Results in chronic hepatitis B patients using tenofovir and entecavir for at least 10 years; HBV clearance rare, disease outcomes good: An observational cohort study

Abstract

This study aims to investigate antiviral effectiveness, side effects, and disease outcomes in patients who have been using entecavir (ETV) and tenofovir disoproxil fumarate (TDF) for a long-term in chronic hepatitis B. Patients with chronic hepatitis B who had been using TDF or ETV for at least 10 years were included in this retrospective study. Co-infected patients, those receiving immunosuppressive therapy, and transplant patients were excluded. Of the study’s total 173 patients (baseline mean age 43.4 ± 11.7 years), 110 (63.6%) were men. Thirty-three (19.1%) patients were cirrhotic, and hepatitis B e-antigen was negative in 131 (75.7%) patients at the baseline. Ninety-two (53.2%) patients used TDF and 81 (46.8%) used ETV for a mean of 156.76 ± 21.60 (120–204) months. Hepatitis B virus (HBV)-DNA negativity (<10 IU/mL) was achieved in 97.7% of all patients. Biochemical remission was achieved in 98.3% of all patients at the last visit. HBsAg became negative in only 5 (2.9%). Hepatocellular carcinoma (HCC) developed in 9 (5.2%) patients. All HCCs occurred after the 5th year of treatment. The age at HBV diagnosis was higher in HCC patients (P = .023), but the most important risk factor for the development of HCC was to have cirrhosis at baseline. Eight (4.6%) patients died in the follow-up, and 2 were due to liver disease and the remaining non-liver disease. At the end of follow-up, HBV-DNA negativity was achieved in almost all patients, and HBsAg sero-clearance was rarely achieved. Very few patients developed HCC and the long-term mortality rate was similar to the general population.

1. Introduction

Hepatitis B virus (HBV) infection is still a global public health problem, despite highly effective vaccination efforts and the availability of highly effective and potent drugs. It is thought that approximately 250 million people in the world still have chronic hepatitis B (CHB) infection.^[1] It is the 10th cause of death in the world due to its very serious complications, especially cirrhosis and hepatocellular carcinoma (HCC),^[2] and the death rate due to these causes is gradually increasing.^[3]

The goals of treatment in CHB are to prevent the development of cirrhosis, decompensation, liver failure, HCC development, the need for liver transplantation, and mortality due to all these. In addition, with treatment, HBV transmission from mother to baby, HBV-related extra-hepatic complications, and HBV reactivation can be prevented.

Currently, the treatment agents for CHB are tenofovir and entecavir (ETV), which are potent and have almost no risk of resistance. This study aims to investigate antiviral effectiveness, side effects, and disease outcomes in patients who have been using oral antivirals for a long-term (at least 10 years) in CHB. This is the study with the longest treatment and follow-up period for CHB patients treated with ETV and tenofovir disoproxil fumarate (TDF) in the literature.

2. Materials and methods

We retrospectively analyzed CHB patients who had been using ETV 0.5 mg once daily or TDF 245 mg once daily for at least 10 years at the gastroenterohepatology clinic of the Istanbul Faculty of Medicine, from January 2006 to December 2023. The University’s ethical committee approved the study protocol.

Inclusion criteria were as follows: (1) adult (≥18 years old) CBH patients, (2) continued with the same antiviral treatment for at least 120 months, (3) using ETV or TDF for at least 120 months, regardless of treatment experience or not. Exclusion criteria were as follows: (1) co-infected patients with HCV, HDV, and HIV, (2) receiving immunosuppressive therapy, and (3) hematologic or solid organ transplant patients, (4) having other accompanying liver disease.

