Renal safety of tenofovir disoproxil fumarate and entecavir in liver transplant patients: a nationwide Korean registry study

Tenofovir disoproxil fumarate (TDF) and entecavir (ETV) have been recommended after liver transplantation to prevent recurrence of hepatitis B virus infection. Despite its proven efficacy, the renal safety of TDF has not been established in liver transplant recipients. We aimed to compare the effects of TDF and ETV on renal function in liver transplant recipients and to evaluate risk factors for renal dysfunction after liver transplantation. This is a retrospective, observational multicenter study of data from the Korean Organ Transplantation Registry. We included adults who underwent liver transplantation for hepatitis B virus-related complications from April 2014 to December 2017 and received TDF or ETV post-transplantation. Renal dysfunction was defined as an estimated glomerular filtration rate decline by at least 20% from baseline (1 month post-transplantation). Median duration of follow-up was 29 months (interquartile range 19–42). A total of 804 liver transplant patients were included. The cumulative probability of renal dysfunction was significantly higher in the TDF group than in the ETV group. Multivariable analysis confirmed that TDF was independently associated with an increased risk of renal dysfunction (hazard ratio = 1.47, 95% confidence interval 1.12–1.92; p = 0.005). Independent risk factors for renal dysfunction included older age, worse baseline renal function, and low body mass index. Overall survival rate was significantly lower in patients with renal dysfunction than in those without. In this nationwide study, the use of TDF was associated with an increased risk of renal dysfunction, when compared with ETV.


Introduction
Hepatitis B virus (HBV) infection remains one of the most important causes of severe liver disease requiring consideration of liver transplantation (LT) in the world [1,2]. Before the advent of effective antiviral prophylaxis, HBV infection was considered a relative contraindication for LT because of poor graft outcomes associated with HBV recurrence [3]. The introduction of nucleos(t)ide analogues (NAs) and hepatitis B immune globulin (HBIG) has significantly decreased 1 3 post-transplant HBV recurrence and improved long-term graft survival [4].
Combinations of HBIG and potent NAs have been recommended to prevent recurrent HBV after LT [5,6]. Entecavir (ETV) and tenofovir disoproxil fumarate (TDF), which have greater potency and higher genetic barriers to resistance than other agents, are first-line NAs recommended by current international guidelines [7,8]. Although TDF and ETV are generally well tolerated, renal dysfunction remains a concern with TDF [9][10][11]. Prior studies in human immunodeficiency virus (HIV)-infected patients have indicated that TDF is associated with a small but significant decline in renal function [12]. By contrast, HBV cohort studies have demonstrated comparable effects on renal function from TDF and ETV [13,14]. However, the majority of studies evaluated patients with preserved renal function and excluded liver transplant recipients.
In the setting of LT, one must always consider the possibility of renal dysfunction because of the concomitant use of calcineurin inhibitors [15,16]. Furthermore, many liver transplant patients exhibit renal dysfunction at the time of transplant under the recent model for end-stage liver disease (MELD) score-based allocation system [17]. However, the renal effects of TDF and ETV in liver transplant recipients have not been well studied [18,19]. In the present study, we evaluated the renal safety of TDF and ETV after LT using nationwide registry data.

Study design and participants
The Korean Organ Transplantation Registry (KOTRY) is a national, multicenter, observational cohort study containing prospectively collected data from solid organ transplant patients. The KOTRY liver transplant database includes demographic information and clinical data, including information regarding immunosuppressive regimens and HBV prophylaxis. Detailed descriptions of collected data have been previously described [20].
For this retrospective study, we retrieved de-identified patient data from the KOTRY liver transplant database and screened all adults who underwent LT for HBV-related complications from April 2014 to December 2017. Patients who received TDF or ETV were included in this study. These exclusion criteria were applied: (1) coinfection with hepatitis C, (2) use of other antiviral agents, (3) delayed use of TDF or ETV (initiated > 1 month after transplantation), (4) baseline (1 month post-LT) estimated glomerular filtration rate (eGFR) < 30 mL/min/1.73m 2 , and (5) missing data (Fig. 1).
This cohort study was based on data from the KOTRY liver transplant database. The KOTRY study protocol satisfied the ethical guidelines of the 1975 Declaration of Helsinki and was approved by the Institutional Review Board of Severance Hospital and at each participating institution. The requirement for written informed consent was waived because of the retrospective nature of this study.

Hepatitis B virus prophylaxis
All patients were treated with TDF or ETV to prevent HBV recurrence. The antiviral agent dosage was adjusted for the patient's renal function during the study period (TDF: 300 mg oral dose every 48 h for an eGFR of 30-49 mL/ min/1.73 m 2 and 300 mg every 72 h for an eGFR of 10-29 mL/min/1.73 m 2 ; ETV: 0.5 mg oral dose every 48 h for an eGFR of 30-49 mL/min/1.73 m 2 and 0.5 mg every 72 h for an eGFR of 10-29 mL/min/1.73 m 2 ). Because reimbursement criteria for ETV and TDF were identical, the antiviral agents were selected at the physician's discretion.
Most patients received HBIG in combination with an antiviral agent, in accordance with institutional protocols. Briefly, HBIG was administered at 10,000-20,000 IU during the anhepatic phase, 10,000-20,000 IU daily for the first 7 days after surgery (20,000 IU for HBV e antigen positive or HBV DNA positive recipients), and then 10,000 IU weekly for the next 3 weeks. Thereafter, 10,000 IU of HBIG was infused once per month to maintain antibodies to HBV surface antigen > 500 IU/L.

