12-month follow-up analysis of a multicenter, randomized, prospective trial in de novo liver transplant recipients (LIS2T) comparing cyclosporine microemulsion (C2 monitoring) and tacrolimus

Authors


  • See Editorial on Page 1452

  • The trial was designed, overseen, and implemented with the support of the trial Steering Committee including representatives of the sponsor. Day-to-day operations were managed by Novartis Pharma AG. The study database is held by an independent data center, PRA International, Swansea, UK. The data center was responsible for statistical analysis of the data. Representatives of the sponsor have commented on the manuscript before submission for publication.

  • (For complete listing of investigators and centers, see Appendix)

Abstract

The LIS2T study was an open-label, multicenter study in which recipients of a primary liver transplant were randomized to cyclosporine microemulsion (CsA-ME) (Neoral) (n = 250) (monitoring of blood concentration at 2 hours postdose) C2 or tacrolimus (n = 245) (monitoring of trough drug blood level [predose]) C0 to compare efficacy and safety at 3 and 6 months and to evaluate patient status at 12 months. All patients received steroids with or without azathioprine. At 12 months, 85% of CsA-ME patients and 86% of tacrolimus patients survived with a functioning graft (P not significant). Efficacy was similar in deceased- and living-donor recipients. Significantly fewer hepatitis C–positive patients died or lost their graft by 12 months with CsA-ME (5/88, 6%) than with tacrolimus (14/85, 16%) (P < 0.03). Recurrence of hepatitis C virus in liver grafts was similar in each group. Based on biopsies driven by clinical events, the mean time to histological diagnosis of hepatitis C virus recurrence was significantly longer with CsA-ME (100 ± 50 days) than with tacrolimus (70 ± 40 days) (P < 0.05). Median serum creatinine at 12 months was 106 μmol/L with CsA-ME and with tacrolimus. More patients who were nondiabetic at baseline received antihyperglycemic therapy in the tacrolimus group at 12 months (13% vs. 5%, P < 0.01). Of patients who were diabetic at baseline, more tacrolimus-treated individuals required anti-diabetic treatment at 12 months (70% vs. 49%, P = 0.02). Treatment for de novo or preexisting hypertension or hyperlipidemia was similar in both groups. In conclusion, the efficacy of CsA-ME monitored by blood concentration at 2 hours postdose and tacrolimus in liver transplant patients is equivalent to 12 months, and renal function is similar. More patients required antidiabetic therapy with tacrolimus regardless of diabetic status at baseline. Liver Transpl 12:1464–1472, 2006. © 2006 AASLD.

Survival rates following liver transplantation have improved markedly in the last decade, and typically over 80% of patients are alive with a functioning graft 1 year after primary liver transplantation.1 Ongoing attrition, however, means that approximately a third of patients with a functioning graft at 1 year will have lost their graft by 10 years posttransplant.2 While disease recurrence remains a major cause of graft loss, cardiovascular events and new-onset malignancy account for more than half of non-graft-related deaths after the first year posttransplant2 and as such are an important target for improving liver transplant outcomes. Liver transplant recipients are at heightened risk of developing ischemic events or cardiovascular death compared to age- and gender-matched controls in the general population,3, 4 largely as a result of metabolic abnormalities associated with the original disease condition and immunosuppressive therapy. A careful assessment of the long-term complications associated with any immunosuppressive regimen is therefore essential in addition to a consideration of efficacy. Despite the development of a range of novel agents, use of calcineurin inhibitors remains routine in liver transplantation, but comparisons of the 2 commercially-available calcineurin inhibitors, cyclosporine microemulsion (CsA-ME) and tacrolimus, have reported conflicting results both in terms of efficacy and relative cardiovascular risk in liver transplantation.5, 6

The (LIS2T) study was a multicenter, prospective trial that compared efficacy and safety outcomes in de novo liver transplant patients randomized to CsA-ME or tacrolimus at 3, 6, and 12 months posttransplant. The trial was the first to compare tacrolimus with CsA-ME, whereby CsA-ME dose is adjusted based on cyclosporine (CsA) blood concentration at 2 hours postdose (C2). As reported previously,7 the LIS2T trial found no significant differences in efficacy (other than among hepatitis C–positive patients, in whom graft failure and death was less common with CsA-ME) or in terms of malignancy or tolerability at 6 months, except for a significantly higher incidence of diabetes mellitus and of diarrhea with tacrolimus. Data collected at 12 months regarding patient and graft survival, occurrence of acute rejection, immunosuppressive regimen, and selected safety measures including incidence of malignancy and requirement for treatment of diabetes mellitus, hyperlipidemia, and hypertension to determine whether any change in outcomes or risk profile developed after the first 6 months are presented here.