Demographic characteristics of the patients (age and gender), laboratory characteristics at the beginning of treatment and the last follow-up visit (hemogram, kidney function tests, liver enzymes, International Normalized Ratio, HBV serology, HBV-DNA level), radiological imaging (liver ultrasound, CT or MRI) features at the beginning of treatment and the last follow-up visit, fibrosis stage of liver disease by noninvasive scores (aspartate aminotransferase to platelet ratio index [APRI] and fibrosis 4 [FIB4]) at the beginning of treatment and the end of follow-up, disease stage in the liver biopsy at the beginning of treatment, treatment-related complications (such as nephropathy, and hypophosphatemia), course of liver disease (transition to cirrhotic or non-cirrhotic stage, decompensation, HCC development), whether the patients had treatment experience before using ETV or TDF, and mortality of the patients were examined. Additionally, the PAGE-B scores predicting HCC risk in patients under antiviral treatment were also calculated. The PAGE-B is a risk-scoring system that includes platelet count, age, and sex. HBV-DNA hybridization test was performed using the bDNA (Versant 3.0, BAYER) method at the beginning of the treatment. The measurement range of the test is between 2.000 and 100,000,000 copies/mL. Quantitative HBV DNA testing was performed on the COBAS 6800 system with the COBAS HBV test for the last 5 years. The measuring range of the test is between 10 IU/mL and 1,000,000,000 IU/mL. HBV-DNA < 10 IU/mL was considered as virological response.

Diagnosis of cirrhosis was determined based on radiological imaging (ultrasonography, computed tomography, and magnetic resonance imaging) or liver biopsy features. Patients with coarse liver parenchyma, irregular surface, and portal hypertension findings in the evaluation made by radiological imaging and/or fibrosis stage ≥ 5/6 according to ISHAK on liver biopsy were considered to have liver cirrhosis.

The requirement for written informed consent was waived due to the study’s retrospective nature and received approval from the ethics committee of the Istanbul Faculty of Medicine. The data of this study is available from the corresponding author.

2.1. Ethics statement

Approval for this study was obtained from the Ethics Committee of Istanbul University Istanbul Faculty of Medicine on December 15, 2023. The protocol number is 2023/2231. All the applied procedures complied with the ethical standards of the Human Testing Committee of our institution and the Helsinki Declaration.

2.2. Statistical analysis

Descriptive statistics for continuous variables in the study; mean, median, and standard deviation; for categorical variables, they were expressed as numbers (n) and percentages (%). Independent T test was performed in the comparison of measurements according to categorical groups. The chi-square test was calculated to determine the relationships between categorical variables. Paired-sample T test was used to analyze the dependent quantitative data. Survival (HCC) curves were estimated using the Kaplan–Meier method, and compared using the log-rank test. Multivariate analysis was performed based on the Cox proportional hazards regression model. Statistical significance level (a) was taken as 5% in the calculations and SPSS (IBM SPSS for Windows, ver.22) statistical package program was used for analysis.

3. Results

3.1. Baseline characteristics of the patients

There were a total of 173 patients in our study. Of these, 110 (63.6%) were male, and 63 (37.4%) were female. The mean age of the patients at the beginning of treatment was 43.4 ± 11.7 years (median, IQR; 44 [35–53]). 92 (53.2%) patients used TDF and 81 (46.8%) patients used ETV. Patients using ETV were significantly older (P = .045). The mean treatment duration was 156.7 ± 21.6 (120–204) months. There was no significant difference between TDF and ETV usage times (154.3 ± 22.7 and 159.6 ± 20 months respectively, P = .109).

In the study, 94 (54.3%) patients were treatment-naive, and 79 (45.7%) patients were treatment-experienced. The mean duration of use of previous treatments in treatment-experienced patients was 41.9 ± 41.4 (3–192) months. The 2 most frequently used agents in treatment-experienced patients are lamivudine (17.3%) and pegylated interferon alpha (15%), respectively. The treatment experience rate was significantly higher in patients receiving TDF than in patients receiving ETV (P = .02).

