Journal Logo

Original Clinical Science—Liver

Improved Survival in Liver Transplant Patients Receiving Prolonged-release Tacrolimus-based Immunosuppression in the European Liver Transplant Registry (ELTR)

An Extension Study

Adam, René MD, PhD1; Karam, Vincent PhD1; Cailliez, Valérie1; Trunečka, Pavel MD, PhD2; Samuel, Didier MD, PhD1; Tisone, Giuseppe MD, PhD3; Němec, Petr MD, PhD4; Soubrane, Olivier MD, PhD5; Schneeberger, Stefan MD, PhD6; Gridelli, Bruno MD, PhD7; Bechstein, Wolf O. MD, PhD8; Risaliti, Andrea MD, PhD9; Line, Pal-Dag MD, PhD10; Vivarelli, Marco MD, PhD11; Rossi, Massimo MD, PhD12; Pirenne, Jacques MD, PhD13; Klempnauer, Jurgen L. MD, PhD14; Rummo, Aleh MD, PhD15; Di Benedetto, Fabrizio MD, PhD16; Zieniewicz, Krzysztof MD, PhD17; Troisi, Roberto MD, PhD18; Paul, Andreas MD, PhD19; Vali, Toomas MD, PhD20; Kollmar, Otto MD, PhD21; Boudjema, Karim MD, PhD22; Hoti, Emir MD, PhD23; Colledan, Michele MD, PhD24; Pratschke, Johan MD, PhD25; Lang, Hauke MD, PhD26; Popescu, Irinel MD, PhD27; Ericzon, Bo-Goran MD, PhD28; Strupas, Kestutis MD, PhD29; De Simone, Paolo MD, PhD30; Kochs, Eberhard MD, PhD31; Heyd, Bruno MD, PhD32; Gugenheim, Jean MD, PhD33; Pinna, Antonio D. MD, PhD34; Bennet, William MD, PhD35; Kazimi, Mirjalal MD, PhD36; Bachellier, Philippe MD, PhD37; Wigmore, Stephen J. MD, PhD38; Rasmussen, Allan MD, PhD39; Clavien, Pierre-Alain MD, PhD40; Hidalgo, Ernest MD, PhD41; O’Grady, John G. MD, PhD42; Zamboni, Frausto MD, PhD43; Kilic, Murat MD, PhD44; Duvoux, Christophe MD, PhD45 all contributing centers (www.eltr.org) and the European Liver and Intestine Transplant Association (ELITA)

Author Information
doi: 10.1097/TP.0000000000002700
  • Free

Abstract

Significant advances in the success of liver transplantation have been made over the past 2 decades, despite the increasing use of extended criteria donors.1 However, although 1-year graft and patient survival rates in Europe are 77% and 83%, respectively, after a first liver transplantation, 10-year graft and patient survival rates remain lower at 54% and 61%, respectively.2 Improving long-term liver transplant outcomes has, therefore, become a primary focus of the transplant community. Many factors can negatively influence outcomes in liver transplantation, including high Model for End-stage Liver Disease (MELD) scores and the viral status of the recipient (eg, hepatitis C virus [HCV] infection).3,4 Nonadherence to immunosuppressive therapy and high intrapatient variability of drug exposure could also potentially reduce long-term transplant outcomes.5-10

Tacrolimus is now the cornerstone of immunosuppression after liver transplantation.11 A once-daily, prolonged-release (PR) formulation of tacrolimus was licensed in Europe in 2007, for use in adult kidney or liver transplant recipients.12 PR tacrolimus has demonstrated comparable efficacy to immediate-release (IR) tacrolimus in clinical studies with de novo liver transplant recipients and is efficacious following conversion of stable liver transplant recipients from IR to PR tacrolimus.13-16 However, it is thought that the PR formulation may offer advantages over twice daily, IR tacrolimus, by reducing nonadherence to immunosuppressant medication and by decreasing intrapatient variability in tacrolimus exposure.17-19 As both of these parameters have been associated with poor liver transplant outcomes,8,20 treatment with PR tacrolimus has the potential to improve long-term outcomes for liver recipients, compared with the IR formulation. However, as clinical trials are of relatively short duration, there is a need for data to assess the effect of PR tacrolimus on long-term outcomes in liver transplantation. In this regard, registry studies can provide prospective and retrospective long-term data.

