Comparative analysis of tenofovir alafenamide and entecavir on renal function and serum lipids in liver transplant recipients: a controlled retrospective cohort study

Abstract

Background

Antiviral treatments can impact serum lipids as well as kidney function. We conducted a retrospective observational study with a 48-month follow-up period, involving liver transplant recipients receiving either entecavir or tenofovir alafenamide, to assess the effects of these antiviral medications on serum lipids and renal function.

Methods

The study was conducted with 47 patients. Research data were retrospectively collected. Serum creatinine and lipid levels were measured at least once per year for each patient. The Friedman test was applied to compare estimated glomerular filtration rate (eGFR) and serum lipids.

Results

. During the 48-month observation period, a mean decrease in eGFR of 10 mL/min/1.73 m² was observed in the tenofovir alafenamide treatment group (p < 0.001), while a decrease of 8 mL/min/1.73 m² occurred in the entecavir group (p < 0.001). However, there was no statistically significant difference in the eGFR decrease between the two groups (p = 0.990). Although entecavir showed more favorable effects on total cholesterol compared to tenofovir alafenamide at the 12th and 24th months (p = 0.025 and p = 0.023, respectively), neither treatment demonstrated a lipid-lowering effect by the end of the 48-month observation period.

Conclusion

We found no significant difference in renal safety between tenofovir alafenamide and entecavir; however, the relatively small sample size and retrospective design limit the power to detect subtle differences. Entecavir was associated with a more favorable lipid profile compared to tenofovir alafenamide during the first two years of treatment, with modest reductions in total cholesterol observed at 12 and 24 months. Overall, both entecavir and tenofovir alafenamide have similar impacts on serum lipids and renal function, making them safe and effective treatment options for liver transplant recipients.

Introduction

Liver transplantation is the only definitive treatment for end-stage liver failure or cirrhosis resulting from acute and chronic liver diseases. The most common indications for liver transplantation include viral hepatitis, alcohol use, drug-induced acute liver failure, and non-alcoholic steatohepatitis [1]. Hepatitis B virus (HBV) is the most common chronic viral infection globally and represents a significant public health concern, as it is one of the leading causes of liver-related diseases and deaths. According to recent WHO data, approximately 30% of the world’s population shows serological evidence of past or present HBV infection, including resolved cases [2]. However, chronic active infection prevalence is lower [2]. An epidemiological study conducted with 5,533 healthy adults in Türkiye found a hepatitis B surface antigen (HBsAg) positivity rate of 4% and a hepatitis B core antibody (anti-HBc) positivity rate of 31%. Türkiye is classified as a moderately endemic region for HBV infection [3]. Although antiviral drugs reduce the long-term complications of chronic HBV infections, some patients still progress to end-stage liver disease. The primary indications for liver transplantation in patients with chronic HBV infection include the development of decompensated cirrhosis (presence of ascites, variceal bleeding, hepatic encephalopathy), recurrent exacerbations of hepatitis leading to liver failure, and HBV-related hepatocellular carcinoma [4]. The most commonly used antiviral agents for HBV infection include adefovir, telbivudine, lamivudine, tenofovir disoproxil fumarate (TDF), tenofovir alafenamide (TAF), and entecavir (ETV) [2].

Antiviral treatments can influence serum lipids, including low-density lipoprotein (LDL), high-density lipoprotein (HDL), total cholesterol, and triglycerides, as well as kidney function. While TDF is associated with a reduction in lipid levels, TAF, another formulation of tenofovir, has been shown to have an increasing effect on these lipids [5, 6]. The long-term use of TAF and ETV is associated with better renal safety compared to TDF [7].

We conducted a retrospective observational study with a 48-month observation period, focusing on liver transplant patients receiving antiviral treatment, to evaluate the impact of antiviral medications on lipid profiles and renal function.

Materials and methods

Study design and patients

This retrospective, observational study was conducted at the Gastroenterology Department of our Institution between January 2011 and January 2024. Liver transplant recipients who received TAF or ETV for HBV prophylaxis were identified through the hospital’s electronic medical records system. From an initial cohort of approximately 700 liver transplant recipients under follow-up, 47 patients (36 on TAF and 11 on ETV) were selected based on predefined inclusion and exclusion criteria. To enhance comparability and reduce confounding, matched reference groups were manually created by 1:1 matching with treated patients based on key baseline clinical characteristics, including age, sex, time since transplantation, and immunosuppressive therapy. Additionally all patients in reference groups did not have HBV or were on antiviral medication. Due to the relatively small sample size and the retrospective nature of the study, manual matching was chosen over propensity score matching. This approach aimed to select controls closely resembling treated patients to strengthen the validity of comparative analyses, resulting in 36 TAF-controls and 11 ETV-controls.