According to radiological imaging and/or liver biopsy features, 33 (19%) CHB patients were compensated cirrhotic at baseline. There weren’t any decompensated cirrhotic patients at the baseline. The rate of cirrhotic patients before treatment was significantly higher in those receiving ETV in our study group (P = .003). There was no significant difference in pretreatment hepatitis B e-antigen (HBeAg), HBV-DNA level, and gender between TDF and ETV. Detailed baseline characteristics of the patients are given in Table 1.

Table 1.

Baseline characteristics of the patients.

Baseline characteristics	All patients (n = 173)	ETV (n = 81)	TDF (n = 92)	P-value
Age (years) (mean ± SD)	43.44 ± 11.74	45.3 ± 10.4	41.7 ± 12.5	.045
Gender (n, %)		81 (46.8)	92 (53.2)	.635
Male	110 (63.6)	53 (65.4)	57 (62)
Female	63 (37.4)	28 (34.5)	35 (38)
HbeAg positiviy (n, %)	42 (24.3)	14 (17.3)	28 (30.4)	.051
HBV-DNA (log10IU/mL) (mean ± SD)	7.6 ± 7.8	7.5 ± 7.8	7.7 ± 7.8	.123
AST (U/L) (mean ± SD)	59 ± 60	72 ± 76	49 ± 43	.029
ALT (U/L) (mean ± SD)	88 ± 103	101 ± 120	77 ± 85	.158
ALP (U/L) (mean ± SD)	163 ± 83	167 ± 86	160 ± 81	.630
GGT (U/L) (mean ± SD)	43 ± 59	42 ± 76	44 ± 74	.875
Total bilirubin (mg/dL) (mean ± SD)	0.9 ± 1.7	1.3 ± 2.6	0.7 ± 0.4	.068
Albumin (g/dL) (mean ± SD)	4.2 ± 0.4	4.2 ± 0.3	4.3 ± 0.4	.211
Platelet (10³/µL) (mean ± SD)	208.4 ± 65.2	196.5 ± 63.2	217.7 ± 65.5	.049
INR (mean ± SD)	1 ± 0.2	1.1 ± 0.2	1 ± 0.2	.685
AFP (ng/mL) (mean ± SD)	5.4 ± 11.6	5.4 ± 8.9	5.4 ± 13.2	.993
Creatinine (mg/dL) (mean ± SD)	0.8 ± 0.2	0.8 ± 0.2	0.8 ± 0.2	.315
eGFR (mL/min/1.73 m2) (mean ± SD)	103.2 ± 18.4	103 ± 17.3	103.4 ± 19.3	.912
P (mg/dL) (mean ± SD)	3.2 ± 0.5	3.2 ± 0.4	3.3 ± 0.7	.794
Cirrhosis (n, %)	33 (19)	23 (28.4)	10 (10.9)	.003
Liver biopsy (n = 102)	16 (15.6)
Radiological imaging (n = 141)	17 (9.8)
Treatment-experienced (n, %)	79 (45.7)	27 (33.3)	52 (56.5)	.002
FIB4 Score (mean ± SD) (< 1.45, ≥ 1.45 - < 3.25, ≥ 3.25)	1.6 ± 1.4	1.8 ± 1.4	1.4 ± 1.4	.073
APRI (mean ± SD) (< 0.5, ≥ 0.5 - < 1, ≥ 1)	1.7 ± 8.7	0.8 ± 0.9	2.4 ± 11.6	.508
Treatment duration (months) (mean ± SD)	156.76 ± 21.60	159.6 ± 20	154.3 ± 22.7	.109

AFP = alpha-fetoprotein, ALP = alkaline phosphatase, ALT = alanine aminotransferase, AST = aspartate aminotransferase, APRI = AST to platelet ratio index, eGFR = estimated glomerular filtration rate, FIB4 = fibrosis 4 score, GGT = gamma-glutamyl transferase, INR = international normalized ratio, P = phosphorus, SD = standard deviation.