The impact of PR versus IR tacrolimus on long-term graft and patient survival was recently assessed using data from the European Liver Transplant Registry (ELTR) in 528 and 3839 patients receiving PR or IR tacrolimus, respectively.21 The study was conducted at 21 European centers between 2008 and 2012.21 Multivariable analyses showed that the use of IR tacrolimus was associated with a higher risk of graft loss (risk ratio, 1.81; P = 0.001) and patient death (risk ratio, 1.72; P = 0.004) compared with PR tacrolimus.21 We report here an analysis of up to 8-year data, collected between 2008 and 2016, in an extension of the previously published study, with the aim of comparing long-term liver transplant outcomes with PR versus IR tacrolimus–based immunosuppression. We also studied the outcome of converting patients from one formulation of the drug to the other, during the posttransplantation follow-up period.

MATERIALS AND METHODS

This was a retrospective analysis of primary liver transplant patients receiving PR tacrolimus (Advagraf; Astellas Pharma Europe BV, The Netherlands) and IR tacrolimus in the ELTR database, as previously described.21 The ELTR currently represents liver transplant data from 174 transplant centers across Europe. Data from participating centers are collected on a voluntary basis at regular intervals using a 2-part, standardized questionnaire designed by the ELTR Coordinating Committee to capture information on donors and recipients, as described previously.21 The methods used to populate the registry and obtain the data have been described elsewhere.22-24 To prevent center bias, only the 44 centers who used both PR and IR tacrolimus at the time of the study were eligible for inclusion in this analysis.

Inclusion Criteria

Data were collected prospectively from patients (aged ≥18 y) who underwent their first liver transplant between January 2008 and June 2016 from contributing centers across Europe. All patients included in this study received PR or IR tacrolimus, with or without concomitant immunosuppressants (including induction agents) within the first month after liver transplantation.

Clinical Efficacy Measures

Efficacy measures were analyzed using the modified intent-to-treat population, which excluded all patients who had <1 month of follow-up after transplantation. This strategy aimed to avoid the potential impact of early postoperative complications not associated with the immunosuppressive regimen. The clinical efficacy measures included univariate and multivariate analyses of the risk factors influencing graft and patient survival; Kaplan-Meier estimates of the incidence of graft and patient survival stratified by PR and IR tacrolimus–based immunosuppression, and causes of graft loss and mortality. Treatment groups were stratified by PR or IR tacrolimus treatment during the first month after transplantation, and patients remained in these allocated groups, regardless of any changes in immunosuppression during follow-up. However, crossover changes from IR to PR and vice versa after 1 month of therapy, regardless of the date of change(s), were considered to measure their impact on graft and patient outcomes. The maintenance immunosuppression data considered in this study were those collected at the last patient follow-up. The number of patients needed to treat with PR versus IR tacrolimus to avoid 1 graft loss in 4 years was calculated. To adjust for the number of patients at risk over the enrollment time between 2008 and 2016 (fewer patients at risk in the PR than IR tacrolimus group at 4 y owing to the gradual increase in the use of PR tacrolimus), the era of transplantation was added to the univariate and multivariate analyses.

Propensity Score Matching

To account for differences in donor and recipient baseline characteristics between groups when estimating the effect of treatment on outcomes, the clinical efficacy measures were repeated on a propensity score–matched population. PR and IR tacrolimus groups were paired on a 1:2 ratio according to 18 items with similar values. The propensity score was based on recipient age (≥60 versus <60 y), donor age (≥60 versus <60 y), full-size organ from a donor after brain death versus all other alternative grafts (living donor, domino, donation after circulatory death, or split grafts from a donor after brain death), MELD score (>24 versus ≤24), recipient hepatitis B virus (HBV) surface antigen, hepatocellular carcinoma (HCC), presence of severe ascites before liver transplantation, United Network for Organ Sharing (UNOS) status (3–4 versus 1–2), total ischemia time (≥6 versus <6 h), graft preservation solution histidine-tryptophan-ketoglutarate versus all other solutions (University Wisconsin, Celsior, IGL-1, Marshall, Ringer, Solution de conservation des Organes et des Tissus [SCOT], or other), and administration of other immunosuppressive medications early posttransplantation (corticosteroids, mycophenolate mofetil, ciclosporin, basiliximab, daclizumab, sirolimus, everolimus, azathioprine). For continuous variables that were converted to discontinuous variables in the model (eg, donor age, recipient age, MELD score), the values that were the most discriminant between PR and IR tacrolimus were selected as cutoffs, based on the calculation of chi-square value and odds ratios (data not shown). All unmatched units in the PR and IR tacrolimus groups were excluded from the propensity score–matched population.

