibitors as a means to improving long-term outcomes in renal transplantation. Several induction agents are available, including rabbit antithymocyte globulin (rATG, Thymoglobulin), the humanized anti-interleukin-2 receptor (IL-2R) antibodies basiliximab (BAS) and daclizumab (DAC), alemtuzumab (ALM), rituximab, and intravenous immunoglobulin (IVIG).rATG is used increasingly in children, living donors, and for immunosuppressive minimization. In a large pediatric, single-center study (n = 34), rATG induction was used with tacrolimus (TAC), mycophenolate mofetil (MMF), and corticosteroids (steroids).[1] During the first year, the rate of acute rejection was 8%, of graft failure 6%, and of patient survival 100%. Cytomegalovirus (CMV) infection was seen in 3%, and only 1 case of posttransplant lymphoproliferative disorder (PTLD) was seen. The acute rejection rate was substantially lower than the 30% rate reported by the North American Pediatric Renal Transplant Cooperative Study.
A single-center, 3-year, observational study of 214 live-donor recipients who received rATG induction and standard maintenance immunosuppression reported a 5% acute rejection rate in the first year.[2] Predicted 5-year patient and graft survival was better than the national data (96% rATG vs 93% national) as was graft survival (82% rATG vs 79% national). Predicted 5-year rejection-free graft survival was superior in related vs unrelated donors (96% vs 81%). rATG induction did not result in higher-than-expected rates of CMV infection (5%), malignancy (3%), or PTLD (.5%).
A large multicenter study enrolled 150 living-donor transplant recipients and compared the use of rATG induction with early steroid withdrawal with standard triple therapy without induction.[3] Despite early steroid withdrawal, acute rejection with rATG induction was 10% vs 20% with standard therapy. Only 1 patient in the rATG group developed CMV and none developed PTLD.
Positive cross-matches are associated with a higher risk of rejection and graft loss. Twenty-one cross-match-positive patients received rATG for a total dose of 6 mg/kg as well as low-dose IVIG 100 mg/kg preoperatively and 2 additional doses postoperatively.[4] Maintenance immunosuppression consisted of TAC, MMF, and prednisone. None had acute cellular rejection. Acute humoral rejection occurred in 10% with anti-HLA antibodies, treated successfully with plasmapheresis, IVIG, rituximab, and steroids. Graft survival was 90% with a mean follow-up of 5.7 months. Two graft losses were due to early vascular thrombosis and hemolytic uremic syndrome.
A retrospective single-center review of 159 deceased-donor renal allograft recipients compared 2-year outcomes in high- and low-risk recipients induced with BAS or rATG (high-risk, only) in combination with sirolimus (SRL), prednisone, and reduced doses of cyclosporine (CsA, C0 25-75 ng/mL) at 2 years post transplantation.[5] Acute rejection in rATG high-immune responders (group 1, n = 53) was 6% (P < .05) vs 22% for BAS-treated low-immune responders (group 2, n =77), and 24% for BAS-treated high-immune responders (group 3, n = 29). Two-year graft survival for rATG was 81% vs 83% for BAS low responders and 69% for BAS high responders. CMV disease occurred in only 2% of recipients, and there were no cases of PTLD.
Both rATG and DAC were used in a single-arm, prospective study of 94 high-risk recipients.[6] All received rATG 1 mg/kg/dose intraoperatively for an average of 6 doses, and DAC 1 mg/kg/dose intraoperatively every 2 weeks with an average of 3 doses; maintenance immunosuppression was with TAC, MMF, and steroids. With a median follow-up of 18.5 months, patient and graft survival were 98% and 95%, respectively. Three percent had delayed graft function and 3% had acute rejection. One patient had PTLD. This was not a randomized trial, and interpretation and applicability of this unconventional approach are unclear.
A small (n = 99) French, multicenter, open-label, randomized trial compared 2 doses of DAC (2 mg/kg) (n = 54) with rATG (n = 55) in patients maintained on CsA, MMF, and prednisone.[7] No CMV prophylaxis was given. At 1-year follow-up, patient and graft survival and biopsy-proven acute rejection (BPAR) did not differ between groups (17% vs 15%). The incidence of CMV infection (39% vs 51%) and ganciclovir-treated CMV infection (33% vs 43%) also did not differ. Rejection rates and CMV rates were more realistic and double those recently reported in similar studies from France comparing rATG with BAS.[8]
rATG was compared with BAS in 83 living- or deceased-donor recipients treated with a steroid-avoidance protocol.[9] All recipients received induction with rATG or BAS. Maintenance immunosuppression was with TAC and MMF. The average follow-up was 244 days. The average MMF dose was 17% higher in the BAS arm than in the rATG arm, and yet BPAR occurred in only 7.5% of the rATG group and 31% of the BAS group. The incidence of infectious complications was similar.