3.2. Biochemical response of CHB patients in TDF group and ETV group

There was a significant decrease in aspartate aminotransferase, alanine aminotransferase (ALT), alkaline phosphatase, and gamma-glutamyl transferase levels in both ETV and TDF groups at the last follow-up visit (P < .05). APRI and FIB4 scores were found to decrease significantly in the group receiving ETV (P < .0001 and P = .049, respectively). In the group receiving TDF, there was no significant change in the FIB4 score (P = .416), while a significant decrease was seen in the APRI score (P < .001).

In the total group, the ALT level of 91 patients was > 42 U/L (the upper limit of normal value of our laboratory) at the baseline. There was no significant difference in basal ALT levels between the ETV and TDF groups (101 ± 120 U/L and 77 ± 85 U/L, respectively, P < .112). ALT levels were normalized (<42 U/L; by ALT normalization, we mean reducing ALT value below the upper limit of the laboratory) in 96.7% of patients with elevated baseline ALT. According to our laboratory cutoff (<42 U/L), 170 (98.3%) patients had normal ALT levels at the last visit. There was no significant difference in ALT levels at the end of follow-up between the ETV and TDF groups (18.7 ± 9.6 U/L and 21.6 ± 11.7 U/L, respectively, P < .144). According to the American Association for the Study of Liver Diseases (AASLD) ALT cutoff (<25 U/L for women, <35 U/L for men), 114 (65.9%) patients had high levels of ALT. At the last visit, only 13 (7.5%) patients had high levels of ALT. The normalization range of liver enzymes was 88.6%, and 160 (92.5%) patients had normal levels of ALT at the last visit according to the AASLD cutoff. Laboratory changes in all patients during the treatment are given in Table 2.

Table 2.

Laboratory changes in all patients during the treatment.

n = 173	Baseline labs	Last follow-up visit	P-value
Creatinine (mg/dL) (mean ± SD)	0.8 ± 0.2	0.8 ± 0.3	.862
eGFR (mL/min/1.73 m2) (mean ± SD)	103 ± 18	92 ± 21	<.001
AST (U/L) (mean ± SD)	59 ± 61	21 ± 17	<.001
ALT (U/L) (mean ± SD)	88 ± 104	19 ± 8	<.001
ALP (U/L) (mean ± SD)	162 ± 85	81 ± 36	<.001
GGT (U/L) (mean ± SD)	44 ± 60	23 ± 29	<.001
T. bil (mg/dL) (mean ± SD)	1 ± 1.8	0.6 ± 0.4	.020
Albumin (g/dL) (mean ± SD)	4.2 ± 0.4	4.4 ± 0.3	<.001
P (mg/dL) (mean ± SD)	3.2 ± 0.5	3.1 ± 0.7	.562
Platelet (10³/µL) (mean ± SD)	208 ± 66	232 ± 70	<.001
AFP (ng/mL) (mean ± SD)	5.5 ± 11	24 ± 228	.385
INR (mean ± SD)	1.1 ± 0.2	1 ± 0.1	.035
FIB4 score (mean ± SD) (< 1.45, ≥ 1.45 - < 3.25, ≥ 3.25)	1.6 ± 1.4	1.5 ± 1.3	.086
APRI (mean ± SD) (< 0.5, ≥ 0.5 - < 1, ≥ 1)	1.3 ± 6.7	0.2 ± 0.2	.065

AFP = alpha-fetoprotein, ALP = alkaline phosphatase, ALT = alanine aminotransferase, AST = aspartate aminotransferase, APRI = AST to platelet ratio index, eGFR = estimated glomerular filtration rate, FIB4 = fibrosis 4 score, GGT = gamma-glutamyl transferase, INR = international normalized ratio, P = phosphorus, SD = standard deviation.

3.3. Viral response, HBeAg loss, HBsAg loss in CHB patients in the TDF group and ETV group

When the entire group was examined, HBV-DNA negativity (<10 IU/mL) was achieved in 98.2% of the patients, at the end of follow-up. HBV-DNA was ≥ 10 IU/mL in 1.8% of the patients. The course of ALT and HBV-DNA negativity during the treatment period is shown in Figure 1.