Statistical Analyses

Statistical analyses were conducted as previously described.21 A univariate Cox regression analysis was performed to evaluate the risk factors influencing graft and patient survival after liver transplantation. Data from the univariate analyses were reported using log-rank P values, with P < 0.05 considered to be statistically significant. A Cox proportional hazards regression evaluation (P < 0.15) was used in a multivariate model to assess the impact of donor and recipient variables on graft and patient survival. Patients with missing data on the ELTR questionnaire were excluded from the multivariate analyses. Kaplan-Meier analyses were used to estimate graft and patient survival stratified by treatment group; statistical analyses were performed using the log-rank test (P < 0.05). Analyses were performed using SAS Enterprise Guide version 5.1 (SAS Institute, Inc., Cary, NC).

RESULTS

Donor and Recipient Characteristics and Demographics

Patient Population

In the initial ELTR study, 4367 primary liver transplant recipients (PR tacrolimus: n = 528; IR tacrolimus: n = 3839) were included in the analysis, between 2008 and 2012.21 In this extension analysis, 13 088 primary liver transplant recipients were included (Figure 1). All patients received either PR tacrolimus (n = 1762) or IR tacrolimus (n = 11 326). Since PR tacrolimus was licensed for use in 2007,12 the proportion of patients who received PR tacrolimus during month 1 gradually increased over enrollment (between 2008 and 2016).

FIGURE 1
FIGURE 1:
Patient populations. *Analysis only in centers using prolonged-release (PR) tacrolimus and immediate-release (IR) tacrolimus. Propensity score matching ratio 1:2, PR tacrolimus/IR tacrolimus. mITT, modified intent to treat.

Baseline Characteristics

Baseline characteristics of donors and recipients are presented in Table 1. Mean recipient age was greater in the PR versus IR tacrolimus group (52.1 ± 18.7 versus 51.8 ± 18.1 y, respectively; P < 0.001). The other main characteristics and their differences between groups are shown in Table 1.

TABLE 1
TABLE 1:
Baseline characteristics of donors and liver transplant recipients

Concomitant Medications

Baseline tacrolimus-associated induction immunosuppressive medications were different in the 2 groups: corticosteroids (69.2% versus 61.4%; P < 0.001), everolimus (7.5% versus 2.5%; P < 0.001), mycophenolate mofetil (75.1% versus 55.8%; P < 0.001), and daclizumab (1.8% versus 0.8%; P < 0.001) were more frequently combined with PR tacrolimus. Azathioprine (3.2% versus 0.5%; P < 0.001), ciclosporin (1.6% versus 0.2%; P < 0.001), basiliximab 25.4% versus 22.0%; P = 0.002), and sirolimus (0.6% versus 0.2%; P = 0.026) were more frequently combined with IR tacrolimus. However, propensity score matching has been used to account for these baseline differences in our study.

Analyses of Patients With ≥1 Month of Follow-up

Univariate Analyses

In the univariate analysis, IR tacrolimus during the first month posttransplantation was identified as a significant risk factor for inferior graft survival (P < 0.001) and patient survival (P = 0.003) over 8 years. Other factors that significantly contributed to reduced long-term graft and patient survival are listed in Table 2.

TABLE 2
TABLE 2:
Univariate analyses of risk factors for reduced graft and patient survival 1, 2, and 4 y posttransplantation, after exclusion of patients with <1 mo of follow-up (mITT population)

Kaplan-Meier Analyses

Kaplan-Meier analysis demonstrated significantly improved graft and patient survival over 4 years with PR versus IR tacrolimus (84% versus 79%; P < 0.001 and 85% versus 81%; P = 0.003, respectively) (Figure 2). At year 4, a 5% and 4% improvement in graft and patient survival, respectively, was observed in the PR versus IR tacrolimus group.

FIGURE 2
FIGURE 2:
Kaplan-Meier analyses of (A) graft and (B) patient survival over 4 y of treatment with prolonged-release (PR) tacrolimus compared with immediate-release (IR) tacrolimus, after exclusion of patients with <1 mo of follow-up (modified intent-to-treat population).

Propensity Score–matched Analyses

The propensity score–matched analysis was performed on 3006 patients (PR tacrolimus: n = 1002; IR tacrolimus: n = 2004). Donor and recipient baseline characteristics were generally comparable between the 2 treatment groups for the propensity score–matched patients, especially for the concomitant immunosuppressive drugs combined with tacrolimus (Table 1).

