Patient and graft survival after liver transplantation are adversely affected by early posttransplant renal dysfunction, which is often, at least in part, attributable to the administration of a calcineurin inhibitor (CI)—cyclosporine or tacrolimus. Thus, we are challenged to develop alternative protocols that are as “renal sparing” as possible. This is particularly relevant today, given the importance of creatinine in determining a patient’s model for end-stage liver disease (MELD) score and priority on the waiting list. Although regimens using no CIs are now being tested (CI avoidance), it is not yet known whether these regimens will result in acceptable outcomes or whether the potential side effects of these regimens will outweigh their advantage of less nephrotoxicity. Indeed, a pilot study using daclizumab induction and CI avoidance after liver transplantation showed a 100% incidence of rejection, including 57% with steroid-resistant rejection (1). As such, CIs remain an integral component of most centers’ early maintenance regimens, but lower doses and delayed initiation of CIs may be a safe and effective approach.
Anti-CD 25 monoclonal antibody induction therapy has been shown to be safe and efficacious in kidney (2–4) and liver (5–10) transplant recipients. After the humanized anti-CD 25 monoclonal antibody, daclizumab, gained approval in renal transplantation, we initially used a combination of two-dose daclizumab induction, low-dose CI, and corticosteroids in liver recipients felt to be at risk of posttransplant renal dysfunction (10). Because of our pilot experience with respect to the efficacy and side-effect profile of daclizumab, we liberalized the criteria that we felt might predispose to renal insufficiency in the early posttransplant period, which led to its use on a more frequent basis, including in many with a normal pretransplant serum creatinine concentration. The purpose of this study, the largest published experience using daclizumab in liver recipients to date, was to evaluate the long-term efficacy (acute and chronic rejection, patient and graft survival) and safety (renal function, hepatitis C virus [HCV] recurrence, opportunistic infection) of this immunosuppressive strategy.
This is a retrospective, nonrandomized observational study of 352 consecutive liver transplants performed from December 1997 to July 2001. Twenty-eight pediatric (≤18 years old) transplant patients, most of whom did not receive daclizumab, were excluded. Overall, 209 adult transplant patients received daclizumab induction (2 mg/kg intraoperatively during the anhepatic phase; 1 mg/kg on postoperative day 5); 115 transplant patients received no induction therapy (control group). The first 98 patients in this series, including 39 patients who received daclizumab induction and 58 who did not, have been reported previously (10). The decision to use induction therapy in subsequent patients was made by the primary surgeon at the time of the transplant and depended on any suspicion of posttransplant renal insufficiency, which was based on less strict criteria than used in our pilot experience (e.g., >4 L of ascites, subjective difficulty of hepatectomy; i.e., patients with more blood loss during the pre-anhepatic phase were more likely to receive daclizumab). Except in our pilot experience (when less than 50% received daclizumab), daclizumab use was evenly distributed over the remaining study period. Maintenance immunosuppression, a CI (primarily tacrolimus), mycophenolate mofetil (MMF; generally in patients without viral hepatitis), and steroids, was similar, but not identical, between the induction and noninduction groups. In the induction group, the starting CI dose was 75% lower (0.05 mg/kg vs. 0.2 mg/kg total daily starting dose), and the targeted trough levels during the first 6 months were lower (8–12 ng/mL vs. 12–15 ng/mL). We were also more willing to delay CI initiation in the induction group. Alternatively, in keeping with a previously published philosophy of using MMF in patients with renal insufficiency (11), induction patients were more likely to receive MMF. In those receiving MMF, the initial total daily dose was 2 g/day, which was gradually tapered and discontinued at 3 months posttransplant, irrespective of whether daclizumab was given. Separate analyses of patients receiving MMF and those not receiving MMF were performed to evaluate the importance of this agent relative to that of daclizumab in preventing acute rejection. Steroids were tapered identically in each group and discontinued at 1 year, except for patients with a recent (<6 months) rejection episode or patients transplanted for an autoimmune disease.