Abbreviations

CsA-ME, cyclosporine microemulsion; CsA, cyclosporine; C2, blood concentration at 2 hours postdose; HCV, hepatitis C virus; C0, trough drug blood level (predose).

MATERIALS AND METHODS

Study design and conduct

LIS2T was a 6-month multicenter, randomized, open-label, parallel-group, prospective study designed to compare the efficacy and tolerability of CsA-ME using C2 monitoring vs. tacrolimus in de novo liver transplant patients receiving steroids with or without azathioprine. Patients were stratified according to hepatitis C virus (HCV) status and randomized to receive either CsA-ME with C2 monitoring or tacrolimus with trough drug blood level (predose) (C0) monitoring within 24 hours of transplantation. The study protocol was approved by local ethics committees and written informed consent was obtained from all patients. Patient care complied with the Declaration of Helsinki and good clinical practice guidelines. Detailed materials and methods relating to the LIS2T study have been described previously.7

The first analysis was performed at 3 months and a second analysis at 6 months posttransplantation on an intent-to-treat basis taking into account all events occurring in patients who completed the 6-month trial and events occurring up to 30 days postwithdrawal in patients who prematurely discontinued study medication and has previously been reported.7 Follow-up information was collected at 3 and 6 months to document the status of patients who withdrew from treatment prematurely and at 12 months posttransplant in all patients. The 12-month analysis was performed on an intent-to-treat basis.

Study Population

All patients enrolled in LIS2T were invited to provide follow-up data at 12 months posttransplantation regardless of whether they had completed the 6-month study or had discontinued study drug prior to the end of study.

Immunosuppression

Both CsA-ME and tacrolimus were initiated within 24 hours of transplantation. It was recommended that CsA-ME be started at 10-15 mg/kg/day in 2 divided doses to achieve a C2 level of 800 to 1,200 ng/mL to month 3, 700 to 900 ng/mL to month 6, and after 6 months, a C2 target range of 500 to 700 ng/mL (midpoint 600 ng/mL). Tacrolimus was initiated at 0.1 to 0.15 mg/kg per day in 2 divided doses, and thereafter the dose was adjusted to target C0 in the range of 5 to 15 ng/mL to month 3 and 5 to 12 ng/mL to month 6, and after month 6 the recommended trough target range for tacrolimus was 5-10 ng/mL. Dose adjustments of CsA-ME and tacrolimus were made to maintain CsA C2 and tacrolimus C0 levels within the protocol target ranges taking into account each patient's general condition. In addition, patients received steroids with or without azathioprine. Doses of azathioprine were defined according to local practice but had to be the same for all patients in both arms within each center. Steroid administration started intraoperatively at a dose of 1 g methylprednisolone. From day 1, patients received 200 mg/day or 3 mg/kg/day of prednisone, tapered to 10 to 20 mg by day 7, and kept unchanged until the end of the first month. During months 2 and 3, the doses were 7.5 to 15 mg/day. From month 3 onward patients received 5 to 10 mg/day. Steroid withdrawal was not permitted during the first 6 postoperative months. For HCV-positive patients, use of steroids could be adjusted to local practice as long as steroid therapy was consistent for both arms.

The use of mycophenolate mofetil or any investigational drug was excluded by protocol during the 6-month study period. No recommendations were made regarding the adjunct immunosuppressants after the end of the 6-month study, but the drugs and dosage used at the time of the month 12 visit were collected.

Evaluation

The 12-month follow-up visit collected data on events that occurred after the previous evaluation—i.e., the 6-month visit for those who completed the study or the 3- or 6-month follow-up visit for those who prematurely discontinued study medication. The following data were collected: (1) graft loss, death, biopsy-proven acute rejection, and malignancies that had occurred since previous evaluation, and (2) calcineurin inhibitor doses and blood levels, adjunct immunosuppressants, treatment for diabetes mellitus, hypertension or lipid abnormalities, and laboratory values (e.g., serum creatinine). The proportion of patients receiving treatment for diabetes mellitus, hypertension, or lipid abnormalities was assessed in all patients at 12 months as well as in those who did or did not present with the condition at baseline.