Univariate and Multivariate Analyses

In the univariate analysis, the use of IR tacrolimus was a significant risk factor for reduced graft and patient survival (P = 0.005 and P = 0.017, respectively) in addition to other factors listed in Table 3. Long-term graft survival was significantly impacted by 13 additional factors: donor age ≥50 years (P < 0.001), recipient age ≥50 years (P = 0.006), recipient dialysis twice in week before transplantation (P = 0.016), negative HBV delta (P = 0.017), positive anti-HCV serology (P < 0.001), positive human immunodeficiency virus serology (P = 0.028), positive HCV RNA (P < 0.001), urgent liver transplant (P = 0.038), UNOS status 1 or 2 (P < 0.001), serum creatinine concentration ≥2 mg/dL (P = 0.01), Milan criteria-out in patients with HCC (P < 0.001), HCC with tumor size >50 mm (P < 0.001), and heterotopic liver transplant (P = 0.003).

TABLE 3
TABLE 3:
Univariate analyses of risk factors for reduced graft and patient survival for propensity score–matched patients

Long-term patient survival was significantly impacted by 15 other factors: donor age ≥50 years (P < 0.001), presence of macro/microvesicular graft steatosis (P = 0.039), male recipient (P = 0.021), recipient age ≥50 years (P = 0.001), recipient dialysis twice in week before transplantation (P = 0.006), negative HBV delta (P = 0.025), positive anti-HCV serology (P < 0.001), positive human immunodeficiency virus serology (P = 0.045), positive HCV RNA (P < 0.001), UNOS status 1 or 2 (P < 0.001), serum creatinine concentration ≥2 mg/dL (P = 0.001), cancer as main indication (P = 0.032), Milan criteria-out in patients with HCC (P < 0.001), HCC with tumor size >50 mm (P < 0.001), and heterotopic liver transplant (P = 0.001).

In the multivariate analysis (Table 4), the use of IR tacrolimus was a significant independent risk factor for reduced graft survival (risk ratio: 1.49; 95% confidence interval [CI]: 1.14-1.96; P = 0.0038) associated with 5 other factors: recipient positive anti-HCV serology (risk ratio: 2.05; 95% CI: 1.66-2.54; P < 0.001), recipient age ≥50 years (risk ratio: 1.74; 95% CI: 1.43-2.11; P < 0.001), UNOS status 1 or 2 (risk ratio: 1.69; 95% CI: 1.35-2.11; P < 0.001), serum creatinine concentration ≥2 mg/dL (risk ratio: 1.66; 95% CI: 1.18-2.35; P = 0.004), and donor age ≥50 years (risk ratio: 1.35; 95% CI: 1.09-1.66; P = 0.005).

TABLE 4
TABLE 4:
Multivariate analyses of risk factors for reduced graft and patient survival for the propensity score–matched patients

The use of IR tacrolimus was also a significant independent risk factor for reduced patient survival (risk ratio, 1.40; 95% CI, 1.05-1.86; P = 0.0215) associated with 6 other factors: recipient positive anti-HCV serology (risk ratio, 1.91; 95% CI, 1.52-2.40; P < 0.001), serum creatinine concentration ≥2 mg/dL (risk ratio, 1.90; 95% CI, 1.33-2.71; P < 0.001), UNOS status 1 or 2 (risk ratio, 1.89; 95% CI, 1.49-2.39; P < 0.001), recipient age ≥50 years (risk ratio, 1.74; 95% CI, 1.41-2.16; P < 0.001), HCC as primary or secondary disease (risk ratio, 1.35; 95% CI, 1.07-1.69; P = 0.01), and donor age ≥50 years (risk ratio, 1.33; 95% CI, 1.07-1.66; P = 0.01).

Kaplan-Meier Analyses and Number of Patients Needed to Treat to Avoid 1 Graft Loss

Kaplan-Meier analysis demonstrated significantly improved graft and patient survival over 4 years with PR versus IR tacrolimus (83% versus 77%; P = 0.005 and 85% versus 80%; P = 0.017, respectively) (Figure 3). At year 4, a 6% and 5% improvement in graft and patient survival, respectively, was observed in the PR versus IR tacrolimus group. The number of patients needed to treat with PR versus IR tacrolimus to avoid 1 graft loss in 4 years was 14.3 patients (95% CI, 9.7-27.3).

FIGURE 3
FIGURE 3:
Kaplan-Meier analyses of (A) graft and (B) patient survival over 4 y of treatment with prolonged-release (PR) tacrolimus compared with immediate-release (IR) tacrolimus for the propensity score–matched patients.