Inclusion criteria for the study were: age ≥ 18 years, a history of liver transplantation, and ongoing treatment with either TAF or ETV. Exclusion criteria included insufficient clinical or laboratory data (n = 8), initiation of lipid-lowering therapy during the observation period (n = 4), familial hyperlipidemia (n = 1), and uncontrolled diabetes mellitus or thyroid disorders.

Data regarding demographics, comorbidities (hypertension and diabetes mellitus), primary liver disease, type of liver transplantation (living or deceased donor), immunosuppressive regimens, use of lipid-lowering medications, and laboratory parameters were collected retrospectively. Laboratory data included serum creatinine, estimated glomerular filtration rate (eGFR), total cholesterol, LDL, HDL, and triglyceride levels. The eGFR was calculated using the 2021 CKD-EPI creatinine Eq. [8]. Laboratory data were collected at three standardized time points: baseline (prior to or at the start of antiviral therapy), 6th months, 12 months, and 24 months. Only patients with complete laboratory data available at each of these time points were included in the analysis. There were no missing values for serum creatinine, lipid profile parameters (total cholesterol, LDL, HDL, triglycerides), or other relevant clinical variables. Thus, no imputation or data correction methods were necessary. All laboratory measurements were conducted at our institution’s central biochemistry laboratory as part of routine post-transplant follow-up using standardized protocols, minimizing inter-assay variability.

Ethics

The study was approved by the Our Instution’s Non-Interventional Research Ethics Committee (Approval No: 2024/16–22) and was conducted in accordance with the ethical principles outlined in the Declaration of Helsinki and the Declaration of Istanbul. Informed consent was not required due to the retrospective nature of the study, in line with institutional policies. All patient data were fully anonymized prior to analysis, and no identifiable information was accessible to the investigators.

Statistics

Statistical analyses were performed using IBM SPSS Statistics version 24.0 (IBM Corp., Armonk, NY, USA). The normality of continuous variables was assessed using the Kolmogorov-Smirnov and Shapiro-Wilk tests. Depending on the distribution of the data, continuous variables were compared using either the independent samples t-test or the Mann-Whitney U test. For comparisons analyzed using t-test, 95% confidence intervals and Cohen’s d effect sizes were calculated and reported to provide estimates of the precision and magnitude of differences. Due to software limitations in SPSS, confidence intervals and effect sizes were not calculated for non-parametric tests (Mann-Whitney U). Categorical variables were compared using the Pearson Chi-Square test or Fisher’s Exact test where appropriate.Variables with normal distribution were presented as mean ± standard deviation (SD), while non-normally distributed data were expressed as median and interquartile range (IQR). Due to the small sample size and the non-normal distribution of repeated measures data, the Friedman test was used as a non-parametric alternative to evaluate within-group changes over time. Effect size (Cohen’s d) and post-hoc power analyses were performed to assess the magnitude and statistical power of between-group differences in renal and lipid parameters over time. A two-tailed p-value < 0.05 was considered statistically significant.

Results

A total of 94 liver transplant recipients were included in the study, with a mean age of 59.9 ± 8.1 years and a mean post-transplant follow-up of 9.3 ± 4.3 years. The majority were male (93.6%). HBV-related indications accounted for the largest proportion of primary liver diseases (48.9%). Half of the patients had received grafts from deceased donors. Hypertension and diabetes mellitus were present in 72.3% and 43.6% of the cohort, respectively. At baseline, the mean serum creatinine was 1.27 ± 0.65 mg/dL, and the mean eGFR was 70 ± 23 mL/min/1.73 m².

No recurrence of HBV was detected in patients with HBV or HBV/HDV coinfection during the 48-month observation period, and there were no deaths or instances requiring renal replacement therapy among the patients during this time.

Comparisons of baseline characteristics of groups

Baseline characteristics including age, sex, time since transplantation, primary liver disease etiology, immunosuppressive regimens, comorbidities (HT and DM), and laboratory parameters showed no statistically significant differences between the treatment and matched reference groups in both TAF and ETV cohorts (Tables 1 and 2).

Table 1.