A retrospective review compared 432 deceased-donor renal transplant recipients receiving rATG (n = 233) 1.5 mg/kg/dose every other day for 2-3 doses or BAS (n = 199) 20 mg intravenously on postoperative days 0 and 4.[10] More rATG patients had panel-reactive antibody levels > 30% (8% vs 2%, P = .005) and cold ischemia time > 36 hours (41% vs 25%, P = .001.), but 1-year BPAR did not differ (12% with rATG vs 16% with BAS). At 3 years, patient survival was 84% in the rATG group and 72% in the BAS group (P = .03).
A single-center, randomized trial compared 3 different antibody-induction agents in 90 first renal transplant recipients from cadaver donors (30 patients per arm): rATG, ALM, and DAC.[11] Maintenance immunosuppressives consisted of TAC and MMF in all 3 arms as well as prednisone in the rATG and DAC arms. At median follow-up of 12 months, acute rejection rates were similar -- 5%, 5%, and 4%, respectively -- in arms 1, 2, and 3. No PTLD occurred and infectious complication rates were similar. This study would have been more useful and interpretable if all 3 arms had continued or discontinued steroids. By design, the groups are not comparable.
IL-2R antagonists reduce rejection compared with placebo in triple-therapy or dual-therapy regimens, but whether they are potent enough to allow for immunosuppressive immunization is unclear. A multicenter, international (4 centers in 3 European countries), prospective study randomized patients to standard therapy with CsA (trough levels 150-250 ng/mL) plus MMF (2 g/day) vs low-dose CsA (defined as 50% of the standard trough levels ranging from 75 to 125 ng/mL) plus DAC (2 mg/kg DAC first dose, followed by 4 additional doses of 1 mg/kg every 2 weeks) and MMF (2 g/day).[12]
Steroids were tapered identically in both groups. Patient and graft survival rates were similar in 121 (59 DAC and 62 standard) patients with a follow-up of at least 6 months. The incidence of BPAR was 7% following DAC induction vs 27% in the standard group (P = .0035).
In ABO-incompatible kidney transplantation, antidonor blood-type antibody titers can be reduced by double filtration plasmapheresis (DFPP). A new protocol, including rituximab, DFPP, and splenectomy, in 7 nonresponders, defined as those whose titers could not be reduced below 1:16 after 3 sessions of DFPP, was presented.[13] Rituximab was infused at a dose of 375 mg/m2 weekly for 3 weeks, and a laparoscopic splenectomy was performed 1 week before transplantation. After the splenectomy, DFPP treatments were performed daily thereafter. Immunosuppression consisted of BAS, CsA, MMF, and steroids. In all cases, the antibody titer dropped below 1:16 and the transplant was performed. Pancytopenia developed in 1 patient after preconditioning, but the patient recovered, allowing transplantation. One patient experienced acute cellular rejection. At a mean follow-up time of 680 days, all transplants were functioning well. The antidonor titer remained under 1:16 in most cases. In the patients who showed an increase after transplantation, the kidneys continued to function well without signs of acute humoral or cellular rejection.
Follow-ups of 2 studies with ALM were presented. The first was the 3-year National Institutes of Health tolerance trial (either alone or combined with infliximab ) and calcineurin minimization (with SRL) in living-donor renal transplant recipients (n = 15) receiving 3 or 4 induction doses of 0.3 mg/kg of ALM preceded by steroids: ALM alone (n = 7), ALM + IFX (n = 4), or ALM followed immediately by SRL monotherapy (n = 4).[14] Three-year patient and graft survival rates were 93%. All patients not receiving SRL experienced early episodes of acute rejection. Only 67% of patients were maintained on monotherapy for 3 years with either SRL (n = 8) or TAC (n = 2). Two patients developed antidonor antibodies and were treated with IVIG and TAC. Notable adverse events included portal vein thrombosis in 1 patient treated with IVIG, SRL-associated pneumonitis (n = 2), peripheral neuropathy (n = 1), late Epstein-Barr virus reactivation (n = 1), and new-onset Grave's disease (n = 1). This study demonstrates that ALM alone or combined with IFX does not produce tolerance, but sustained minimization of immunosuppression can be achieved. However, these results are no better than those reported with CsA (Sandimmune) monotherapy, which was used 20 years ago with less rejection and fewer serious side effects.