Because the entire study period was in the preMELD era, the previous United Network of Organ Sharing classification was used. Status 1 (all fulminant hepatic failure patients in this series) and 2A patients were in the intensive care unit (ICU) immediately pretransplant, with a life expectancy less than 7 days; in addition, status 2A patients were required to have chronic liver disease and a Pugh score (12) of 10 or greater. Status 2B patients had chronic liver disease, a Pugh score of 10 or greater, and were not in the ICU; status 3 patients had chronic liver disease and a Pugh score 7 to 10.
All episodes of acute and chronic rejection and HCV recurrence were diagnosed by biopsy, which was only performed for abnormal liver biochemistries after vascular/biliary complications were excluded by appropriate studies. All rejection episodes were treated with 3 g of intravenous methylprednisolone, 1 g every other day for a total of three doses. Patients without biochemical/clinical improvement were rebiopsied to identify steroid-resistant rejection. Histologic grading of rejection was not routinely performed.
Cytomegalovirus (CMV)-seropositive patients were given acyclovir (800 mg by mouth twice daily) for 2 weeks for CMV prophylaxis regardless of the donor’s CMV status; CMV-seronegative patients received ganciclovir (5 mg/kg intravenously twice daily, adjusted for renal function) for 2 weeks, followed by acyclovir (800 mg by mouth twice daily) for 3 months. Once available, valganciclovir (450 mg by mouth once daily) was used instead of acycolvir/ganciclovir, with seropositive patients receiving 2 weeks of prophylaxis and seronegative patients receiving 3 months of prophylaxis.
Opportunistic infections were documented with standard blood assays/cultures or histologically/immunohistologically. CMV infections were treated with ganciclovir (5 mg/kg intravenously twice daily, adjusted for renal function) for at least 2 weeks and longer if viremia was still detectable.
Survival curves were generated by the Kaplan-Meier method and compared with the Wilcoxon test. Continuous variables were compared with a two-tailed t test (paired or unpaired, as appropriate), categorical variables with Fisher’s exact test. Where appropriate, values are expressed as mean±standard error of the mean. P≤0.05 was considered statistically significant.
Important patient and transplant characteristics are shown in Table 1. Patients receiving induction were older, more likely to be status 2A, and were more likely to receive MMF as part of their initial maintenance immunosuppression. Because of less restrictive use of daclizumab induction in this cohort, the difference in serum creatinine concentrations pretransplant was less than that reported in our pilot experience (10), but the induction group still had significantly worse pretransplant renal function. Expectedly, the average intraoperative packed red blood cell requirement was higher in induction patients. Fresh frozen plasma use was also higher among induction patients, but this did not quite reach statistical significance. Platelet use was similar. No grafts in either group were lost to primary nonfunction. One graft in the control group was lost to hepatic artery thrombosis, which became clinically evident 2 weeks after transplant; this patient soon received a retransplant and did well with daclizumab induction. No patient experienced clinically evident side effects of daclizumab injection, and no effect was noted on the complete blood count.
There were no statistically or clinically significant differences in patient (Fig. 1) or graft (Fig. 2) survival between the groups. Similarly, the incidence of chronic rejection was not different: 9 of 209 (4.3%) versus 3 of 115 (2.6%; P=0.55). Acute rejection-free survival was significantly higher in patients receiving daclizumab induction (Fig. 3), despite receiving significantly less and significantly delayed initiation of a CI (see below); the overall acute rejection incidence within 6 months was also better in the induction group: 53 of 209 (25.4%) versus 45 of 115 (39.1%; P=0.01). One patient in each group developed steroid-resistant rejection, and both were successfully treated with OKT3.
To evaluate the possibility that our improved rejection rate seen in patients receiving daclizumab was caused by an effect of MMF, we separately analyzed those receiving and those not receiving MMF. Acute rejection was significantly less in MMF-free patients who received daclizumab compared with MMF-free patients not receiving daclizumab: 20 of 83 (24%) versus 25 of 61 (41%; P=0.04); in addition, patients receiving MMF had less acute rejection if daclizumab induction was used, 33 of 93 (26%) versus 20 of 54 (37%), although this did not reach statistical significance (P=0.16). Alternatively, MMF did not provide effective rejection prophylaxis overall (P=0.81), in induction patients (P=0.87), or in controls (P=0.71).