Figure 1.

The course of ALT levels and HVB-DNA negativity achievement during the follow-up. ALT = alanine aminotransferase, HBV = hepatitis B virus.

HBV-DNA negativity was achieved in 97.8% (n = 90) of those who received TDF and in 97.5% (n = 79) of those who received ETV, and there was no statistical difference between them (P = .867). When we investigated the patients who remained HBV-DNA positive in detail, we found out that these patients group was non-compliant with treatment. In patients whose HBV-DNA did not become negative, HBV-DNA levels were 20 IU/mL (ETV), 88 IU/mL (TDF), 252 IU/mL (ETV), and 2633 IU/mL (TDF). Genetic resistance tests were performed in 3 patients (by INNO-LiPA HBV DR), but no resistance was detected.

When the entire group was examined, no difference was detected between those with treatment experience and treatment-naive patients in terms of HBV-DNA negativity at the end of the study (P = .103). HBV-DNA negativity was achieved in 100% of treatment-experienced patients and 96.7% of treatment-naive patients.

Before treatment, 42 (24.3%) of the patients had positive HbeAg in the total group. HBeAg loss was observed in 40.5% (n = 17) of these patients (Fig. 2). There was no significant difference between the use of ETV or TDF in terms of HBeAg negativity (P > .05).

Figure 2.

Cumulative HBeAg negativity rates in HBeAg positive patients. HBeAg = hepatitis B e-antigen.

HBsAg loss was seen in only 5 (2.9%) patients, 4 of them were under TDF and 1 was under ETV. There was no significant difference between them. Two of these patients developed anti-HBs spontaneously. HBsAg loss was observed at the mean of 11.2 ± 2.7 (min 10–max 16) years. Of these, 4 were male and 1 was female. The mean age was 54.6 ± 11.9 years. HBeAg was negative at the baseline and the last visit. Patients were non-cirrhotic at the baseline and the last visit. Only 1 male patient had baseline ALT above the laboratory upper limit of 42 U/L. The other patients had normal baseline ALT levels. All patients had normal ALT levels at the last visit.

3.4. Disease course

At the beginning of treatment, 33 (19.1%) patients were cirrhotic. According to liver biopsy features (ISHAK stage ≥ 5/6), 16 patients, and according to radiological imaging (coarse liver parenchyma, irregular surface, and portal hypertension findings) and clinical features, 17 patients were in the cirrhotic stage. During the follow-up, 13 (39.4%) of 33 patients who were cirrhotic at baseline regressed to the non-cirrhotic stage with the treatment according to radiological imaging and FIB4 (<3.25) or APRI (<1) scores. The rate of cirrhotic patients was significantly higher in the group receiving ETV than receiving TDF at the last visit in our study group (n = 18 [22.2%] and n = 10 [10.8%], respectively, P = .043). However, it should not be forgotten that the rate of cirrhosis was higher in the group receiving ETV at the beginning of treatment. During the follow-up, 8 (5.7%) of 140 non-cirrhotic patients at baseline progressed to the cirrhotic stage. The change in the cirrhotic stages of the patients is shown in Table 3.

Table 3.

Rates of progression and regression of cirrhosis during treatment.

Radiological imaging and/or biopsy characteristics	n = 173	Cirrhotic at last visit	Non-cirrhotic at last visit
Cirrhotic at baseline (n, %)	33 (19.1) 23 patients with ETV 10 patients with TDF	20 (60.6)	13 (39.4)
Non-cirrhotic at baseline (n, %)	140 (80.9)	8 (5.7)	132 (94.3)
Total (n, %)	173 (100)	28 (16.2) 18 patients with ETV 10 patients with TDF	145 (83.8)

ETV = entecavir, TDF = tenofovir disoproksil fumarate.