Statistical Analysis

Categorical variables were summarized as number of patients and percentage and continuous variables described as median and range. Fisher exact test was used to compare incidences of clinical events between the 2 treatment groups.

RESULTS

Patient Population

Of 500 patients enrolled in LIS2T, 495 patients were transplanted and randomized, and they received study medication (CsA-ME, n = 250; tacrolimus, n = 245). As reported elsewhere,7 there were no significant differences between patient populations in the 2 treatment groups in terms of recipient age, gender, race, donor age or type, cold ischemia time, indication for transplantation, or severity of liver disease. Twelve-month assessments were available for 221 (88%) patients in the CsA-ME group, comprising 159 patients who completed the 6-month study and 62 patients who prematurely withdrew from the study but provided data at the 12-month follow-up visit. Two hundred and twenty-three (91%) patients in the tacrolimus group provided 12-month data, including 187 patients who completed the 6-month study and 36 patients who withdrew prematurely but provided 12-month data.

Patient and Graft Survival

At 12 months, the proportion of patients surviving with a functioning graft was 85% in the CsA-ME group and 86% with tacrolimus (not significant), irrespective of whether they received dual therapy or triple therapy (i.e., including azathioprine) at time of transplant. These results are in line with patient and graft survival rates at 6 months, when no significant differences were found between the CsA-ME and tacrolimus cohorts. Table 1 summarizes the timing of death or graft loss and whether events occurred while the patient was receiving study medication or had discontinued. Causes of death and graft loss that occurred prior to 6 months and between 6 and 12 months are shown in Tables 2 and 3, respectively.

Table 1. Timing of Patient Death and Graft Loss During the First 12 Months Posttransplant
 All PatientsHCV Positive
CsA-ME (n = 250)Tacrolimus (n = 245)CsA-ME (n = 88)Tacrolimus (n = 85)
  1. NOTE: Patients who withdrew from the study but experienced an event within 30 days of their withdrawal have been accounted for together with the patients who experienced the event while on study. Patients who experienced an event more than 30 days after withdrawal have been analyzed separately, and the results are mentioned in the lower part of the table.

Death or graft loss while remaining in the study2932514
 <6 months2729513
 6–12 months2301
Death or graft loss recorded >30 days after study withdrawal8252
 <6 months5131
 6–12 months3121
Table 2. Causes of Death and Graft Loss Prior to 6 Months Posttransplant
 All PatientsHCV Positive
CsA-ME (n = 250)Tacrolimus (n = 245)CsA-ME (n = 88)Tacrolimus (n = 85)
  1. Abbreviation: GVHD, graft vs. host disease; —, none.

  2. NOTE: Figures shown are patient numbers. Each patient appears in 1 category only.

  3. Analysis included patients both actively enrolled in the study or who had experienced an event within 30 days of withdrawal.

Death with a functioning graft    
 Infections8423
 Shock liver11
 Bleeding11
 GVHD111
 Cardiac/respiratory failure22
 Large bowel ischemia1
 Recurrence of hepatocellular carcinoma1
 Multiorgan failure2
 Other1
Graft loss with subsequent death    
 Acute Rejection11
 Primary graft nonfunction21
 Hepatic artery thrombosis222
 Portal vein thrombosis1
 Other2
Graft loss with retransplantation    
 Acute Rejection111
 Chronic rejection1
 Primary graft nonfunction11
 Hepatic artery thrombosis4611
 Portal vein thrombosis1 
 Surgical complications: Bleeding13
Graft loss without retransplantation and loss of subsequent follow-up    
 Hepatic artery thrombosis11
 Recurrence of liver disease11
Total number of deaths or graft loss2729513
Table 3. Causes of Death or Graft Loss Occurring Between Six and 12 months Posttransplant
 All PatientsHCV Positive
CsA-ME (n = 250)Tacrolimus (n = 245)CsA-ME (n = 88)Tacrolimus (n = 85)
Death or graft loss while remaining in the study1
 Infection (death)11
 Chronic rejection (graft loss/retransplant)1
 Hepatocellular carcinoma (death)11
 Multiorgan failure (death)    
Death or graft loss recorded >30 days after study withdrawal    
 HCV cirrhosis (death)1111
 Hepatic artery thrombosis (graft loss/retransplant)11
 Chronic rejection (graft loss/retransplant)1

The proportion of HCV-positive patients who died or lost their graft by 12 months while remaining in the study was significantly lower in the CsA-ME arm (5/88, 6%) than in the tacrolimus cohort (14/85, 16%) (P < 0.03), confirming the results observed at 6 months.7 In addition, 7 HCV-positive patients died or lost their graft more than 30 days after withdrawing from the study, such that in total 89% of HCV-positive patients survived with a functioning graft in the CsA-ME group compared to 81% in the tacrolimus group.