The initial results of 29 renal transplant patients treated with ALM induction with initial triple maintenance immunosuppression have been reported.[15] Although 45% of patients had acute rejection and 50% were C4d-positive, the 3-year results showed graft and patient survival rates of 96% and 100%, with 1 patient lost to follow-up.[16] Twenty patients were eventually maintained on steroid-free immunosuppressive regimens, and 15 patients were eventually maintained on monotherapy (12 on SRL). The results of this small, nonrandomized study with multiple variations in maintenance immunosuppression, including steroid withdrawal and SRL substitution, are so confounded that interpretation of the findings is difficult.
Acute rejection > grade Ib has been associated with a high percentage of monocytes, regardless of ALM treatment.[17] A follow-up study showed negative CD52 expression in renal glomeruli, tubules, and vessels, suggesting no nephrotoxicity of ALM.[18] However, regardless of ALM induction, CD52-positive inflammatory cells may be present in some, but not all, acute rejection. In the absence of ALM induction, 80% to 90% of inflammatory cells involved in acute rejection stained positive for CD52. CD52 inflammatory cells were not present on protocol biopsies of kidney transplants without rejection.
The complications associated with chronic corticosteroid (steroid) administration have led to the development of a large number of steroid-avoidance and steroid-withdrawal protocols. Outcomes of protocols for near avoidance, early withdrawal, and late withdrawal of steroids from the immunosuppression regimen in renal transplant recipients are described.
Three presentations from the group at Hahnemann University Hospital, Philadelphia, Pennsylvania, described outcomes in protocols that employed antibody induction with basiliximab, 2 doses of steroids, and a number of maintenance regimens. All 3 studies were presented by Mysore S. Anil Kumar, MD.[1-3]
A randomized trial compared 4 different maintenance regimens: (1) tacrolimus-mycophenolate mofetil (TAC-MMF), (2) TAC-sirolimus (SRL), (3) cyclosporine (CsA)-MMF, and (4) CsA-SRL.[1] Fifty patients were in each group; 82% of the kidneys were from deceased donors and 18% were from expanded criteria donors. Fifty-two percent of the recipients were African-American (AA). Serial protocol biopsies were done. Three-year patient and graft survival rates were not different among the 4 groups: 91% to 93% and 77% to 83%, respectively. The lowest incidence of acute rejection was in the TAC-SRL group (8%) and the highest incidence of rejection was in the CsA-MMF group (26%). There were no differences in the serum creatinine (SCr) and calculated creatinine clearance levels among the 4 groups. Delayed or slow graft function was observed in 42% to 58% of the patients. The overall incidence of chronic allograft nephropathy (CAN) was 50% to 60%, and was similar among the 4 groups; however, moderate-to-severe CAN was seen least often in the patients receiving SRL combined with either TAC or CsA.
The second study was a randomized trial of TAC-MMF vs TAC-SRL in 150 patients, again with basiliximab induction and steroid avoidance (2 doses).[2] Eighty-seven percent of the transplants were with deceased donors, and 60% of the recipients were AA. Two-year patient and graft survival rates were not different between the 2 groups: 90% and 82% to 84%, respectively. SCr levels were 1.8 mg/dL in both groups. Acute rejection rates were also similar: 12% in the TAC-MMF patients and 8% in the TAC-SRL patients. More subclinical rejection was observed in the TAC-MMF group. Less moderate-to-severe CAN was seen in the TAC-SRL-treated group.
The final study from this group compared outcomes between 50 AA and 50 non-AA recipients.[3] Basiliximab induction with just 2 doses of steroids plus either TAC-MMF or TAC-SRL were administered to all patients. Three-year patient and graft survival rates were comparable, as were the SCr levels, the rates of acute rejection, delayed or slow graft function, CAN, and posttransplant diabetes mellitus (PTDM, 6% in AA and 4% in non-AA recipients).
Taken together, these results represent remarkably good outcomes in a very difficult patient and donor graft population.
A. Humar, MD, of Minneapolis, Minnesota, presented 5-year outcomes in 589 patients undergoing transplantation who received 5 days of rabbit antithymocyte globulin (rATG) induction and 5 days only of steroids.[4] Patient and graft survival were 91% and 84%, respectively, the incidence of rejection was 16%, and the mean SCr was 1.7%. Three-year graft survival rates in patients with no acute rejection, acute cellular rejection, or humoral rejection were 92%, 88%, and 79%, respectively. Cytomegalovirus infection was seen in 12%, posttransplant lymphoproliferative disorder in 1%, and posttransplant diabetes mellitus in 1% of patients. Outcomes and complication rates were compared with an historical control group receiving steroids and were superior in all parameters studied. This study provides important insights into the durability of steroid avoidance with medium-term follow-up.