At the end of the follow-up, 7 (4%) out of all patients were decompensated. Only 2 of these patients were cirrhotic at baseline, and they were compensated cirrhosis. Decompensation characteristics were as follows: variceal bleeding and ascites in 2 patients, ascites and hepatic encephalopathy in 2 patients, ascites only in 2 patients, and hepatic encephalopathy only in 1 patient. There were no statistical differences in decompensation between ETV and TDF groups (P = .806).

HCC developed in 9 (5.2%) patients. Of the patients who developed HCC, 1 used TDF, and 8 used ETV. According to Kaplan–Meier survival analysis, treatment choice was not found to be statistically significant in the development of HCC (P = .056). All HCCs occurred after the 5th year of treatment. PAGE-B score was significantly higher in those who developed HCC (12.9 ± 4.8 and 8 ± 5.2, respectively, P = .009). There was no difference between genders in the development of HCC (P = .378) in survival analysis. There was no difference between treatment-naive and treatment-experienced patients in the development of HCC (P = .055) in survival analysis. The AUC of the age for HCC detection was 0.778 (95% confidence interval [CI]:.707–.850, P = .005]. At an optimal cutoff of 50, the age had a sensitivity of 89% and a specificity of 76.2% for HCC detection. Being over the age of 50 years at the beginning of the treatment was significantly associated with the development of HCC (P < .001) in survival analysis. The age at CHB diagnosis was significantly higher in HCC patients (48 ± 6.4 years and 37.7 ± 13.2 years respectively, P = .023). Being cirrhotic stage at baseline had a statistically significant effect on HCC development (P = .001) (Fig. 3). In multivariate Cox regression analysis, age (hazard ratio = 12.513; 95% CI 1.542–101.552; P = .018) and cirrhotic stage (hazard ratio = 5.571; 95% CI 1.382–22.453; P = .016) were associated with the development of HCC (Table 4).

Figure 3.

Kaplan–Meier curves for cumulative HCC of cirrhotic or non-cirrhotic patients at the beginning of the treatment. HCC = hepatocellular carcinoma.

Table 4.

Factors associated with hepatocellular carcinoma on multivariate Cox regression analysis.

	Hepatocellular carcinoma hazard ratio (95% CI)	P-value
Age	12.513 (1.542–101.552)	.018
Cirrhosis	5.571 (1.382–22.453)	.016

CI = confidence interval, ETV = entecavir, TDF = tenofovir disoproksil fumarate.

The model included baseline age and cirrhotic condition which are significantly different in survival analysis.

During the follow-up period, 8 (4.6%) patients died, and 2 were due to liver disease and the remaining non-liver disease. There was no significant difference in mortality between TDF and ETV (P > .05).

3.5. Drug side effects

All side effects evaluated developed after at least 10 years of treatment. During follow-up nephropathy (GFR < 60 mL/min) developed in 9 (5.2%) patients, hypophosphatemia (P < 2.5 mg/dL), and nephropathy developed in 4 (2.3%) patients, and hypophosphatemia developed in 19 (11%) patients. The rate of both side effects was significantly higher in those receiving TDF than in those receiving ETV (P = .03). Rates of hypophosphatemia and nephropathy in ETV and TDF groups are given in Table 5. During the follow-up, 18 patients had a change in medication due to hypophosphatemia and/or nephropathy; 17 patients (2 with ETV, 15 with TDF) were switched to tenofovir alafenamide, and 1 patient was switched from TDF to ETV. Other patients underwent follow-up or dose adjustment.

Table 5.

Rates of hypophosphatemia and nephropathy in ETV and TDF groups.

Adverse effects	Hypophosphatemia (P < 2.5 mg/dL)	Nephropathy (eGFR < 60 mL/min/1.73 m2)	Hypophosphatemia (P < 2.5 mg/dL) + nephropathy (eGFR < 60 mL/min/1.73 m2)
ETV (n)	5	4	-
TDF (n)	14	5	4
Total (n, %)	19 (11)	9 (5.2)	4 (2.3)

eGFR = estimated glomerular filtration rate, ETV = entecavir, TDF = tenofovir disoproksil fumarate.