All liver biopsies performed during the first 12 months posttransplant that were driven by clinical events were analyzed (107 biopsies performed in 58 patients [66%] in the CsA-ME arm and 109 biopsies in 57 patients [67%] in the tacrolimus arm); protocol biopsies performed at preset dates were excluded to avoid jeopardizing the evaluation. Of interest, results showed that the same number of patients in each group (n = 27) showed histological signs of hepatitis C in biopsies undertaken in response to clinical events, but that the mean time to histological diagnosis of hepatitis C recurrence was significantly longer with CsA-ME (100 ± 50 days) than with tacrolimus (70 ± 40 days, P < 0.05, Wilcoxon test). Data on alanine aminotransferase levels was assessed in 60 patients (CsA-ME, n = 31; tacrolimus, n = 29), and 24 patients in the tacrolimus group (83%) compared to 22 patients in the CsA-ME group (72%) had elevated alanine aminotransferase levels, although this difference was not statistically significant.

Information on the presence of HCV RNA in the serum after liver transplantation was available only for 68 patients. Of these, 29/33 (88%) in the CsA-ME group and 30/35 (86%) in the tacrolimus group had detectable HCV RNA posttransplant, and the levels of RNA were broadly equivalent (difference <0.5 log at all time points analyzed) (P = not significant). Analysis of liver biopsies showed that the degree of fibrosis and activity index were equivalent between the 2 groups (P = not significant, Kaplan-Meier). The proportion of patients with fibrosis stages 0 or 1 was 58.6% in the CsA-ME group and 44.4% in the tacrolimus group (P = not significant). Fibrosis ≥2 was seen in 41.4% of CsA-ME treated patients and 55.6% of the tacrolimus group (P = not significant).

The donor age for patients transplant for HCV was similar in both groups (CsA-ME 43 ± 15 years vs. tacrolimus 46 ± 16 years [P = not significant]) as well as the proportion of living donors (CsA-ME 10% vs. tacrolimus 8% [P = ns]). The severity of the HCV disease as assessed by the Child-Turcotte-Pugh score was comparable (38% of CsA-ME and 42% of tacrolimus patients were Child-Turcotte-Pugh class C [P = not significant]). Antiviral therapy was administered to 7 patients in the CsA-ME group and 8 patients in the tacrolimus group after HCV recurrence was confirmed by biopsy. No other patients received HCV antiviral treatment.

Biopsy-Proven Acute Rejection

Between 6 and 12 months, 8 patients in the CsA-ME group and 6 in the tacrolimus group experienced acute rejection which were biopsy confirmed and treated. Three additional patients were treated for rejection in the absence of a biopsy (1 in the CsA-ME group and 2 in the tacrolimus group). Response to treatment is detailed in Table 4. The severity of rejection was not recorded between 6 and 12 months, but of rejection episodes prior to 6 months (CsA-ME, n = 65; tacrolimus, n = 58, P = not significant), 8% were severe in the CsA-ME group vs. 15% in the tacrolimus group, 55% (CsA-ME) and 59% (tacrolimus) were moderate and 37% (CsA-ME), and 26% (tacrolimus) were mild. Out of the patients who withdrew prematurely from the study, biopsy-proven and treated rejection occurred in 6 CsA-ME-treated patients and in 5 tacrolimus-treated patients after withdrawal.

Table 4. Response to Rejection Treatment for Episodes of Rejection Occurring Between 6 and 12 Months Posttransplant
 CsA-ME (n = 9)Tacrolimus (n = 8)
Resolved43
Improved but residual impairment23
No improvement11
Unknown21

Malignancies

Between months 6 and 12, 5 patients receiving CsA-ME developed malignancies (4 skin cancers and 1 pulmonary metastasis). Two cases of malignancy were reported in tacrolimus-treated patients (1 skin cancer and 1 recurrent hepatocellular carcinoma).