Two reports from the University of Cincinnati, Cincinnati, Ohio, described outcomes associated with early (1-week) steroid withdrawal. In the first study, E. Steve Woodle, MD, reported on 84 AA recipients who were compared with 216 non-AA recipients, all transplanted between 1995 and 2004.[5] The AA recipients received fewer living-donor kidneys (25%) than the non-AA recipients (63%) and had higher peak panel-reactive antibody levels (31% vs 10%). However, comparable 2-year patient survival rates (98.5% and 98%) and graft survival rates (95% and 94%) were observed in the AA and non-AA recipients, respectively. Acute rejection was observed more in the AA recipients (27%) than in the non-AA recipients (15%); however, a multivariate analysis did not implicate race as a risk factor. SCr levels were not significantly different at 2 years between the AA and non-AA recipients. These outcomes in AA recipients undergoing early steroid withdrawal were exceptionally good.
Rike and colleagues,[6] of Cincinnati, examined cardiovascular risk and cardiovascular events in 258 patients undergoing early steroid withdrawal and 149 patients maintained on steroids with a 2-year follow-up. Although cardiovascular risk scores were not different between the 2 groups, the incidence of cardiovascular events was significantly lower in the steroid-withdrawal patients (10%) than in the steroid-maintenance patients (20%). This is one of the first studies to demonstrate an impact of steroid withdrawal on cardiovascular events.
Lorenzo Gallon, MD, of Chicago, Illinois, compared TAC-SRL (n = 37) and TAC-MMF (n = 45) as maintenance regimens for early steroid withdrawal[7]; all patients received anti-interleukin-2 receptor antibodies and 3 days of steroids. Patient survival was comparable between the 2 groups, but graft survival and renal function were better in the TAC-MMF group through 24 months of follow-up. Ninety-seven percent of the TAC-MMF group remained off steroids compared with 87% of the TAC-SRL group.
Somewhat worrisome were outcomes presented by Mark Laftavi, MD, of Buffalo, New York, from a randomized trial comparing early (1-week) steroid withdrawal (n = 34) with low-dose (5 mg/day at 1 month) steroids (n = 31) in patients receiving rATG, TAC, and MMF.[8] Protocol biopsies were performed at 1, 6, and 12 months. Patient and graft survival rates were comparable at 1 year, as was the incidence of acute rejection and the glomerular filtration rate. However, in the 1-year protocol biopsies, more CAN and fibrosis were observed in the steroid-withdrawal group. The investigators expressed concern over the long-term outcomes associated with early steroid withdrawal.
Donald Hricik, MD, of Cleveland, Ohio, described somewhat more discouraging, suboptimal outcomes with a 51-month follow-up in a group of 34 AA recipients who received TAC-SRL and were withdrawn from steroids 3-5 months after transplantation.[9] Although early outcomes had been encouraging, with an acute rejection rate of 7% and an ability to withdraw steroids in 90% of patients, the cumulative acute rejection rate was 32%, with a noncompliance rate of 35% and a 23% graft loss rate. Forty-four percent of patients remained off steroids. Even in patients without late rejection, there was a steady rise in the SCr over time. Although the 5-year graft survival rate of 72% was better than the average United Network for Organ Sharing outcomes, the high rate of late rejection and noncompliance was discouraging.
Wael El Haggan, MD, of Caen, France, examined bone mineral densities (BMD) in patients withdrawn (6 months after transplantation) and not withdrawn (7.5-10 mg/day) from steroids.[10] Thirty-eight patients were studied; 27 had 2 years of follow-up, whereas 17 were off steroids and 10 were on steroids. Increased BMD in both the lumbar spine and the femoral neck were seen in the patients withdrawn from steroids; increased osteocalcin and alkaline phosphatase levels were also observed in these patients.
Taken together, these reports describe generally favorable outcomes in patients, including AA recipients, transplanted without steroids, although a couple of the reports were less optimistic. Ultimately, long-term outcomes, including half-life determinations, will be needed to assess the outcomes of these studies. In general, short-term and medium-term outcomes remain promising.