Because induction patients had worse renal function and may have been inherently less able to mount an alloimmune response, we separately analyzed rejection in patients with a pretransplant serum creatinine concentration 1.5 mg/dL or less. This subgroup still had a lower incidence of rejection if treated with daclizumab induction: 44 of 171 (25.7%) versus 42 of 108 (38.9%; P=0.02).
CI therapy was initiated significantly later in induction patients: 2.39±0.11 versus 1.73±0.10 days posttransplant (P<0.0001). By posttransplant day 5, 2 (1.7%) control patients had not received their first dose of CI compared with 20 (9.6%) induction patients (P=0.005). Both control patients not exposed to CI by day 5 rejected within 6 months compared with 30% (6 of 20) of induction patients not exposed to CI by day 5 (P=0.12).
Overall, 34.8% (40/115) of controls had temporary cessation of their CI during the first week because of an increase in creatinine concentration compared with 28.2% (59/209) of induction patients (P=0.26). The risk of rejection within 6 months in this subgroup was significantly greater in controls: 20 of 40 (50%) versus 16 of 59 (27.1%; P=0.03).
Of those receiving tacrolimus on day 7, the mean total daily dose on that day was significantly higher in the control group: 13.8±0.45 versus 9.9±0.31 mg (P<0.0001). In addition, despite more temporary CI cessation, the total amount of tacrolimus given during the first week posttransplant was higher in controls: 66.4±2.6 mg versus 38.4±1.7 mg (P<0.0001); this was also true for patients treated with cyclosporine: 3,743±460 mg versus 1,500±276 mg (P=0.0002).
Tacrolimus levels were not statistically different on posttransplant day 7, 7.4±0.6 ng/mL (induction group) versus 8.4±0.8 ng/mL (P=0.39), but were lower in induction patients at 1 month (10.1±0.5 ng/mL versus 12.7±0.7 ng/mL; P=0.006); tacrolimus levels at all subsequent time points analyzed were similar.
The overall incidence of CMV infection was similar in each group: 25 of 209 (12%) induction patients versus 11 of 115 (9.6%) controls (P=0.49). Treatment of acute rejection significantly increased the risk of developing CMV: 19 of 98 (19.4%) patients treated for rejection developed CMV infection compared with 17 of 226 (7.5%) patients not treated for rejection (P=0.003). This effect of acute rejection therapy on CMV infection was more pronounced in the induction group: 13 of 53 (24.5%) induction patients treated for rejection developed CMV infection compared with 12 of 156 (7.7%) induction patients not treated for rejection (P=0.003). In the control group, 6 of 45 (13.3%) patients treated for rejection developed CMV infection compared with 5 of 70 (7.1%) not treated for rejection (P=0.34). One induction patient, whose indication for transplantation was fulminant hepatic failure, died from systemic aspergillus, which became evident less than 1 week posttransplant.
Histologic hepatitis C recurrence-free survival is shown in Figure 4 and was not increased in patients receiving daclizumab induction. Of note, daclizumab also provided similar prophylaxis against rejection in HCV recipients as in the overall cohort (Fig. 5), but this did not reach statistical significance (presumably because of smaller group size). Also, HCV recurrence was associated with treatment for acute rejection only in the control group: 15 of 18 (83.3%) treated for rejection developed recurrence compared with 15 of 31 (48.4%) not treated for rejection (P=0.018). There was no association between treatment for rejection and HCV recurrence in the induction group (P=0.77)
Renal function, as measured by serum creatinine concentration, at 36 months is similar in each group, despite a significantly higher pretransplant value in the induction group (Table 2). However, in the control group, the serum creatinine concentration significantly increased (by paired t test) during the first week (P=0.004), month (P=0.0004), 6 months (P=0.0004), 12 months (P=0.0004), 24 months (P=0.0004), and 36 months (P=0.0004); excluding the two liver/kidney recipients, this also remained true at all time points (all P≤ 0.0004). In addition, induction, liver-only (i.e., excluding the 4 combined liver/kidney recipients) patients with a pretransplant creatinine concentration 1.5 mg/dL or less experienced an increase in mean creatinine concentration at each time point compared with the pretransplant value (all P≤ 0.013). However, induction, liver-only patients with a pretransplant creatinine concentration greater than 1.5 mg/dL (n=34) had a significantly decreased creatinine concentration on average at each time point; specifically, the mean pretransplant creatinine concentration in this subgroup was 2.17 mg/dL, and the posttransplant values ranged from 1.42 to 1.64 mg/dL (Table 2, daclizumab subgroup). By paired t test, the pretransplant creatinine concentrations in this group were significantly higher on average than the 1 week (P=0.0004), 1 month (P=0.0004), 6 month (P=0.004), 12 month (P=0.023), 24 month (P=0.029), and 36 month (P=0.021) values.