Baseline clinical characteristics of TAF and TAF-matched reference groups

	TAF (n:36)	TAF-reference (n:36)	p
Age mean ± sd	61.2 ± 8.6	61.1 ± 8.0	0.952
Sex n (%)
Male  Female	35 (97.2) 1 (2.8)	35 (97.2) 1 (2.8)	N/A
Time elapsed since the transplantation years mean ± sd	11.3 ± 3.9	11.2 ± 3.8	0.978
Primary disease n (%)
HBV  HBV/HDV coinfection  HCV  Alcohol  Autoimmune Hepatitis  PSC  Other	30 (83.3) 6 (16.6) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0)	0 (0) 0 (0) 7 (19.4) 19 (52.8) 4 (11.1) 2 (5.6) 4 (11.1)	N/A
Transplant type n (%)
Deceased donor  Living donor	17 (47.2) 19 (52.8)	17 (47.2) 19 (52.8)	N/A
Lipid-lowering medication n (%)	11 (30.5)	11 (30.5)	N/A
Immunosuppressive agent n (%)
Tacrolimus  MMF  Everolimus  Cyclosporin  Other	11 (30.6) 28 (77.8) 24 (66.7) 6 (16.7) 3 (8.3)	14 (38.9) 31 (86.1) 19 (52.8) 7 (19.4) 5 (13.9)	0.458 0.358 0.230 0.759 0.453
HT n (%)	27 (75.0)	28 (77.8)	0.781
DM n (%)	15 (41.7)	18 (50.0)	0.478
Baseline laboratory mean ± sd
Creatinine (mg/dl)  eGFR (mL/min/1.73 m*2)  Triglyceride (mg/dl)  Total cholesterol (mg/dl)  LDL (mg/dl)  HDL (mg/dl)	1.22 ± 0.36 69 ± 18 140 ± 62 201 ± 36 129 ± 33 43 ± 9	1.11 ± 0.37 77 ± 23 120 ± 63 188 ± 38 120 ± 28 44 ± 5	0.198 0.108 0.164 0.133 0.178 0.452

Abbreviations: DM Diabetes mellitus, eGFR Estimated glomerular filtration rate, HBV Hepatitis B virus, HDL High-density lipoprotein, HDV Hepatitis D virus, HT Arterial hypertension, MMF Mycophenolate mofetil, N/A Not Applicable, LDL Low-density lipoprotein, TAF Tenofovir alafenamide

Table 2.

Baseline clinical characteristics of ETV and ETV-matched reference groups

	ETV (n:11)	ETV-reference (n:11)	p
Age mean ± sd	55.8 ± 6.8	55.9 ± 5.7	0.973
Sex n (%)
Male  Female	9 (81.8) 2 (18.2)	9 (81.8) 2 (18.2)	N/A
Time elapsed since the transplantation years median (IQR)	5 (3)	5 (2)	0.713
Primary disease n (%)
HBV  HBV/HDV coinfection  HCV  Alcohol  Autoimmune Hepatitis  PSC  Other	9 (81.8) 2 (18.2) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0)	0 (0) 0 (0) 2 (18.2) 4 (36.4) 1 (9.1) 1 (9.1) 3 (27.3)	N/A
Transplant type n (%)
Deceased donor  Living donor	6 (54.5) 5 (45.5)	7 (63.6) 4 (36.4)	0.665
Lipid-lowering medication n (%)	2 (18.2)	2 (18.2)	N/A
Immunosuppressive agent n (%)
Tacrolimus  MMF  Everolimus  Cyclosporin  Other	7 (63.6) 7 (63.6) 3 (27.3) 3 (27.3) 2 (18.2)	6 (54.5) 7 (63.6) 4 (36.4) 4 (36.4) 2 (18.2)	0.665 N/A 0.647 0.647 N/A
HT n (%)	7 (63.6)	6 (54.5)	0.665
DM n (%)	4 (36.4)	4 (36.4)	N/A
Baseline laboratory median (IQR)
Creatinine (mg/dl)  eGFR (mL/min/1.73 m*2)  Triglyceride (mg/dl)  Total cholesterol (mg/dl)  LDL (mg/dl)  HDL (mg/dl)	1.03 (1.18) 81 (69) 141 (56) 184 (25) 113 (35) 42 (10)	1.55 (0.29) 53 (19) 117 (69) 162 (18) 91 (23) 38 (10)	0.902 0.121 0.783 0.190 0.235 0.308

Abbreviations: DM Diabetes mellitus, eGFR Estimated glomerular filtration rate, ETV Entecavir, HBV Hepatitis B virus, HDL High-density lipoprotein, HD Hepatitis D virus, HT Arterial hypertension, LDL Low-density lipoprotein, MMF Mycophenolate mofetil, N/A Not Applicable, IQR Interquartile range, SD Standard deviation

There were no significant differences between the TAF and ETV groups in most baseline clinical and laboratory characteristics. However, the time elapsed since transplantation was significantly longer in the TAF group (median 11 vs. 5 years, p = 0.001), and everolimus use was more common in the TAF group (p = 0.035). Apart from these, baseline characteristics were generally similar between the groups (Table 3).