Analysis of Crossover Groups

In the nonconverted patients over the study period, graft and patient survival were significantly higher with (induction-last follow-up immunosuppressive regimen available) PR-PR than with IR-IR tacrolimus (88% versus 82% at 4 y; P = 0.019 and 89% versus 83% at 4 y; P = 0.047, respectively) (Figure 4). Patients converted from IR to PR tacrolimus after 1 month had a significantly higher graft and patient survival rate compared with patients who were started on and still receiving PR tacrolimus at the last follow-up (92% versus 88% at 4 y; P = 0.019 and 94% versus 89% at 4 y; P = 0.004, respectively) or started on and still receiving IR tacrolimus at the last follow-up (P < 0.001 for both).

FIGURE 4
FIGURE 4:
Kaplan-Meier analyses of (A) graft and (B) patient survival over 4 y of treatment in crossover groups for the propensity score–matched patients. IR, immediate release; PR, prolonged release.

Causes of Graft Loss and Mortality

The most common cause of graft loss was infection in both groups (Table 5). Over 8 years of treatment, the proportion of patients with bacterial infection that resulted in graft loss was higher with PR versus IR tacrolimus (17.2% versus 9.2%, respectively; P = 0.031) (Table 5). Compared with patients receiving IR tacrolimus, “other” causes of graft loss were less frequent in patients receiving PR tacrolimus (P = 0.01). However, there were no significant differences between groups in the incidence of graft loss due to acute or chronic rejection, cardiovascular, or renal causes.

TABLE 5
TABLE 5:
Causes of graft loss and mortality over 8 years of treatment for the propensity score–matched patients

The most common cause of patient mortality was infection in both groups (Table 5). The proportion of patients with bacterial infection that resulted in patient death was similar with PR and IR tacrolimus (18.3% versus 10.4%, respectively; P = 0.057). “Other” causes of mortality were less frequent in patients receiving PR versus IR tacrolimus (P = 0.005). There were no significant differences between treatment groups in the proportion of patients with cardiovascular or renal causes of mortality.

DISCUSSION

The initial ELTR study was the first large retrospective registry study in Europe, evaluating PR tacrolimus–based immunosuppression in primary liver transplantation.21 The study showed that the use of IR tacrolimus was associated with a higher risk of graft loss and patient death, compared with PR tacrolimus. Additionally, PR tacrolimus significantly improved graft and patient survival over 3 years posttransplantation, compared with the IR formulation.21 However, the cohort size was relatively limited, as was the length of follow-up.

This extension to the ELTR study, reporting up to 8-year data from adult primary liver transplant recipients, confirmed that PR tacrolimus was associated with improved graft and patient survival (over 4 y posttransplantation), compared with IR tacrolimus. Consistent with Adam et al,21 IR tacrolimus was an independent risk factor for graft loss and mortality over 8 years of treatment. In addition to the longer time period assessed, the number of patients included in the current study exceeded 13 000, compared with the 4367 patients included in the initial ELTR study.21 This provided enhanced statistical robustness, increasing the reliability of the results.

Univariate and multivariate analyses substantiated the independent prognostic value of typical risk factors, including donor age (≥50 y), recipient viral status (HCV positivity), and UNOS status 1 or 2 in impairing month 1 to year 8 graft and patient survival.25,26 As reported by Adam et al,21 IR tacrolimus was also identified in this study as a significant predictor of graft loss and patient death in univariate analyses. Furthermore, after accounting for differences in baseline characteristics between treatment groups, IR tacrolimus formulation remained a significant predictor of graft and patient loss, in both univariate and multivariate analyses.

In the previous ELTR study, improvements in graft and patient survival with PR versus IR tacrolimus were observed as early as 3 months after transplantation and continued over 3 years.21 Our current data demonstrate that the survival benefit associated with PR tacrolimus continues over 4 years of treatment. Furthermore, while improved 3-year patient survival with PR versus IR tacrolimus did not reach statistical significance for the unmatched patient cohort in the initial ELTR study,21 the benefit was statistically significant by year 4 in our current study. Indeed, there was a 5% and 4% graft and patient survival advantage, respectively, by year 4 with the PR versus IR formulation. The difference between the PR and IR groups for graft and patient survival rates also seemed to increase with time.