There were six patients requiring hospitalization for more than 2 months posttransplant, five of which were in the induction group. Including these six patients, the length of posttransplant hospital stay (LOS) ranged from 4 to 91 days in controls and 5 to 154 days in induction patients and averaged 11.8±1.0 days versus 14.2±1.0 days, respectively (P=0.11); the total number of inpatient days for the first month posttransplant averaged 12.1±0.6 days versus 14.1±0.5 days (P=0.013). After excluding patients in the ICU pretransplant (status 1 and 2A), the average LOS was 12.0±1.0 in controls and 13.3±0.9 days in induction patients (P=0.35); the total number of inpatient days during the first month in this subgroup averaged 12.4±0.6 days and 13.6±0.5 days (P=0.14), respectively.
Renal dysfunction is common early after liver transplantation, and minimizing it is clearly advantageous. Our initial efforts using daclizumab induction sought this aim, and, as such, this immunosuppressive strategy was used only in patients felt to be at high risk for posttransplant renal dysfunction (10). Finding its “renal-sparing” efficacy in those patients, along with improved rejection rates and no demonstrable adverse side effects, led to its more liberal use. Others have also used daclizumab in a “renal-sparing” manner (7,9). We have now shown that this approach more effectively prevents acute rejection in all patients—those with and without renal dysfunction—in the early posttransplant period (Fig. 3) compared with patients not receiving anti-CD 25 induction; this is true despite higher maintenance CIs in controls. It is also apparent that renal function is preserved and may actually be allowed to improve, given the delayed initiation and lower doses of a CI. Liver-related renal dysfunction is most effectively reversed by restoring liver function (e.g., with transplantation), as long as other renal insults (e.g., CI toxicity) can be minimized. Our strategy of delaying CI initiation and using lower doses/target levels reduced the renal insults associated with liver transplantation and perhaps aided in renal recovery after liver function was restored. As evidence, the induction group, who had later CI initiation and lower total CI exposure, had stable mean creatinine concentrations during the critical first month compared with controls who had significantly higher values at each posttransplant time point (Table 2). Moreover, induction, liver-only recipients with a pretransplant creatinine concentration greater than 1.5 mg/dL had significantly lower values on average at each time point posttransplant (Table 2, daclizumab subgroup). This subgroup is particularly relevant in today’s MELD era, given the importance of renal function in determining a patient’s priority on the waiting list.
It is also noteworthy that induction therapy with daclizumab appeared to protect against rejection if no CI had been given by day 5 and also if the CI had to be temporarily discontinued. It is impossible to accurately predict pretransplant who will require temporary cessation of tacrolimus/cyclosporine, and our data suggest those who receive induction do not incur an increased risk of rejection if doses have to be held (27.1% compared with 25.4% overall). Although not quite statistically significant (P=0.10), controls did appear to incur an increased risk after temporary CI cessation (50% compared with 39.1% overall).
The overall incidence of CMV infection was similar in both groups. Interestingly, however, induction patients had a stronger association between CMV infection and rejection than did controls. It is thus likely that our method of rejection therapy combined with daclizumab induction is overly immunosuppressive. As a result of our data, perhaps our induction patients undergoing treatment for rejection should subsequently receive more intense CMV prophylaxis, or our rejection therapy should be modified in this group.