Table 3.

Baseline clinical characteristics of TAF and ETV groups

	TAF (n:36)	ETV (n:11)	p
Age mean ± sd	61.2 ± 8.6	55.8 ± 6.8	0.071
Sex n (%)
Male  Female	35 (97.2) 1 (2.8)	9 (81.8) 2 (18.2)	0.132
Time elapsed since the transplantation years median (IQR)	11 (4)	5 (3)	0.001
Primary disease n (%)
HBV  HBV/HDV coinfection	30 (83.3) 6 (16.6)	9 (81.8) 2 (18.2)	0.853
Transplant type n (%)
Deceased donor  Living donor	17 (47.2) 19 (52.8)	6 (54.5) 5 (45.5)	0.671
Lipid-lowering medication n (%)	11 (30.5)	2 (18.2)	0.702
Immunosuppressive agent n (%)
Tacrolimus  MMF  Everolimus  Cyclosporin  Other	11 (30.6) 28 (77.8) 24 (66.7) 6 (16.7) 3 (8.3)	7 (63.6) 7 (63.6) 3 (27.3) 3 (27.3) 2 (18.2)	0.076 0.435 0.035 0.419 0.131
HT n (%)	27 (75.0)	7 (63.6)	0.467
DM n (%)	15 (41.7)	4 (36.4)	0.754
Baseline laboratory median (IQR)
Creatinine (mg/dl)  eGFR (mL/min/1.73 m*2)  Triglyceride (mg/dl)  Total cholesterol (mg/dl)  LDL (mg/dl)  HDL (mg/dl)	1.21 (0.32) 70 (24) 140 (70) 200 (52) 130 (47) 43 (15)	1.03 (1.18) 81 (69) 141 (56) 184 (25) 113 (35) 42 (10)	0.479 0.577 0.911 0.081 0.123 0.833

Abbreviations: DM Diabetes mellitus, eGFR Estimated glomerular filtration rate, HBV Hepatitis B virus, HDL High-density lipoprotein, HDV Hepatitis D virus, HT Arterial hypertension, MMF Mycophenolate mofetil, N/A Not Applicable, LDL low-density lipoprotein, TAF Tenofovir alafenamide

Comparison of renal function and lipid profile changes between TAF and ETV

Cohen’s d effect size and post-hoc power analysis were conducted to evaluate the magnitude and reliability of observed group differences. For changes in eGFR (ml/min/1.73 m²), Cohen’s d values at 3rd, 6th, 12th, 24th, 36th, and 48th months were 0.457, 0.795, 0.130, − 0.199, − 0.208, and − 0.093, respectively, corresponding to post-hoc power values of 25.5%, 61.8%, 6.6%, 8.8%, 9.1%, and 5.8%. For total cholesterol changes (mg/dL), effect sizes at 6th, 12th, and 24th months were 0.259, 0.625, and 0.830, with associated power values of 11.4%, 42.7%, and 65.4%, respectively.

At the end of the 48-month follow-up period, a decrease in mean eGFR of 9.6 ± 15.9 was observed in all patients (p < 0.001). Specifically, the TAF group experienced a decrease of 9.9 ± 15.7 (p < 0.001), while the ETV group showed a decrease of 8.4 ± 17.7 (p < 0.001). Over the 48-month observation period, changes in renal function, as measured by eGFR, did not differ significantly between the TAF and ETV groups at any time point. Both groups showed modest declines in eGFR, with slightly greater reductions observed in the TAF group at later follow-ups (Table 4). At the end of the 24-month observation period, levels of LDL, HDL, triglycerides, and total cholesterol increased compared to baseline levels in TAF and ETV groups (p < 0.001). Regarding lipid parameters, the TAF group demonstrated a significant increase in total cholesterol at 12 and 24 months compared to the ETV group (p = 0.008 and p = 0.020, respectively). Additionally, HDL levels were significantly higher in the TAF group at 12 months (p = 0.006), while changes in LDL and triglyceride levels showed no statistically significant differences between the two groups (Table 4).

Table 4.