Renal Function

Median serum creatinine at 12 months was 106 μmol/L (range, 34-627 μmol/L) in the CsA-ME cohort and 106 μmol/L (range, 44-539 μmol/L) in the tacrolimus treatment arm. These values were similar to those recorded at month 6 (CsA-ME, 106 μmol/L; tacrolimus, 103 μmol/L).

Treatment for Diabetes, Hyperlipidemia, or Hypertension

A statistically significantly higher proportion of patients who were not diabetic at baseline were treated for diabetes at 12 months in the tacrolimus group compared to those that were treated with CsA-ME (Table 5), a difference that was sustained when patients who switched calcineurin inhibitor were excluded from the analysis.

Table 5. Use of Antihyperglycemic, Antihypertensive, and Lipid-Lowering Medication at 12 Months Among Patients Without the Condition at Baseline
 CsA-METacrolimusP Value
  1. Abbreviation: NS, not significant.

  2. NOTE: Figures shown are the number of patients receiving treatment at 12 months; the denominator is the number of patients without the condition at baseline.

Treatment for diabetes9/179 (5%)24/181 (13%)<0.01
Treatment for hypertension59/196 (30%)50/206 (24%)NS (0.217)
Treatment for hyperlipidemia11/239 (5%)7/240 (3%)NS (0.350)

No difference was observed between the 2 groups regarding the 12-month incidence of treatment for de novo hypertension or hyperlipidemia (Table 5).

A substantial proportion of patients who presented with diabetes, hypertension, or hyperlipidemia at baseline were not treated for the condition at 1-year posttransplant (Table 6). Among patients who were diabetic at baseline, significantly more of those receiving tacrolimus required antidiabetic treatment at 12 months. The percentage of patients with pretransplant diabetes who received treatment for diabetes at baseline was similar in both arms (40%) as well as the proportion of insulin-dependant diabetes (4% in each arm). No difference was observed between treatment arms regarding the need of antihypertensive or lipid-lowering treatment at 1 year in patients with pretransplant hypertension or hyperlipidemia. No patients received statin therapy at entry to the study, and only 3 patients in each treatment group received statins during the first 12 months posttransplant.

Table 6. Use of Antihyperglycemic, Antihypertensive, and Lipid-Lowering Medication at 12 Months Among Patients Who Presented With the Condition at Baseline
 CsA-METacrolimusP Value
  1. Abbreviation: NS, not significant.

  2. NOTE: Figures shown are the numbers of patients receiving treatment at 12 months; the denominator is the number of patients with the condition at baseline.

Treatment for diabetes30/61 (49%)34/50 (70%)0.02
Treatment for hypertension32/47 (68%)19/29 (66%)NS (0.817)
Treatment for hyperlipidemia2/10 (20%)1/5 (20%)NS (1.0)

The lipid levels were comparable throughout the study. At 12 months, the median levels and ranges of total cholesterol (CsA-ME 4.7 mmol/L [range, 1.2-11.7] vs. tacrolimus 4.3 mmol/L [range, 1.4-7.7]) and triglycerides (CsA-ME 1.6 mmol/L [range, 0.5-16.1] vs. tacrolimus 1.4 mmol/L [0.4-9.4]) were not statistically different.

Immunosuppression

The mean dose of CsA-ME at 12 months was 3 ± 1.2 mg/kg/day with a mean C2 level of 640 ± 284 ng/mL; mean tacrolimus dose was 0.08 ± 0.05 mg/kg/day with a mean C0 level of 8.7 ± 3 ng/mL. Azathioprine was used in a similar proportion of patients in the CsA-ME and tacrolimus cohorts at 12 months (14% vs. 11%, not significant) with a median dose of 75 mg/day in each group. At baseline, 43% and 41% of CsA-ME-treated and tacrolimus-treated patients, respectively, were receiving azathioprine. At 12 months, mycophenolate mofetil was prescribed for 12% and 10% of patients in the CsA-ME and tacrolimus groups, respectively (not significant), with sirolimus used in 2% and 0.5% of patients. Corticosteroid usage at 12 months was similar with CsA-ME and tacrolimus (35% and 37% of patients, respectively) with a median dose of 5 mg/day in each group.

Adverse Events and Reason for Discontinuation

The number and the severity of adverse events were similar in both groups as previously reported.7 All patients experienced at least 1 event by 6 months; the incidence of serious adverse events was similar in both groups of patients (CsA-ME 56% vs. tacrolimus 55.5%). Of particular interest, there was no significant difference in the incidence of infections, neoplasms, or renal adverse events (Table 7).