The remarkable improvement in short-term graft and patient survival noted over the last decade has altered the focus of clinical research in renal transplantation from short- to long-term outcomes.[1] Recently, in a dramatic change from tradition, more than 1 expert has seen fit to compare longevity after a transplant, not with its usual comparator (survival on chronic dialysis), but with longevity in the general population. To even dream that chronic kidney disease (with its attendant hypertension, cardiovascular risk, and metabolic bone disease) may someday not be perceived as a threat to longevity seems outrageous folly. On the other hand, 1-year graft survival rates routinely in excess of 90% must have been perceived as an equally foolish goal by our predecessors.
Currently, long-term graft survival is thought to require chronic administration of potent pharmacologic immunosuppression.[2] Although such an approach is effective, at least over the intermediate term, morbidity and mortality directly attributable to adverse effects of immunosuppressive drugs (ie, infection, malignancy, cardiovascular risk, and nephrotoxicity, among others) compromise outcomes. Clearly, if our newfound emphasis on enhancing long-term results is to bear fruit, the paradigm must be challenged. Although immunologic tolerance in its classic interpretation remains beyond our reach, current efforts to enhance long-term outcomes are devoted to "optimizing" or "minimizing" immunosuppression.[3,4] The 2005 American Transplant Congress in Seattle, Washington, hosted the presentation of several key advances in this regard.
In solid organ transplantation, it has long been recognized that more intense immunosuppression is necessary to prevent rejection early in the posttransplant course. This phenomenon, termed "accommodation," has only been loosely characterized and never easily quantified. Nonetheless, it is well known that at least some subsets of kidney allograft recipients require very little, if any, immunosuppression over time. Witness the 50% graft survival achieved under azathioprine (AZA)/corticosteroid (steroid)-based regimens.[2] The benefit of newer immunosuppressants has been to reduce the risk of acute rejection (and its associated comorbidities) for these patients, and to double the number of recipients for whom excellent outcomes can now be expected.[5] In the United States, what is now considered traditional immunosuppression includes triple therapy (calcineurin inhibitor , antiproliferative agent, and steroids), with or without antibody induction at the time of transplantation.[6] Optimizing therapy implies that the best outcomes for our patients will result from reducing reliance on more toxic agents in protocols, and achieving maximal effectiveness from the less toxic agents that remain.
A long-held goal of clinical immunosuppressive management has been to directly monitor the immunologic status of the recipient rather than rely on the current indirect standard -- serial measurement of immunosuppressant levels in the blood.[7-9] CNIs block lymphocyte production of NFAT (nuclear factor of activated T cells)-stimulated cytokines, notably, interleukin-2 and interferon-alfa.[10] Sommerer and colleagues[11] presented data expanding on their previous work with real-time polymerase chain reaction (PCR) to measure CNI inhibition of these cytokines in renal transplant recipients. In 5 subjects, cytokine production from stimulated peripheral blood lymphocytes drawn 2 hours after oral dosing was reduced by > 80% compared with sampling immediately prior to dosing. Over time, and with stepwise dose reduction of cyclosporine (CsA) with a 30% decline in blood levels, these parameters remained relatively stable and the patients maintained stable kidney function, with perhaps fewer side effects. The study authors offer the hope that this technique may someday supplant drug levels.
van de Wetering and colleagues[12] examined cytotoxic T-lymphocyte precursor frequencies (CTLpf) as an indicator of which patients may be successfully withdrawn from CNI therapy; those with lower CTLpf may prove more amenable to CNI withdrawal. This group previously demonstrated that patients with low CTLpf tolerated CNI withdrawal.[13] In this study, however, virtually all patients were successfully withdrawn from CNI, regardless of CTLpf status. Unfortunately, this finding is in keeping with the lack of clinical applicability of several other previously promising markers.[7,8,14]
Mycophenolate mofetil (MMF) was developed as a fixed-dose drug, without understanding of or recommendation for therapeutic drug monitoring.[15] More recently, enterically coated mycophenolate sodium (EC-MPS [Myfortic]) was made available simply by documenting equivalent efficacy and tolerability to MMF in fixed-dose regimens with CsA.[16] However, data accumulating in the 1990s from Europe indicated that MMF exposure influenced outcomes, even in traditional regimens with CsA, implying that monitoring may be of benefit.[17] In addition, due to the absence of the interactions that accompany MMF usage with CsA, mycophenolic acid (MPA) exposure varies widely when used alone or in combination with tacrolimus (TAC) or sirolimus (SRL).[18] Thus, it seems likely that the potential benefit of therapeutic drug monitoring of MPA may be even greater in "alternative" protocols. El Haggan and colleagues[19] reported area under the curve (AUC) for MPA in 15 stable patients receiving variable doses of MMF in combination with SRL. In the SRL-treated patients, MPA exposure with a 1500-mg daily dose of MMF was identical to that noted with a 2-g dose in combination with CsA. For most patients in the SRL group, a daily dose of 1 g provided what many would consider adequate exposure (AUC >/= 30 mg/hour/L). As AUC measurements are cumbersome, this group reported good correlation between trough (C0)-MPA and AUC (r =.67), with trough levels of 2.3 ± 1.7 and 3 +/ 1.9 mg/L in the 1000-mg and 1500-mg dose groups, respectively.