Difference in eGFR, HDL, LDL, triglyceride and total cholesterol over the observation period in between Tenofovir and Entecavir groups

	TAF (n:36)	Entecavir (n:11)	Mean difference (95% CI)	Cohen’s d	p
∆ eGFR ml/min/1.73m2 mean ± sd
3rd month  6th month  12th month  24th month  36th month  48th month	−1.7 ± 10.1 0.4 ± 8.5 −1.3 ± 11.1 −6.8 ± 12.6 −7.6 ± 14.8 −9.9 ± 15.7	−6.2 ± 8.9 −7.3 ± 13.0 −2.9 ± 15.8 −4.1 ± 16.4 −4.4 ± 17.4 −8.4 ± 17.7	4.5 (−1.7/10.7) 7.6 (0.9/14.4) 1.6 (−6.9/10.1) −2.7 (−12.1/6.7) −3.6 (−13.8/6.6) −1.5 (−12.7/9.7)	0.47 0.70 0.12 −0.18 −0.25 −0.09	0.150 0.091 0.707 0.566 0.481 0.792
∆ Triglyceride mg/dl median (IQR)
6th month  12th month  24th month	9 (52) 18 (51) 36 (58)	−14 (71) 9 (60) 25 (122)	N/A	N/A	0.543 0.449 0.106
∆ Total cholesterol mg/dl mean ± sd
6th month  12th month  24th month	5.2 ± 25.5 13.9 ± 35.0 24.8 ± 34.6	−2.2 ± 37.4 −5.8 ± 13.3 −8.6 ± 55.7	7.4 (−15.0/29.8) 19.8 (5.5/34.1) 33.4 (5.5/61.4)	0.18 0.72 0.77	0.464 0.008 0.020
∆ LDL mg/dl median (IQR)
6th month  12th month  24th month	3 (22) 8 (38) 12 (34)	−5 (33) −4 (11) −8 (48)	N/A	N/A	0.594 0.150 0.106
∆ HDL mg/dl median (IQR)
6th month  12th month  24th month	7 (10) 5 (9) 6 (10)	1 (6) 0 (9) −1 (8)	N/A	N/A	0.081 0.006 0.057

Abbreviations: CI Coefficient interval, eGFR Estimated glomerular filtration rate, ETV Entecavir, HDL High-density lipoprotein, LDL Low-density lipoprotein, N/A Not Applicable, IQR Interquartile range, TAF Tenofovir alafenamide

Comparison of renal function and lipid profile changes between TAF and TAF-control

Renal function was better preserved in TAF-treated patients compared to TAF-controls, with a statistically significant difference observed only at the 6th month in eGFR values (p = 0.004). Additionally, TAF treatment was associated with greater increases in lipid parameters, particularly total cholesterol levels at both the 12th and 24th months (p = 0.010 and p < 0.001, respectively). Triglyceride levels also showed a more pronounced rise in the TAF group by the 24th month (p < 0.001), whereas the increases in HDL and LDL levels were not statistically significant (Table 5).

Table 5.

Difference in eGFR, HDL, LDL, triglyceride and total cholesterol over the observation period in between Tenofovir and Tenofovir reference groups

	TAF (n:36)	TAF-ref. (n:36)	Mean difference (95% CI)	Cohen’s d	p
∆ eGFR ml/min/1.73m2 mean ± sd
6th month  12th month  24th month  36th month  48th month	0.4 ± 8.5 −1.3 ± 11.1 −6.8 ± 12.6 −7.6 ± 14.8 −9.9 ± 15.7	−4.8 ± 6.3 −3.6 ± 12.6 −4.7 ± 13.2 −10.1 ± 14.9 −15.9 ± 17.3	5.2 (1.7/8.7) 2.4 (−3.2/7.9) −2.1 (−8.3/3.9) 2.4 (−4.6/9.4) 6.0 (−1.7/13.8)	0.69 0.20 −0.17 0.16 0.38	0.004 0.399 0.503 0.493 0.126
∆ Triglyceride mg/dl median (IQR)
12th month  24th month	16 (51) 36 (58)	7 (40) 15 (43)	N/A	N/A	0.164 < 0.001
∆ Total cholesterol mg/dl mean ± sd
12th month  24th month	13.9 ± 35.0 24.8 ± 34.6	5.4 ± 26.3 3.9 ± 35.3	19.4 (4.8/33.9) 20.8 (4.4/37.3)	0.64 0.59	0.010 0.014
∆ LDL mg/dl mean ± sd
12th month  24th month	6.8 ± 29.4 10.0 ± 27.1	1.9 ± 25.2 2.5 ± 31.4	4.9 (−8.0/17.7) 7.5 (−6.3/21.3)	0.18 0.25	0.455 0.283
∆ HDL mg/dl mean ± sd
12th month  24th month	4.2 ± 7.8 5.7 ± 7.7	0.8 ± 9.3 2.7 ± 5.6	3.5 (−0.5/7.5) 3.0 (−0.1/6.2)	0.43 0.46	0.089 0.061

Abbreviations:CI Coefficient interval, eGFR Estimated glomerular filtration rate, ETV Entecavir, HDL High-density lipoprotein, LDL Low-density lipoprotein, N/A Aot Applicable, TAF Tenofovir alafenamide

Comparison of renal function and lipid profile changes between ETV and ETV-control

ETV-treated patients showed a slower decline in renal function compared to ETV-controls; however, the difference was not statistically significant. Lipid parameters showed a more favorable profile in the ETV group. Total cholesterol and LDL levels decreased significantly in the ETV group, whereas they increased in controls at both the 12th and 24th months (p-values for total cholesterol: <0.001 and 0.011; LDL: 0.001 and 0.008). Moreover, HDL levels remained stable or declined slightly in the ETV group, while significantly increasing in the controls at 12 months (p = 0.001) (Table 6).