Table 7. Adverse Events Reported by Six Months Posttransplant
 CsA-ME (n = 250)Tacrolimus (n = 245)P Value
  1. Abbreviation: NS, not significant.

  2. NOTE: Data shown as patient numbers (% patients). Each patient may have more than 1 adverse event.

Any infection158 (63%)148 (61%)NS
Serious infection48 (19%)43 (18%)NS
Psychiatric disorders109 (44%)121 (50%)NS
Hypertension105 (42%)89 (37%)NS
Renal impairment42 (17%)34 (14%)NS
Headache42 (17%)45 (19%)NS
Tremor38 (15%)51 (21%)NS
Renal failure38 (15%)39 (16%)NS
Diarrhea34 (14%)70 (29%)<0.001
Diabetes18 (7%)35 (14%)<0.02
Convulsions14 (5%)15 (6%)NS
Pruritus13 (5%)19 (8%)NS
Hirsutism10 (4%)0%
Alopecia5 (2%)4 (2%)NS
Gingival hyperplasia5 (2%)0%
Neoplasm5 (2%)3 (1%)NS

Ninety-one patients (36%) in the CsA-ME arm and 58 patients (24%) in the tacrolimus arm withdrew from the study prior to 6 months. The main reason for withdrawal was the occurrence of adverse events. More patients were withdrawn from the CsA-ME group as a response to the occurrence of adverse events than from the tacrolimus group (17% vs. 8%) (P = 0.004). Renal impairment was the most frequent adverse event accounting for discontinuation. Although the incidence of renal impairment was similar in tacrolimus and CsA-ME treated patients, investigators made the decision to withdraw 11 CsA-ME patients within the first month of the study, whereas despite similar renal dysfunction in the tacrolimus group, only 1 patient was withdrawn within the same time period. The patients who were withdrawn within the CsA-ME group had increasing but still normal serum creatinine levels, and thus the reason for withdrawal probably reflected a lack of willingness to increase the dose of CsA-ME to reach the target C2 level. This was the major reason accounting for the difference in patient withdrawals. Overall, 22% of patients randomized to CsA-ME were receiving tacrolimus at 12 months, and 5% of patients randomized to tacrolimus were receiving CsA-ME.

DISCUSSION

At 1-year post-transplant, both CsA-ME- and tacrolimus-treated patients showed equivalent outcomes in terms of patient survival, graft survival, and incidence of rejection. These findings are consistent with those of a previous prospective study of CsA-ME vs. tacrolimus that showed no significant difference in rejection rates between the 2 therapies in liver transplant recipients, even when CsA-ME is monitored by C0 level.6 Improved protection against rejection has been reported with C2 monitoring compared to C0 monitoring in liver transplantation,8 and it is feasible that earlier reports of inferior graft outcome with CsA-ME may have been the result of inappropriate exposure levels.5 However, more patients withdrew from the CsA-ME group, most frequently as a consequence of adverse events, although the number and severity of adverse events were similar in both groups. This may have been due to the limited experience of many centers in the management of CsA-ME using C2 monitoring, and the difficulties faced in reaching C2 target levels (for example in delayed absorbers of CsA) that led to higher doses than usual especially early in the posttransplant period. This led to the decision to withdraw the patients with associated adverse events or concern that increasing the dose would be deleterious to a patient's welfare. In favor of this was the fact that withdrawals decreased as investigators enrolled more patients. A second analysis of the data was performed excluding patients who switched therapy, and the results of this analysis did not differ from the intent to treat analysis. Guidance of use of C2 monitoring based on the experience from this study has been recently published,9 as well as further data on delayed absorption.10

An increase in patient and graft survival was observed at 6 months and maintained at 12 months among HCV-positive patients from the CsA-ME group compared to the tacrolimus group, whereas no such difference in death or graft loss was seen in non-HCV transplanted patients. The 2 cohorts were comparable at baseline. Factors known to influence outcome, including donor age and cold ischemia time, were similar as well as other factors such as proportion of living donors, severity of disease as indicated by Child-Turcotte-Pugh score and the need and use for antiviral therapy posttransplant.