With MMF, bioavailability of MPA increases over the first 6-8 weeks after transplantation, after which the levels tend to stabilize.[17] In addition, within-patient variability has been thought to be large. Thus, when or how to best assess MPA exposure has been uncertain. van Hest and colleagues[20] analyzed MPA exposure in 30 transplanted patients receiving CsA at 3, 7, 11, and 21 days post transplantation, as well as every 30 days for 5 months afterward. Their work demonstrated within-patient variability to be quite low, but greater in C0 measures than AUC during the first few weeks. To optimize MPA exposure, the investigators recommended 1 measurement, preferably by AUC, during the first week after transplantation, with a second measurement after 2 months to document adequate exposure for the long term.
The difficulty of actually implementing MPA monitoring in clinical practice was demonstrated by Pereira and colleagues,[21] with the Modification of Doses to Improve Function through the Years (MoDIFY) trial. In this open-label, prospective, randomized study in 82 patients, prospective MPA monitoring was utilized with the goal of optimizing MMF dosing to reduce reliance on CNIs. TAC levels were kept lower in the monitored group (7-10 vs 11-14 ng/mL, respectively) during the first 6 months after transplantation. However, interim analysis disclosed no significant differences in MPA exposure at 14 and 180 days after transplantation among treatment groups, and clinical outcomes were similar in the monitored group and 2 control groups receiving full doses of CNI without MMF monitoring. Nonetheless, the investigators interpret their data to this point as indicating a role for MMF monitoring in facilitating reduced CNI exposure.
As previously noted, standard immunosuppressive protocols dose CsA or TAC to achieve target blood levels. These levels, in turn, correlate at least to some degree with the clinical goals of maximizing efficacy and reducing adverse events. Trough (C0) levels of TAC correlate better with clinical outcomes than do similar levels of CsA.[9] In recent years, it has become evident that CsA levels measured at their peak, 2 hours post dosing (C2), correlate better with overall drug exposure after a dose than traditional C0 levels.[22] The value of monitoring C2 levels in minimization protocols was tested by Saaed and coworkers[23] in 100 patients receiving steroid-free immunosuppression. Although clinical outcomes (patient survival, graft survival, acute rejection, and renal function) were similar in patients monitored with trough and peak levels, 1-year protocol biopsies disclosed more moderate-to-severe chronic allograft nephropathy (CAN) in the C0 group.
Given the potency of currently available maintenance protocols, the role of antibody administration at the time of transplantation continues to evolve.[6] The traditional benefits of antibody induction include a reduction of early acute rejection episodes and avoiding early use of nephrotoxic immunosuppressants.[24] Indeed, in an era when kidneys from expanded criteria donors and aging recipients are increasingly common, these traditional benefits remain relevant. Osuna and colleagues[25] used standard daclizumab dosing in 133 older recipients of expanded criteria donor kidneys, delaying TAC for as long as a week and then minimizing target C0 levels (5-8 ng/mL). Overall rejection rates were < 10%; infection was manageable; and ultimately renal function was quite good (median serum creatinine level of 1.5 mg/dL) in these challenging patients. In a similar cohort of 74 older patients receiving expanded criteria donor kidneys in Brazil, Keitel and coworkers[26] used 2 doses of daclizumab to test CNI minimization in 3 treatment groups. The first group received low-dose CsA; the second group received low-dose TAC; and the third group received SRL -- all in combination with MMF and steroids. Rates of delayed graft function were high in all groups; however, the best outcomes in terms of graft survival after 1 year accrued to patients in the TAC arm.
Increasingly, however, antibody induction is being viewed in a new light, as a tool to facilitate long-term minimization of maintenance immunosuppression.[27] Indeed, results from 2 multicenter studies indicate a benefit of anti-CD25 antibodies in facilitating the relatively conservative goal of reduced long-term CNI exposure. Budde and coworkers[28] administered basiliximab to 89 patients also receiving CsA, EC-MPS, and steroids, then randomized after 30 days to reduced-dose vs full-dose CsA. Despite 10% to 15% lower CsA exposure in the reduced-dose group, clinical outcomes after 1 year were identical between groups. Similarly, Vincenti and colleagues[29] presented 18-month data from an international trial involving 536 patients. Administration of a standard regimen of daclizumab in combination with low-dose CsA, MMF, and steroids resulted in less rejection and better renal function than in 2 comparator groups that underwent either complete CsA withdrawal or remained on full-dose CsA.