Table 6.

Difference in eGFR, HDL, LDL, triglyceride and total cholesterol over the observation period in between Tenofovir and Tenofovir reference groups

	ETV (n:11)	ETV-ref. (n:11)	Mean difference (95% CI)	Cohen’s d	p
∆ eGFR ml/min/1.73m2 mean ± sd
6th month  12th month  24th month  36th month  48th month	−7.3 ± 13.0 −2.9 ± 15.8 −4.1 ± 16.4 −4.4 ± 17.4 −8.4 ± 17.7	−5.5 ± 6.2 −7.5 ± 11.4 −11.3 ± 10.6 −12.3 ± 12.0 −14.8 ± 11.7	−1.8 (−10.9/7.3) 4.6 (−7.6/16.9) 7.2 (−5.1/19.4) 8.2 (−3.5/19.9) 6.4 (−6.9/19.7)	0.18 0.35 0.51 0.65 0.43	0.683 0.439 0.237 0.159 0.332
∆ Triglyceride mg/dl median (IQR)
12th month  24th month	9 (60) 25 (122)	28 (65) 35 (24)	N/A	N/A	0.088 0.047
∆ Total cholesterol mg/dl mean ± sd
12th month  24th month	−5.8 ± 13.3 −8.6 ± 55.7	4.1 ± 30.7 5.1 ± 42.3	−4.2 (−6.8/−2.6) −5.9 (−10.3/−15.5)	1.93 1.25	< 0.001 0.011
∆ LDL mg/dl median (IQR)
12th month  24th month	−4 (11) −8 (48)	20 (23) 36 (31)	N/A	N/A	0.001 0.008
∆ HDL mg/dl median (IQR)
12th month  24th month	0 (9) −1 (8)	6 (7) 7 (8)	N/A	N/A	0.001 0.088

Abbreviations:CI Coefficient interval, eGFR Estimated glomerular filtration rate, ETV Entecavir, HDL High-density lipoprotein, LDL Low-density lipoprotein, N/A Not Applicable, TAF Tenofovir alafenamide

Discussion

In this observational study involving 94 liver transplant recipients, including 36 patients on TAF, 11 on ETV, and their respective matched reference groups (36 TAF-control and 11 ETV-control), we observed several important clinical outcomes over a 48-month period. TAF and ETV groups showed a gradual decline in eGFR over the 48-month follow-up period; however, the magnitude of decline was similar between the groups, with no statistically significant differences at any time point. While changes in triglyceride levels were not significantly different between the groups, patients receiving TAF experienced significantly greater increases in total cholesterol at 12 and 24 months. Similarly, HDL levels were significantly higher in the TAF group at 12 months. Although LDL levels tended to increase more in the TAF group, this difference did not reach statistical significance. Overall, TAF was associated with more pronounced increases in lipid parameters, whereas renal function decline remained comparable between the two groups.

The mean age of the study group was 60, and 94% of the participants were male. Liver transplant patients in Turkey are predominantly male [9], and our findings indicated an even higher proportion, as we included a subgroup of liver transplant patients receiving antiviral therapy. These demographic patterns are consistent with the HBV-endemic nature of our region and local liver transplant epidemiology. Nonetheless, they may limit the extrapolation of our findings to populations with differing etiologic or demographic characteristics.