Graft dysfunction secondary to HCV recurrence is arguably the most important problem after liver transplantation today, and any protocol felt to increase the severity or incidence of HCV recurrence is difficult to justify. Interestingly, our control patients were at increased risk of HCV recurrence if requiring treatment for rejection, an association not seen if daclizumab induction was used. We found it reassuring that histologic HCV recurrence-free survival in our patients did not appear to be adversely affected by daclizumab induction (Fig. 4). This is in contrast with a report by Nelson et al. (13) whose data suggested that daclizumab and MMF may be associated with early HCV recurrence and more rapid histologic progression of the disease. Close inspection of their data, however, revealed potential confounding factors (14). At least one randomized, double-blind clinical trial (8) showed basiliximab, a chimeric (mouse/human) anti-CD 25 monoclonal antibody, was superior to placebo when comparing the composite endpoint of HCV recurrence, acute rejection, death, and graft loss. Although not stated in their report, it is discernible from their data that the single endpoint of HCV recurrence was clearly not increased in their patients receiving anti-CD 25 therapy. Separately, Heffron et al. (9) showed similar HCV recurrence rates, histologic severity of recurrence, and time to recurrence among liver recipients treated with daclizumab induction compared with noninduction recipients. Data from this same group, however, showed that those with HCV recurrence were less likely to achieve normalization of serum aminotransferases after 3 months of therapy with interferon-α and ribavirin if they had received daclizumab or MMF, although no impact on graft survival was reported (15). On the basis of the relative strengths of previous reports and our data, we think it is unlikely that anti-CD 25 antibody induction therapy increases the incidence of HCV recurrence or HCV-related graft loss. An ongoing randomized, clinical trial with daclizumab will hopefully further clarify this issue.
The differences in total number of inpatient days within the first month and intraoperative packed red blood cell requirement, important outcomes in today’s cost-containment era, were statistically significantly higher in the daclizumab group. These differences are cause for concern, but it is debatable whether the differences seen are clinically or financially significant in overall patient care after liver transplantation; selection bias also conceivably accounts for these differences. Indeed, significantly more induction patients received transplants from the ICU, and the differences were neither clinically nor statistically significant when excluding this subgroup (blood utilization data not shown). In addition, although difficult to measure, overall management in the induction group was seemingly simplified; it is our subjective opinion that this was generally true. Temporary cessation of CIs was less necessary, and the improved renal function allowed fewer drug dosing adjustments than may have otherwise been necessary. Less rejection obviously meant less admissions and treatment for rejection and the potential associated morbidity, particularly in patients with hepatitis C.
There are limitations to our study. First and foremost, it is retrospective and nonrandomized. There are also many variables that interplay to impact outcomes after liver transplantation, and our study design does not allow the individual influence of these variables to be determined; however, at least two important negative factors (older age, higher acuity of illness) were more common in the induction group, which would have biased against induction with respect to patient and graft survival. Also, acute rejection, which was our primary endpoint, is not the primary enemy in liver transplantation today, and it has been shown that acute hepatic rejection can actually be associated with improved survival (16). In our experience, however, more than two acute rejection episodes within the first 6 months is an independent predictor of chronic rejection (17). Thus, although acute liver rejection may not have the same influence on long-term outcomes as in other solid-organ transplants, we should still strive to prevent it as long as there is no increased cost in terms of over-immunosuppression. Our results do allow us to conclude that lower rejection rates can be achieved with anti-CD 25 induction without this cost. That it provides superior rejection prophylaxis with less maintenance immunosuppression in a “renal-sparing” manner is also a valuable asset. Last, the possibility exists that the decrease in creatinine concentrations noted in induction patients with higher pretransplant values (Table 2, daclizumab subgroup) is attributable to the aforementioned restoration of liver function; however, this subgroup has yet to experience the progressive deterioration in renal function seen in many liver recipients (18).