Consistent with data reported in a 4-year follow-up of de novo liver transplant recipients from a phase II study,27 PR tacrolimus was associated with 4-year graft and patient survival rates of ~90%. As a complement to our previous study, we also evaluated the impact of crossover changes from IR to PR and vice versa after 1 month of induction therapy with regards to graft and patient outcomes. Both graft and patient survival were higher in patients who converted from IR to PR tacrolimus, compared with those who received induction PR tacrolimus and were still receiving PR tacrolimus at last follow-up. Although the cause of this is unclear, the data suggest that the use of PR tacrolimus at last follow-up therapy is associated with improved outcomes, irrespective of the timing of conversion. The transplant community is now interested in identifying whether earlier conversion (<6 mo after liver transplantation) from IR to PR tacrolimus is associated with better outcomes than conversion >6 months posttransplantation.28

The survival advantages observed in patients treated with PR versus IR tacrolimus reported at 3 years in Adam et al,21 and at 4 years in this study were not observed in short-term, randomized, controlled trials. For example, Trunečka et al29 reported 12-month graft survival rates of 85.3% and 85.6%, and patient survival rates of 89.2% and 90.8%, with PR and IR tacrolimus, respectively. The potential survival advantages associated with PR tacrolimus in de novo liver transplant recipients may, therefore, not become apparent until beyond 1-year posttransplantation.

In an independent editorial that accompanied the initial ELTR study, Asrani and O’Leary30 considered the potential mechanisms underlying the improvement in long-term graft and patient survival with PR versus IR tacrolimus-based immunosuppression. Compared with IR tacrolimus, PR formulation reduces variability of tacrolimus exposure19,31 and offers a simpler regimen comprising a single, morning dose,12 which can improve medication adherence.17,18 Indeed, in an expert literature review, improved adherence with treatment is highlighted as a main advantage of PR versus IR tacrolimus.28 Given that high intrapatient variability in tacrolimus exposure and medication nonadherence have been associated with poor transplant outcomes,8,20 PR tacrolimus may improve long-term graft and patient survival compared with the IR formulation.

As observed in the initial ELTR study,21 the overall proportion of patients with graft loss was lower in the PR versus IR tacrolimus group. The reasons for graft loss and mortality were generally comparable between groups in this study and, consistent with the previous ELTR study, infections were the most frequent cause.21

In this study, not all factors could be controlled by the transplant team to improve outcomes. Only the type of preservation solution, the ischemia time, and the immunosuppressive regimen used could be altered, as donor and recipient characteristics cannot be changed in the MELD allocation system used in most countries. Therefore, it is important to consider the numbers of patients needed to treat with PR tacrolimus to avoid 1 graft loss. In this study, 14.3 patients needed to be treated with PR versus IR tacrolimus to avoid 1 graft loss in 4 years. These data are consistent with those of Muduma et al,32 who developed a model using UK liver transplant data and showed that, over a 3-year time period, 1 graft would be saved for every 14 patients treated with PR versus IR tacrolimus, with minimal impact on costs. To place these data in clinical context, 15 patients needed to be treated with nicotine replacement therapy for 1 patient to cease smoking,33 20 required treatment with calcium and vitamin D for 3 years to prevent 1 hip fracture,34 23 required treatment with flu vaccine to prevent 1 flu episode,35 and 35 needed primary treatment with statins for 5 years to prevent a cardiac event.36

Despite the improvements to our study design and analytical methods, any conclusions drawn from our findings must be made within the context of the limitations of our study, which have been described in detail previously.21 These include the retrospective nature of the study and the long period over which data were collected, which may be associated with changes in clinical practice. Furthermore, our study design carries a risk of bias in terms of patient and treatment selection. To control for these differences, propensity score matching was undertaken against a larger number of characteristics compared with the initial ELTR study. However, it is recognized that propensity score matching can only be used to balance measured variables and cannot entirely exclude inherent differences, such as socioeconomic factors, ethnicity, or other unknown variables. A major limitation concerns the lack of data on drug exposure, as the dose and trough levels of tacrolimus were not captured in the ELTR; it is also not known which IR tacrolimus preparation patients were receiving. Despite these limitations, to our knowledge, we report on the largest population of liver transplant recipients to date, comparing the impact of PR and IR tacrolimus administration. Furthermore, this analysis builds on our previous publication by including a 3-fold larger cohort, providing extended follow-up, evaluating the impact of crossover between IR and PR tacrolimus therapy, and reporting the clinical implication of the results in terms of number needed to treat to avoid 1 graft loss.