In the study conducted by Sara Jeong et al. on patients with HBV infection, there was no significant difference in the renal safety profiles among TDF, TAF, and ETV treatments, similar to our findings [10]. In our study, 24 patients (66.7%) in the TAF group were receiving everolimus and 11 patients (30.6%) were on tacrolimus, whereas in the ETV group, everolimus was used by 3 patients (27.3%) and tacrolimus by 7 patients (63.6%). It is important to note that the use of tacrolimus in these patients could further complicate the interpretation of the results. Specifically, patients in the ETV group, who exhibited better baseline creatinine clearance, may have experienced a negative impact on renal function due to tacrolimus, a drug known to reduce creatinine clearance [11]. This could have led to a further decline in renal function despite initially favorable baseline levels. Conversely, in patients receiving TAF, the transition from other antiviral medications to TAF was predominantly motivated by concerns regarding renal impairment [7]. As these patients already had lower baseline creatinine clearance, they were more likely to be prescribed everolimus, which has a more favorable renal profile and does not significantly impact creatinine clearance [12]. While the use of antiviral agents such as ETV and TAF may directly influence renal function, concurrent medications, particularly tacrolimus, also play a significant role in determining renal outcomes. Our findings indicate that renal function, as measured by eGFR, declined modestly over time in both treatment groups, with no statistically significant difference between TAF and ETV at most time points. This trend is consistent with previous studies suggesting that renal function may gradually deteriorate after liver transplantation due to multifactorial causes, including immunosuppressive therapy, comorbidities, and the cumulative effect of antiviral agents. Although TAF has been associated with a more favorable renal safety profile compared to TDF, our results suggest that its advantage over ETV in this context may be limited or clinically marginal in stable liver transplant recipients.

Cardiovascular mortality is the most common cause of death among liver transplant recipients within the first year post-transplantation and is the second most common cause of death overall, following malignancy [13]. Metabolic syndrome is one of the most significant contributing factors to cardiovascular mortality, with a prevalence of 50% among liver transplant recipients [14]. Dyslipidemia also significantly increases cardiovascular mortality, with a prevalence of 68% among liver transplant recipients [15]. Recipient age, calcineurin inhibitor use, and the duration of prednisolone treatment have been identified as important risk factors for post-transplant dyslipidemia [16]. Therefore, it is crucial to implement appropriate measures and optimize the medication regimens of liver transplant recipients in order to reduce cardiovascular mortality. Some studies found that TDF has a lipid-lowering effect on serum levels, whereas TAF and ETV are associated with an increase in serum lipid levels [17, 18]. In our study, we found that ETV decreased total cholesterol levels at the 12th and 24th months, while TAF increased total cholesterol. Entecavir was associated with a more favorable lipid profile compared to tenofovir alafenamide during the first two years of treatment, with modest reductions in total cholesterol observed at 12 and 24 months. These findings are consistent with those reported by Rui-Min Lai et al. [19]. They reported similar results in their study on patients with chronic HBV infection, which evaluated the effects of TAF and ETV on serum lipid levels over a one-year treatment period [19]. In the meta-analysis by Kexin Tong et al., it was concluded that both TAF and ETV increase serum lipid levels within an acceptable range, with no significant difference in their dyslipidemic effects [20]. We observed similar findings in our study; by the end of the 24th month, there was no significant difference in the changes in serum lipid levels between the TAF and ETV groups, except for total cholesterol. Another meta-analysis evaluating the effect of TAF on serum lipids reported that TAF increases serum lipid levels, particularly total cholesterol, over time, a finding consistent with our results [21]. Given that ETV showed a more favorable lipid profile on total cholesterol and LDL levels, it may be preferable in terms of cardiovascular risk protection. However, the clinical relevance of these lipid changes warrants cautious interpretation. In the context of liver transplant recipients, who already have elevated cardiovascular risk, the observed increases may have implications for long-term risk stratification and the need for lipid-lowering interventions. Further studies integrating cardiovascular outcome measures and risk scoring systems are needed to clarify the significance of these findings.

This study has certain limitations that should be acknowledged. A key limitation of our study is the relatively small sample size, particularly in the ETV group (n = 11), which significantly reduces the statistical power to detect clinically meaningful differences. Post-hoc power analysis revealed that, with the exception of total cholesterol change at 24 months (Cohen’s d = 0.83; power = 65.4%), most comparisons across eGFR and lipid parameters exhibited insufficient power (< 80%). Specifically, power values for changes in eGFR over time ranged from 5.8 to 61.8%, and for total cholesterol changes from 11.4 to 65.4%, indicating a considerable risk of Type II error. These findings suggest that non-significant results should be interpreted with caution, as the study may be underpowered to detect smaller but clinically relevant differences, especially in safety-related outcomes. Future studies with larger, more balanced cohorts are warranted to validate our observations and to ensure sufficient power for subgroup analyses.