In summary, this is the largest published series evaluating the use of daclizumab induction therapy in liver transplantation. Induction patients with or without renal dysfunction had significantly less acute rejection, despite having significantly delayed CI initiation and lower early CI exposure compared with contemporary controls. Moreover, the currently relevant subgroup of induction patients with a pretransplant creatinine concentration greater than 1.5 mg/dL experienced sustained improvement in renal function throughout follow-up. The lower acute rejection rate was not attended by an increased risk of complications of over-immunosuppression, including HCV recurrence, although intensified CMV prophylaxis may be indicated in induction patients requiring treatment for acute rejection. Our data suggest that routine anti-CD 25 induction therapy is a useful adjunct in liver-transplant recipients with and without pretransplant renal dysfunction.
1. Hirose R, Roberts JP, Quan D, et al. Experience with daclizumab
in liver transplantation: renal transplant dosing without calcineurin inhibitors is insufficient to prevent acute rejection in liver transplantation. Transplantation
2000; 69: 307.
2. Ahsan N, Holman MJ, Jarowenko MV, et al. Limited dose monoclonal IL-2R antibody induction protocol after primary kidney transplantation. Am J Transplant
2002; 2: 568.
3. Ciancio G, Burke GW, Suzart K, et al. Daclizumab
induction, tacrolimus, mycophenolate mofetil and steroids as an immunosuppression regimen for primary kidney transplant recipients. Transplantation
2002; 73: 1100.
4. Bumgardner GL, Hardie I, Johnson RW, et al. Phase III Daclizumab
Study Group. Results of 3-year phase III clinical trials with daclizumab
prophylaxis for prevention of acute rejection after renal transplantation. Transplantation
2001; 72: 839.
5. Yan LN, Wang W, Li B, et al. Single-dose daclizumab
induction therapy in patients with liver transplantation. World J Gastroenterol
2003; 9: 1881.
6. Niemeyer G, Koch M, Light S, et al. Long-term safety, tolerability and efficacy of daclizumab
(Zenapax) in a two-dose regimen in liver transplant recipients. Am J Transplant
2002; 2: 454.
7. Emre S, Gondolesi G, Polat K, et al. Use of daclizumab
as initial immunosuppression in liver transplant recipients with impaired renal function. Liver Transplant
2001; 7: 220.
8. Neuhaus P, Clavien PA, Kittur D, et al. Improved treatment response with basiliximab immunoprophylaxis after liver transplantation: results from a double-blind randomized placebo-controlled trial. Liver Transplant
2002; 8: 132.
9. Heffron TG, Smallwood GA, Pillen T, et al. Liver transplant induction trial of daclizumab
to spare calcineurin inhibition. Transplant Proc
2002; 34: 1514.
10. Eckhoff DE, McGuire B, Sellers M, et al. The safety and efficacy of a two-dose daclizumab
(zenapax) induction therapy in liver transplant recipients. Transplantation
2000; 69: 1867.
11. Eckhoff DE, McGuire BM, Frenette LR, et al. Tacrolimus (FK506) and mycophenolate mofetil combination therapy versus tacrolimus in adult liver transplantation. Transplantation
1998. 65: 180.
12. Pugh RH, Murray-Lyon IM, Dawson JL, et al. Transection of the oesophagus for bleeding oesophageal varices. Br J Surg
1973; 60: 646.
13. Nelson DR, Soldevila-Pico C, Reed A, et al. Anti-interleukin-2 receptor therapy in combination with mycophenolate mofetil is associated with more severe hepatitis C recurrence after liver transplantation. Liver Transplant
2001; 7: 1064.
14. Charlton MR. Mycophenolate and hepatitis C: salve on a wound or gasoline on a fire? Liver Transplant
2002; 8: 47.
15. De Vera ME, Smallwood GA, Rosado K, et al. Interferon-α and ribavirin for the treatment of recurrent hepatitis C after liver transplantation. Transplantation
2001; 71: 678.
16. Wiesner RH, Demetris AJ, Belle SH, et al. Acute hepatic allograft rejection: incidence, risk factors, and impact on outcome. Hepatology
1998; 28: 638.
17. Haustein SV, McGuire BM, Eckhoff DE, et al. Impact of non-compliance and donor/recipient race matching on chronic liver rejection. Transplant Proc
2002; 34: 1497.
18. Ojo AO, Held PJ, Port FK, et al. Chronic renal failure after transplantation of a nonrenal organ. N Engl J Med
2003; 349: 931.