Our results, based on up to 8-year data, confirm observations from the initial 3-year study21 that PR tacrolimus-based immunosuppression can improve long-term outcomes in liver transplantation compared with IR tacrolimus. Furthermore, IR tacrolimus–based immunosuppression is a significant predictor of long-term graft loss and patient mortality. Conversion from IR to PR tacrolimus after 1 month was also associated with a better outcome compared with maintaining patients on IR tacrolimus–based immunosuppression, or starting and maintaining patients on PR tacrolimus–based immunosuppression. Importantly, our findings confirm that PR tacrolimus continues to provide ongoing benefits for graft and patient survival beyond 3 years posttransplantation.

ACKNOWLEDGMENTS

Thanks to all the centers who contribute to the ELTR. The Organ Sharing Organizations the French ABM (Sami Djabbour and Alain Jolly), the Dutch NTS (Cynthia Konijn), the Eurotransplant Foundation (Marieke Van Meel and Erwin de Vries), the Spanish ONT (Gloria de la Rosa), and the UK-Ireland NHSBT (Mike Chilton and Julia Micciche) are acknowledged for the data cross-check and sharing with the ELTR. Astellas is acknowledged for the initial idea of the analysis and for participation in further scientific discussion. The authors also acknowledge Daniella T. Draper, PhD, CMPP, and Anna Thompson, PhD, from Cello Health MedErgy (Europe) assisted in drafting the initial version of the manuscript under the direction of the authors and provided editorial support throughout its development.