More radical applications of antibody induction invoke principles of lymphocyte depletion, with much interest in the pan T-cell anti-CD52 antibody, alemtuzumab (Campath-1H).[27] Flechner and colleagues[30] presented preliminary results from a 2-center pilot trial involving 22 recipients of renal allografts. After two 30-mg doses of alemtuzumab (on days 0 and 1, along with two 500-mg doses of methylprednisolone), patients were maintained on MMF 500 mg twice daily and concentration-controlled SRL (target range, 8-12 ng/mL). Thirty-six percent of patients experienced acute rejection, several episodes of which were quite severe, and adverse events were relatively common. Nonetheless, with follow-up approaching 1 year, 85% remained off CNI and 95% remained off steroids. The investigators appropriately caution that longer follow-up will be required before additional studies should be attempted. Similarly, Barth and colleagues[31] implemented a single-center, randomized trial of alemtuzumab induction. All subjects received a single 30-mg dose of alemtuzumab, along with CNI, MMF, and steroids. Patients were randomized to remain on or be weaned off CNI after 2 months. In a small number of subjects (15 approaching 1 year of follow-up), rejection rates remained quite low in both groups.
A recent article by Nankivell and colleagues,[32] implying that CAN is the inescapable consequence of long-term CNI therapy, has renewed enthusiasm for protocols avoiding use of CsA and TAC (even though the long-term impact of the latter may not be so ominous).[33] Approaches and implications of CNI-sparing protocols are linked to the timing of the intervention; although some studies examine avoidance of CNI from the time of transplantation, others focus on minimizing exposure only after the onset of graft dysfunction.
Glotz and colleagues[34] of the French-Belgium Sirolimus Study Group randomized 141 patients to receive either rabbit antithymocyte globulin (rATG) and SRL or TAC without antibody induction. All subjects also received MMF and steroids. Study withdrawal occurred much more often in the SRL-treated patients (37% vs 6%), most often due to adverse events, including impaired wound healing, lymphocoele formation, cytopenia, and hyperlipidemia. After 6 months, however, rejection rates were comparable (< 15%), and the glomerular filtration rate (GFR) was higher (74 vs 62 mL/min) in those patients off CNI therapy.
Lebranchu and colleagues[35] compared outcomes in 145 patients 1 year post transplantation, all of whom received rATG, MMF, and a 6-month course of steroids. Patients had been prospectively randomized at the time of transplantation to receive either SRL (target C0 levels of 10-15 ng/mL) or CsA (target C0 levels of 75-250 ng/mL). Graft survival, patient survival, and acute rejection rates were identical between groups, and 88% of those in each group remained off steroids at 1 year. The primary study end point, calculated GFR, was higher (69 vs 61 mL/min) in the SRL group. However, as also noted in the previous study, withdrawal for adverse effects was substantially more common among those randomized to SRL. In addition, urinary protein excretion was significantly greater in the SRL-treated patients (640 vs 180 mg/day).
In a study with a 2-year follow-up, Morales and coworkers[36] randomized 87 renal allograft recipients on triple therapy (TAC, SRL, and steroids) to either continue or discontinue TAC after the third month post transplantation. After 2 years, among the 52 patients remaining on assigned therapy, GFR was higher and blood pressure was lower in those off CNI. However, in the overall, intention-to-treat cohort, there were no differences in GFR, graft survival, or patient survival.
Although the results of CNI avoidance in de novo patients can best be termed "mixed," the benefit of discontinuing CNI therapy in patients with allograft dysfunction may be more readily apparent. In a multicenter, international study of 143 renal allograft recipients with CAN on CsA, patients were randomized to remain on CsA or convert to MMF.[37] Both groups remained on steroids. After 2 years, renal function was significantly more likely to stabilize or improve in the MMF group, with no between-group differences in acute rejection, graft survival, or other complications.