Immunosuppressive regimens
Most patients received basiliximab for induction therapy (20 mg on days 0 and 4 post-LT). In the majority of patients, the maintenance regimen was tacrolimus and glucocorticoid, with or without mycophenolate mofetil (MMF).

Study endpoints and definitions
The primary study endpoint was renal dysfunction, and the secondary endpoint was patient survival. Renal dysfunction was defined as an eGFR decline of at least 20% from baseline that persisted until the end of follow-up (at the end of the study period, when the data were analysed) [21,22]. Transient dysfunction with the recovery of renal function during the study period was not counted as an event. Baseline eGFR was defined as the eGFR at 1 month after LT. eGFR was calculated using the Chronic Kidney Disease Epidemiology Collaboration equation [23]. Patient survival was evaluated from the date of transplantation to the date of death or December 31, 2019 (end of the follow-up period).

Statistical analysis
Results were expressed as frequency (percentage), mean and standard deviation, or median and interquartile range (IQR), depending on the type of data. The Chi-square test was used to compare categorical variables. Continuous variables were compared using Student's t test, for parametric data, or the Wilcoxon rank-sum test, for nonparametric data. Cox proportional hazard regression analysis was used to evaluate the associations between the use of TDF and time to renal dysfunction. Patient survival was analysed using the Kaplan-Meier method and compared using the log-rank test. All statistical analyses were performed using SAS version 9.4 (SAS, Inc., Cary, NC, USA) and SPSS version 25.0 (SPSS Inc., Chicago, IL, USA), and p values < 0.05 were considered statistically significant.

Patients characteristics
During the study period, 1444 patients who underwent LT for HBV were screened. Of these, 640 were excluded because of hepatitis C virus coinfection, use of other antiviral agents, delayed use of antiviral agents (initiated > 1 month after transplantation), baseline eGFR < 30 mL/min/1.73 m 2 , or missing data. In total, 804 recipients who received TDF or ETV for antiviral prophylaxis after LT were included in the final analysis.
Recipient and donor characteristics are presented in Table 1. No significant differences were observed between groups for age, sex, body mass index (BMI), the proportion of living donors, or proportions of patients with hypertension or diabetes mellitus. The proportion of

Immunosuppression
Immunosuppression data are presented in

Renal outcomes
During the median follow-up period of 29 months, 363 (45.1%) patients experienced a significant decline in eGFR, with a decrease of at least 20% below baseline. In 128, renal function recovered before the end of follow-up (i.e., transient dysfunction), whereas 235 (29.2%) patients had persistent renal dysfunction. The cumulative probability of renal dysfunction was significantly higher in the TDF group than in the ETV group (Fig. 2) On univariable analysis, use of TDF was significantly associated with an increased risk of renal dysfunction. After adjusting for age, diabetes, BMI, immunosuppression, and baseline renal function, use of TDF was independently associated with an increased risk of renal dysfunction [hazard ratio (HR) = 1.47, 95% confidence interval (CI) 1.12-1.92; p = 0.005]. Older age (HR 1.45, 95% CI 1.07-1.96, p = 0.016), worse baseline renal function (HR 0.98, 95% CI 0.97-0.99, p < 0.001), and low BMI (HR 1.70, 95% CI 1.30-2.24, p = 0.002) were also independent risk factors for renal dysfunction (Table 3).

Patient survival
As shown in Fig. 3, renal dysfunction was associated with a significant reduction in overall patient survival (p < 0.001).
Overall survival rates of patients with and without renal dysfunction were 88.9% and 96.8%, respectively, at 36 months post-LT.
Overall patient survival was comparable in the TDF and ETV groups. At 36 months after LT, the overall survival rate was 95.3% in the TDF group and 93.4% in the ETV group (p = 0.349).