REFERENCES

1. European Liver Transplant Registry. Donor data. Available at www.eltr.eu. 2015. Accessed February 2019
2. European Liver Transplant Registry. Mortality and retransplantation. Available at www.eltr.eu. 2015. Accessed July 2019
3. Gambato M, Frigo AC, Rodríguez Castro KI, et al. Who fares worse after liver transplantation? Impact of donor and recipient variables on outcome: data from a prospective study. Transplantation. 2013; 95:1528–1534
4. Onaca NN, Levy MF, Sanchez EQ, et al. A correlation between the pretransplantation MELD score and mortality in the first two years after liver transplantation. Liver Transpl. 2003; 9:117–123
5. Mor E, Gonwa TA, Husberg BS, et al. Late-onset acute rejection in orthotopic liver transplantation–associated risk factors and outcome. Transplantation. 1992; 54:821–824
6. Berlakovich GA, Langer F, Freundorfer E, et al. General compliance after liver transplantation for alcoholic cirrhosis. Transpl Int. 2000; 13:129–135
7. O’Carroll RE, McGregor LM, Swanson V, et al. Adherence to medication after liver transplantation in Scotland: a pilot study. Liver Transpl. 2006; 12:1862–1868
8. Lieber SR, Volk ML. Non-adherence and graft failure in adult liver transplant recipients. Dig Dis Sci. 2013; 58:824–834
9. Cervelli M, Russ G. Intrapatient variability with tacrolimus. Aust J Pharm. 2012; 93:83–86
10. Burra P, Germani G, Gnoato F, et al. Adherence in liver transplant recipients. Liver Transpl. 2011; 17:760–770
11. Kim WR, Lake JR, Smith JM, et al. OPTN/SRTR 2015 annual data report: liver. Am J Transplant. 2017; 17Suppl 1174–251
12. Advagraf European Public Assessment Reports (EPAR). European Medicines Agency. Available at http://www.ema.europa.eu/docs/en_GB/document_library/EPAR_-_Scientific_Discussion/human/000712/WC500022237.pdf. 2007. Accessed February 2019
13. TruneČka P, Klempnauer J, Bechstein WO, et al; DIAMOND† Study Group. Renal function in de novo liver transplant recipients receiving different prolonged-release tacrolimus regimens-the DIAMOND study. Am J Transplant. 2015; 15:1843–1854
14. Dumortier J, Guillaud O, Boillot O. Conversion from twice daily tacrolimus to once daily tacrolimus in long-term stable liver transplant recipients: a single-center experience with 394 patients. Liver Transpl. 2013; 19:529–533
15. Shamilin S, Ranju M, Bismi ES, et al. A comparative study of once daily versus twice daily tacrolimus in liver transplantation. J Young Pharm. 2017; 9:605–609
16. Altieri M, Delaval G, Kimmoun E, et al. Conversion from once-daily prolonged-release tacrolimus to once-daily extended-release tacrolimus in stable liver transplant recipients. Exp Clin Transplant. 2018; 16:321–325
17. Beckebaum S, Iacob S, Sweid D, et al. Efficacy, safety, and immunosuppressant adherence in stable liver transplant patients converted from a twice-daily tacrolimus-based regimen to once-daily tacrolimus extended-release formulation. Transpl Int. 2011; 24:666–675
18. Valente G, Rinaldi L, Sgambato M, et al. Conversion from twice-daily to once-daily tacrolimus in stable liver transplant patients: effectiveness in a real-world setting. Transplant Proc. 2013; 45:1273–1275
19. Sańko-Resmer J, Boillot O, Wolf P, et al. Renal function, efficacy and safety postconversion from twice- to once-daily tacrolimus in stable liver recipients: an open-label multicenter study. Transpl Int. 2012; 25:283–293
20. Christina S, Annunziato RA, Schiano TD, et al. Medication level variability index predicts rejection, possibly due to nonadherence, in adult liver transplant recipients. Liver Transpl. 2014; 20:1168–1177
21. Adam R, Karam V, Delvart V, et al; European Liver Intestine Transplant Association (ELITA). Improved survival in liver transplant recipients receiving prolonged-release tacrolimus in the European Liver Transplant Registry. Am J Transplant. 2015; 15:1267–1282
22. Adam R, Karam V, Delvart V, et al; All contributing centers (www.eltr.org); European Liver and Intestine Transplant Association (ELITA). Evolution of indications and results of liver transplantation in Europe. A report from the European Liver Transplant Registry (ELTR). J Hepatol. 2012; 57:675–688
23. Jurewicz WA. Tacrolimus versus cyclosporin immunosuppression: long-term outcome in renal transplantation. Nephrol Dial Transplant. 2003; 18(Suppl 1i7–i11
24. Burra P, Senzolo M, Adam R, et al; ELITA; ELTR Liver Transplant Centers. Liver transplantation for alcoholic liver disease in Europe: a study from the ELTR (European Liver Transplant Registry). Am J Transplant. 2010; 10:138–148
25. Cuervas-Mons V, de la Rosa G, Pardo F, et al; en representación del Registro Español de Trasplante Hepático. [Activity and results of liver transplantation in spain during 1984-2012. Analysis of the Spanish Liver Transplant Registry]. Med Clin (Barc). 2015; 144:337–347
26. Gwiasda J, Schrem H, Klempnauer J, et al. Identifying independent risk factors for graft loss after primary liver transplantation. Langenbecks Arch Surg. 2017; 402:757–766
27. van Hooff JP, Alloway RR, Trunečka P, et al. Four-year experience with tacrolimus once-daily prolonged release in patients from phase II conversion and de novo kidney, liver, and heart studies. Clin Transplant. 2011; 25:E1–E12
28. Coilly A, Calmus Y, Chermak F, et al. Once-daily prolonged release tacrolimus in liver transplantation: experts’ literature review and recommendations. Liver Transpl. 2015; 21:1312–1321
29. Trunečka P, Boillot O, Seehofer D, et al; Tacrolimus Prolonged Release Liver Study Group. Once-daily prolonged-release tacrolimus (ADVAGRAF) versus twice-daily tacrolimus (PROGRAF) in liver transplantation. Am J Transplant. 2010; 10:2313–2323
30. Asrani SK, O’Leary JG. Can one pill a day keep rejection away? Am J Transplant. 2015; 15:1135–1136
31. Florman S, Alloway R, Kalayoglu M, et al. Conversion of stable liver transplant recipients from a twice-daily prograf-based regimen to a once-daily modified release tacrolimus-based regimen. Transplant Proc. 2005; 37:1211–1213
32. Muduma G, Odeyemi I, Pollock RF. Evaluating the cost-effectiveness of prolonged-release tacrolimus relative to immediate-release tacrolimus in liver transplant patients based on data from routine clinical practice. Drugs Real World Outcomes. 2016; 3:61–68
33. Green G. Nicotine replacement therapy for smoking cessation. The NNT Group. Available at www.thennt.com/nnt/nicotine-replacement-therapy-for-smoking-cessation. May 1, 2013. Accessed July 20, 2011
34. Bandolier. Calcium and vitamin D for preventing hip fractures. Available at http://www.bandolier.org.uk/Conflicts%20Folder/b37-4.html. Accessed February 2019
35. Bandolier. Influenza vaccination in the elderly. Available at http://www.bandolier.org.uk/band11/b11-5.html. Accessed February 2019
36. Bandolier. Statins. Available at http://www.bandolier.org.uk/band47/b47-2.html. 1998. Accessed February 2019
Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.