The issue of proteinuria in CNI-sparing protocols with SRL was highlighted in a single-center study presented by Bumbea and colleagues.[38] The investigators converted 43 recipients from CNI to SRL for allograft dysfunction (n = 38) and cutaneous malignancy (n = 5). All patients remained on steroids and either MMF or AZA. Over 2 years, creatinine clearance did not deteriorate but actually improved (from 49 ± 15 to 52 ± 20 mL/min) after conversion. Again, anemia, leukopenia, and thrombocytopenia complicated the courses of the conversion patients, with increased use of erythropoietin therapy. De novo proteinuria (> 500 mg/day) occurred in 37% of converted patients, with 12% demonstrating nephrotic-range urinary protein excretion. The pathogenesis and implication of proteinuria associated with SRL-based therapy remain uncertain, but have been increasingly reported.[39]
How is one to interpret these results of CNI withdrawal in the face of CAN? Mulay and colleagues[40] offer a review of the existing literature, most of which originated in nonrandomized trials. Their principal finding is that conversion from CNI to SRL-based therapy generally results in short-term improvement in GFR at the cost of increased risk of hyperlipidemia, proteinuria, and other adverse effects. Not surprisingly, the investigators conclude their work with a call for more randomized, long-term studies.
Complications associated with chronic steroid therapy are well documented and for years have provided the impetus for steroid-sparing trials.[41,42] Traditionally, steroids were weaned in stable patients at some predefined point after transplantation, a practice that continues at some centers. Touchard and colleagues[43] presented a study in which stable patients underwent steroid withdrawal during the first posttransplant year under an immunosuppressive protocol that included antibody induction, CsA, and MMF. During the second year, patients considered at low immunologic risk were randomized to stay on CsA/MMF, discontinue MMF, or switch from MMF to AZA. Not surprisingly, subsequent follow-up revealed no significant differences in clinical outcomes among these low-risk patients. However, it seems increasingly clear that, for most patients, the most effective approach to steroid sparing is avoidance, rather than weaning off steroids after stability is achieved. Successful steroid sparing means never introducing the agents in the first place.
A lingering question is which CNI best supports steroid avoidance. In a presentation with the intriguing title, "Monotherapy Head to Head: Tacrolimus versus Cyclosporine," Coupes and coworkers[44] offered data accumulated in 92 patients over 2 years. After that time, 63% of the TAC-treated patients and 72% of the CsA-treated patients remained on monotherapy. Rejection rates were comparable. Renal function was better in the TAC-treated patients, who also had less hyperlipidemia. However, glucose tolerance tended to be worse with TAC. Which CNI is best for steroid-sparing protocols? The question remains unanswered.
Even with combination therapy, acceptable trade-offs in a steroid-free regimen remain difficult to determine. Marks and colleagues[45] presented single-center data from Seattle, comparing a CsA-MMF regimen to TAC-SRL after rATG induction in 124 patients. Graft and patient survival at 1 year were excellent in both groups. Early acute rejection occurred less commonly with the TAC-SRL regimen (4% vs 12%), but viral infections (including polyomavirus and cytomegalovirus) and posttransplant diabetes mellitus (15% vs 5%) occurred more frequently in the TAC-SRL-treated patients.
Are all renal allograft recipients candidates for steroid-sparing protocols? Traditionally, African-American recipients have not been thought to fare as well as whites.[42] Gruber and coworkers[46] presented data in 47 African-American recipients, 27 of whom were treated with a steroid-avoidance protocol; all received rATG, TAC, and MMF. Preliminary short-term findings indicated identical outcomes with and without steroids. These findings support other recent reports of successful steroid-free immunosuppression in African-Americans.[47,48]
Much of the literature regarding steroid avoidance in kidney transplantation derives from open-label studies. Thus, the ongoing blinded trial conducted under the auspices of the Fujisawa Steroid Withdrawal Study Group[49] is of interest. In this multicenter study involving 386 patients, patients receiving rATG, TAC, and MMF are randomized to continue or discontinue steroids after 7 days post transplantation. Although the trial will remain blinded for 5 years, data after 2 years were presented in Seattle. Early steroid-withdrawal patients experienced more acute rejection at 1 year (12% vs 6%), but had identical patient and graft survival rates and a significantly more favorable cardiovascular risk profile compared with those remaining on steroids. The data that will continue to emanate from this trial are likely to prove extremely informative in upcoming years.
Regardless of how one perceives evolving practices in immunosuppression, the paradigm is changing. A decade ago, renal transplant immunosuppression was remarkably homogeneous around the world. Now, it is obvious that outstanding results may be achieved with many different approaches. Whether one chooses to focus on antibody induction, therapeutic drug monitoring, or sparing of toxic immunosuppressants, there are benefits and hazards associated with each. The ongoing challenge is to translate this rapidly developing information into clinically applicable protocols that best serve the interests of recipients.