Unfortunately we did not ask investigators to collect HCV genotype, although we would expect that the randomization and stratification would ensure an equal distribution between the 2 groups. However, it has been suggested that genotype impacts only response to therapy rather than outcome to liver transplantation. With regard to levels of HCV RNA, the limited data provided by investigators and reported here suggests that levels of HCV RNA posttransplant were similar in the CsA-ME and tacrolimus-treated patients. Thus, the reason for the difference in survival rates is unclear at present, especially in view of the fact that losses occurred early posttransplant. As reported earlier,7 most of the deaths and graft losses were not obviously associated with hepatitis C recurrence. However, other causes of graft loss, including acute rejection, infections, and hepatic artery thrombosis, were slightly more frequent in the tacrolimus group (Table 2). Whether this signals a genuine difference between treatments requires assessment in a trial, which is now underway, specially designed to study outcomes in the HCV-positive population. The hypothesis can be raised that the dose of tacrolimus, although within the recommended range, might be too high. On the other hand, in vitro studies have previously demonstrated that CsA suppresses HCV replication, an effect not shared by tacrolimus.11–13 Over the last 10 years, most trials comparing CsA-ME and tacrolimus in HCV-positive patients have enrolled fewer than 100 patients per treatment such that they were not powered to detect a difference in survival rates. Nevertheless, some authors have reported better outcomes in patients receiving CsA-ME.14–18 Only 3 trials have enrolled over 100 patients per arm.19–21 The observational study of 283 HCV-positive liver transplant recipients by Berenguer and colleagues showed that use of tacrolimus compared to CsA-ME was independently associated with development of cirrhosis by multivariate analysis (P = 0.009).19 Ghobrial et al. reported a shorter time to HCV recurrence with tacrolimus,20 while other researchers have identified a trend suggesting that immunosuppressive treatment with tacrolimus could promote the early induction of fibrosis.21 In this regard, it is interesting that the time to HCV recurrence in our study was significantly shorter with tacrolimus than CsA-ME.

Clinical assessment at 12 months included identification of patients receiving de novo medication for diabetes, dyslipidemia, or hypertension to determine whether differences between CsA-ME and tacrolimus identified at 6 months were sustained or increased. Significantly more patients who were nondiabetic at time of transplant were receiving antihyperglycemic therapy in the tacrolimus group at 12 months, in line with the higher incidence of new-onset diabetes mellitus recorded at 6 months with tacrolimus and with consistent reports from the literature.22–23 While newly instigated treatment for hypertension was numerically higher with CsA-ME, the difference was not statistically significant. Again, this is compatible with data from other published studies that have shown inconclusive results regarding a differential in the hypertensive effects of CsA-ME and tacrolimus in liver transplant patients.5, 6 Use of antihyperlipidemic agents at 12 months in patients not receiving such treatment at baseline was similar in both groups. It was recently reported that the incidence of new-onset diabetes mellitus was lower in patients receiving statin therapy.24 In our study, the use of statins was similar and low in both groups and could not explain the difference observed in incidence of new-onset diabetes mellitus.

Among patients who presented with diabetes, hypertension, or hyperlipidemia at entry to the trial, we observed that the proportion of patients who did not require treatment for hypertension or hyperlipidemia at 12 months was similar in both arms, but that more patients receiving CsA-ME did not require diabetes treatment. The proportion of Type 1 diabetes was the same in both arms and there was no difference in the proportion of Type 1 diabetics receiving treatment at 1 year between the 2 groups, suggesting that the differential between treatments was restricted to patients with Type 2 diabetes. To our knowledge this is the first observation of such a finding. Whether Type 2 diabetes is more readily managed in patients receiving CsA-ME than tacrolimus would require clarification in further trials.

The effect of immunosuppressive therapy on renal function in non-renal transplant patients is another area of potential concern. In this trial, renal function was equivalent with CsA-ME or tacrolimus at 6 and 12 months, with no evidence of a trend to deteriorating function in either group.

In conclusion, CsA-ME with C2 monitoring and tacrolimus show similar efficacy in liver transplant patients to 1-year posttransplant. The increased incidence of diabetes mellitus in patients receiving tacrolimus at 6 months was also observed at 12 months, and control of preexisting diabetes appeared to be facilitated in patients receiving CsA-ME.