Discussion
ETV and TDF are the two first-line NAs for preventing HBV recurrence after LT. However, limited data exist regarding the renal safety of TDF and ETV in liver transplant recipients, who are more susceptible to nephrotoxicity [7,8,18]. In the present study, we evaluated the renal effects of TDF and ETV in 804 liver transplant patients. Multivariable analysis showed that the risk of renal dysfunction was significantly higher in the TDF group than in the ETV group, and patients who developed renal dysfunction had higher mortality than those who did not. To our knowledge, this is the first nationwide cohort study evaluating the renal safety of TDF and ETV in liver transplant patients.
TDF, a prodrug of tenofovir diphosphate, is widely prescribed for the treatment of HIV and HBV infections. Although TDF is generally well tolerated, there are increasing concerns about its nephrotoxicity [9][10][11]. The mechanism of TDF nephrotoxicity involves mitochondrial toxicity in renal proximal tubule cells [24]. There have been conflicting results regarding the effects of TDF on renal function. In HIV patients, the authors of a meta-analysis concluded that patients treated with TDF experienced a small but significant loss of renal function during treatment, compared with controls [12]. However, findings in HIV patients should be extrapolated to other populations with caution because of HIV-associated nephropathy and concomitant use of antiretroviral agents [25].
In HBV patients, several randomized clinical trials have shown that TDF has no significant effects on renal function [26,27]. In contrast, some real-world observational studies reported that TDF significantly increases the risk of renal dysfunction, compared with ETV [28,29]. Possible reasons for these discrepancies include the use of different definitions for renal dysfunction and differences in inclusion/exclusion criteria. As a reflection of these conflicting  results, recent international guidelines for selecting antiviral agents differ in their recommendations related to renal safety issues. The European Association for the Study of the Liver guidelines recommended selecting ETV or tenofovir alafenamide fumarate (TAF) over TDF for patients with an old age (> 60 years), eGFR < 60 mL/min per 1.73 m 2 , or albuminuria, as well as for patients receiving hemodialysis [7]. By contrast, the American Association for the Study of Liver Disease guidelines suggests no preference for TDF or ETV regarding renal safety issues [8]. Of note, there is a distinct paucity of data regarding the renal safety of TDF and ETV in liver transplant recipients, who are particularly susceptible to nephrotoxicity because of the concomitant use of calcineurin inhibitors. Considering that liver transplant patients with HBV require lifelong use of antiviral agents in combination with calcineurin inhibitors, there is an urgent need for comprehensive research regarding the renal safety of these agents in this patient population [18,19].
Our results suggest that TDF significantly increases the risk of renal dysfunction after LT. Previous studies revealed that older age, diabetes mellitus, low BMI, worse baseline renal function, and use of nephrotoxic agents are associated with TDF-induced nephrotoxicity [24,30]. In agreement with these findings, we also found that older age, low BMI, and worse baseline renal function were significantly associated with renal dysfunction in liver transplant recipients. Therefore, TDF should be used with caution in these patients.
Renal dysfunction is a common and serious complication after LT [15,16]. Despite advances in perioperative management and reduction of calcineurin inhibitor exposure, the risk of renal dysfunction following LT has substantially increased in the MELD era [17]. Our data showed that patients who developed renal dysfunction after LT had a significantly lower overall survival rate than those who did not. Patients using TDF are at increased risk of post-LT renal dysfunction and therefore need meticulous follow-up.
Despite its higher risk of renal dysfunction, TDF cannot be totally replaced by ETV [27]. Prior exposure to other NAs increases ETV resistance, whereas TDF is effective for treating HBV strains resistant to ETV, lamivudine, telbivudine, or even multiple drugs [7,8,31]. Although TAF has shown better renal safety than TDF, limited data exist regarding its use in liver transplant recipients [32]. Thus, TDF continues to have an important role in the treatment of liver transplant patients, and more attention should be given to patients at risk for renal dysfunction.
The present study has several limitations. First, as with any registry study, selection bias and confounders are the major limitations of this study. Although reimbursement criteria for ETV and TDF are identical in South Korea, patients at higher risk for renal dysfunction might be more likely to receive ETV rather than TDF. In addition, it is difficult to determine the specific effects of TDF on renal dysfunction in liver transplant patients receiving calcineurin inhibitors. To minimize potential confounding factors as much as possible, we adjusted for tacrolimus serum trough levels, as well as other well-known risk factors for renal dysfunction. Second, this study included a large number of living donor transplant recipients with a low MELD Na score. Thus, our results may not be generalizable to deceased donor transplant patients with a high MELD Na score. Third, we were unable to obtain data about tubular dysfunction, such as urinary β2 microglobulin and fractional excretion of phosphate. However, these tests are not commonly used in real-world practice. Fourth, renal dysfunction was defined as a persistent eGFR decline of at least 20% from baseline. Declines in eGFR have been suggested as surrogate endpoints for kidney disease progression and have been widely used in recent clinical trials [22]. Previous studies evaluating TDF-associated renal outcomes also used cutoff of 20-25% decline of eGFR [21,28,29].
Despite these limitations, the study design involving the use of data from a nationwide registry has enabled us to analyse a large patient cohort. This study includes realworld clinical data in liver transplant patients receiving first-line antiviral agents with calcineurin inhibitors. Previous scientific evidence is limited regarding the renal safety of TDF in this patient population, which has resulted in inconsistencies between recent international guidelines. Thus, our data provide valuable insights into the renal safety of TDF in liver transplant recipients.
In conclusion, the use of TDF in liver transplant recipients was associated with a higher risk of renal dysfunction, when compared with ETV. TDF should be used with caution in this patient population, especially in individuals with other risk factors for renal dysfunction.
Author contributions JL, JGL, and MSK involved in study concept and design. JL and MSK drafted the manuscript. All authors interpreted the data and critically revised the manuscript for important intellectual content. All authors approved the final manuscript.

Availability of data and material
The data that support the findings of this study are available from the corresponding author upon seasonable request.