Although we used the 2021 CKD-EPI creatinine equation to estimate eGFR, which is widely accepted for clinical use, its accuracy in liver transplant recipients may be limited due to altered muscle mass, hepatic function, and non-steady-state conditions in this population. Validation studies of eGFR equations in transplant cohorts are limited, and this should be considered when interpreting renal function estimates in our study. The imbalance in group sizes and the longer post-transplant period observed in the TAF group may have influenced clinical outcomes, especially considering that everolimus use was more frequent in this group. These factors could have affected both renal function and lipid metabolism, potentially impacting the comparative analysis. An additional limitation is the presence of potential temporal and selection biases due to differences in the timing of transplantation between groups. Specifically, the median time since transplantation differed significantly between the TAF and ETV groups, which may have introduced confounding by both patient characteristics and evolving clinical practices. Another limitation of this study is the manual matching of reference groups instead of using more rigorous methods like propensity score matching, which may leave residual confounding and affect result validity. Although we attempted to match patients based on baseline characteristics, we acknowledge that unmeasured confounding variables such as adherence to treatment, baseline cardiovascular risk, prior lipid-lowering therapy, and dietary/lifestyle factors could have influenced the results. These were not consistently available in retrospective records and could have introduced residual bias in comparative analyses.

Despite these limitations, baseline clinical characteristics were largely comparable across groups, and matched reference groups were carefully matched to enhance the reliability of comparisons. Nevertheless, the retrospective nature of this study inherently introduces a risk of selection bias and confounding, particularly given the lack of randomization in antiviral therapy assignment. The choice between TAF and ETV may have been influenced by clinicians’ perceptions of renal or bone-related safety concerns, leading to indication bias. Patients switched to TAF may have had pre-existing renal impairment, which could have confounded eGFR trajectories irrespective of the drug’s effect. Although the data were collected through the hospital’s electronic medical record system, reducing the likelihood of information bias, the lack of randomization remains a methodological constraint. Notably, differences in post-transplant duration and concomitant immunosuppressive use between groups may reflect indication bias, as patients switched to TAF often had prior renal or bone concerns motivating therapy changes. Despite these limitations, this study offers valuable insight with one of the longest follow-up periods evaluating renal function and lipid profile changes under antiviral treatment in liver transplant recipients. Given the retrospective design and small sample size, particularly in the entecavir group, we opted for non-parametric analyses such as the Friedman test. Although mixed-effects modeling may provide more robust estimates, we refrained from such analyses to avoid overfitting and misleading results due to the low number of subjects. Furthermore, we did not adjust for multiple comparisons due to the exploratory nature of the study, which may increase the risk of Type I error. These limitations should be considered when interpreting the results. It should be noted that this was a single-center study conducted at a tertiary care center in Turkiye. This limits the generalizability of the results to broader populations, especially considering geographic and demographic differences in liver transplant practices. Lastly, another limitation of this study is the incomplete baseline characterization, as key variables such as baseline HBV DNA levels, HBsAg status, prior antiviral treatment history, and detailed immunosuppressive dosing were not consistently available, which may affect the interpretation of treatment outcomes. Current literature mainly addresses HBV or HIV-infected individuals, with limited data available for this specific population. The results contribute to this gap and offer meaningful clinical insight, supporting the need for further prospective, multicenter studies with larger sample sizes to confirm these observations.

Conclusion

In this study, TAF and ETV demonstrated comparable renal safety profiles over a 48-month follow-up period in liver transplant recipients. However, their effects on lipid metabolism differed markedly. TAF was associated with significant increases in total cholesterol, LDL, and triglyceride levels, reflecting a less favorable lipid profile. In contrast, ETV not only showed a neutral effect on lipids but was also associated with reductions in total cholesterol and LDL levels, suggesting a potential lipid-lowering benefit in this patient population.

Authors’ contributions

HD: Conceptualization, Data Curation, Formal analysis, Interpretation of Data, Writing - original draft. ND: Conceptualization, Data Curation, Interpretation of Data, Supervision, Writing – review & editing. TU: Conceptualization, Interpretation of data, Writing – review & editing. TE: Conceptualization, Interpretation of data, Writing – review & editing. CA: Conceptualization, Interpretation of data, Writing – review & editing. MO: Conceptualization, Interpretation of data, Writing – review & editing. HE: Conceptualization, Formal Analysis, Interpretation of Data. MA: Conceptualization, Supervision, Writing – review & editing. All authors critically revised the manuscript, approved the final version to be published, and agreed to be accountable for all aspects of the work.

Funding statement

None declared.

Data availability

All data used in this study are available from the corresponding author upon request: Hüseyin Döngelli; drhuseyindongelli@gmail.com. This statement is consistent with the manuscript file.

Declarations

Ethics approval and consent to participate

This study was performed in line with the principles of the Declaration of Helsinki. The study was approved by the Dokuz Eylul University Non-Interventional Research Ethics Committee on May 8, 2024, under approval number 2024/16–22. Since the study design was retrospective, the ethics committee did not find it necessary to obtain informed consent from the patients, and the patients’ identities were processed after being anonymized.

Consent of publication

Not applicable.

Competing interests

The authors declare no competing interests.