APPENDIX I

The LIS2T Investigator Study Group

Argentina: F. Villamil and V. Descalzi, Fundacion Favaloro, Buenos Aires; A. Gadano and A. Villamil, Hospital Italiano, Buenos Aires. Australia: R. Jones and P. Angus, Austin and Repatriation Centre, Melbourne; S. Lynch and G.A. Balderson, Princess Alexandra Hospital, Brisbane; G. Jeffreys and B. Chester, Sir Charles Gairdner Hospital, Nedlands. Austria: F. Muehlbacher and R. Steininger, Universitatsklinik fuer Chirurgie, Wien. Belgium: B. de Hemptinne and R. Troisi, UZ Ghent. Brazil: G. Cantisani and M.L. Zanotelli, Santa Casa de Misericordia, Porto Alegre; L. Leonardi and I. Boin, Hospital das Clinicas–Unicamp Cidade Universidad Zeferino Vaz, Campinas; S. Mies and B.D. Guardia, Hospital das Clinicas FMUSP, São Paulo. Canada: G. Levy and L. Lilly, University Health Network, The Toronto Hospital, Toronto; P. Marotta and W. Wall, London Health Science Centre, London; K. Peltekian, QE11 Health Sciences Centre, Halifax; A. Roy and D. Marleau, CHUM-St-Luc, Montreal; C. Scudamore and M. Yoshida, West Vancouver; J. Tchervenkov and M. Cantarovich, Royal Victoria Hospital, Montreal. France: O. Boillot and P. Bernard, Hôpital Edouard Herriot, Lyon; Y. Calmus and F. Conti, Hôpital Cochin, Paris; D. Cherqui and C. Duvoux, Hôpital Henri Mondor, Creteil; C. Ducerf and R. Maar, Hôpital de la Croix Rousse, Lyon; J. Gugenheim and A. Myx, Hôpital de l'Archet 2, Nice; D. Neau-Cransac and K. Mouette, Hôpital Pellegrin-Tripode, Bordeaux; D. Samuel and F. Saliba, Hôpital Paul Brousse, Villejuif; L. Rostaing and N. Kamar, Hôpital de Rangueil, Toulouse. Germany: G. Otto and C. Moench, Universitaetsklinikum, Mainz. Ireland: A. McCormick and M. O'Rourke, St. Vincent's University Hospital, Dublin. Italy: D. D'Amico and U. Cillo, Policlinico–Università degli Studi, Padova; G.L. Grazi and G. Varotti, Policlinico S Orsola–Malpighi, Bologna; A. Maffei-Faccioli and G. Gerunda/R Merenda, Policlinico–Università degli Studi, Padova; A. Risaliti and U. Baccarani, Policlinico Universitario (PUDG), Udine; M. Rossi and P. Berloco, Policlinico Umberto 1, Roma. Japan: K. Tanaka and F. Oike, Kyoto University, Kyoto. New Zealand: S. Munn and E. Gane, New Zealand Liver Transplant Unit, Auckland Hospital, Auckland. Norway: I. Brekke and K. M Boberg, Transplantasjonskirurgisk Seksjon, Oslo. Spain: A. Bernardos and I. Garcia, Hospital Virgen del Rocio, Seville; V. Cuervas and J.M. Moreno, Clinica Puerta de Hierro, Madrid; A. Rimola and I. Cirera, Hospital Clinico y Provincial, Barcelona; F. Sanjuan and M. Prieto, Hospital La Fe, Valencia; M. Salcedo and G. Clemente, Hospital Gregorio Mara, Madrid. Sweden: B. Ericzon and H. Gjertsen, Huddinge University Hospital, Huddinge; S. Friman and G Herlenius, Sahlgrenska University Hospital, Gothenburg. Switzerland: P.A. Clavien and Z. Kadry, Unispital Zürich. United Kingdom: D. Mayer, Queen Elizabeth Hospital, Edgbaston, Birmingham; P. Hayes and J. Davidson, Royal Infirmary of Edinburgh, Edinburgh; S. Pollard and J. Hodgson, St. James's University Hospital, Leeds. United States: J.S. Bynon, University of Alabama Medical Center, Birmingham; G. Klintmalm and G. Davis, Baylor U Medical Center, Dallas; J. Ortiz and C. Manzarbeita, Albert Einstein Medical Center, Philadelphia; R. Mennon and R.H. Wiesner, Mayo Clinic Transplant Center, Rochester; Y. Wu and R-Y Chenshu, University of Iowa Hospitals and Clinic, Iowa City; J.R. Lake and A. Humar, University of Minnesota, Minneapolis.

Ancillary