4. Discussion

CHB is a major public health problem worldwide. CHB is a significant cause of morbidity and mortality with the risk of liver cirrhosis, decompensation development, liver failure, and HCC, and also poses a significant burden on the health economy with all these complications.^[4] The most important step in preventing this process, which starts with CHB and progresses to liver failure, is antiviral treatment. ETV and TDF, the most potent antivirals with minimal or no antiviral resistance, form the basis of CHB treatment. However, in the literature, information on the outcome and efficacy of long-term ETV and TDF treatments in the treatment of CHB is limited.

A study from Taiwan shows that regression of fibrosis and maybe reversal of cirrhosis can be achieved with prolonged suppression of viral replication.^[5] Among serum fibrosis models FIB4 and APRI scores, which are noninvasive indicators of liver fibrosis, often eliminate the need for liver biopsy and guide in evaluating liver stiffness.^[6] A few studies have shown that FIB4 and APRI scores can be reliable noninvasive serum markers for detecting significant fibrosis and cirrhosis in CHB patients.^[7,8] In this study, we concluded that the rate of cirrhotic patients before antiviral treatment decreased after at least 10 years of antiviral treatment with ETV or TDF, compared to the beginning of treatment. We also achieved significant regression in FIB4 and APRI scores with ETV and TDF treatments.

The most important step in preventing CHB-related morbidity and mortality is to ensure continuous viral suppression.^[9] Both ETV and TDF have very satisfactory virological and biochemical suppression.^[10] Very effective results have been obtained with ETV and TDF treatments, both in reducing ALT and HBV-DNA levels and in HBeAg loss.^[11,12] In our study, we achieved a significant decrease in liver enzymes and a very good virological response in both TDF and ETV groups. The rate of HBeAg loss was 23.8% with TDF and 16.6% with ETV in our study. These rates were similar to previous studies.^[13,14] HBsAg loss, which is quite rare,^[15] was also found at a low rate in our study (2.9%).

Antiviral treatment is very important in preventing decompensation and HCC, which are important causes of morbidity and mortality in cirrhosis.^[16] Unfortunately, the risk of HCC has not been eliminated despite treatment. In our study, HCC was most commonly seen in patients with cirrhosis and older age, and it is consistent with the literature.^[17] Although TDF and ETV are known to be effective in preventing progression to cirrhosis and the development of HCC in CHB patients,^[18,19] there is little information regarding the comparison of effectiveness between TDF and ETV. In our study, it was clearly understood that the choice of drug therapy was not an independent variable affecting the development of HCC in multivariate Cox regression analysis. In a French study, there was no difference between the drugs in the 5-year development of HCC in CHB patients receiving TDF and ETV.^[20] The predominant genotype with CHB patients in Türkiye is genotype D.^[21] In a multicenter study consisting mostly of genotype D CHB patients, it was observed that there was no difference in the development of HCC between patients receiving TDF and ETV.^[22] Some risk scores predict the development of HCC in CHB patients receiving antiviral therapy. In PAGE-B, one of these scores, platelet count, age, and gender are evaluated. Its effectiveness in predicting HCC in CHB patients receiving ETV and TDF treatment has been known.^[23] In this study, it was observed that PAGE-B scores of patients who developed HCC were significantly higher.

It is known that nucleos(t)ide analogs used in the treatment of CHB have negative effects on kidneys and bone metabolism, especially TDF, in the long-term.^[24] In our study, drug side effects were more common in the TDF group.

There are some limitations to the study. First of all, this study was conducted in a single center and in the Turkish population. The predominant genotype with CHB patients in Türkiye is genotype D.^[25] Therefore, the data here belong to patients with CHB due to genotype D. Studying a specific population infected with a single genotype will also contribute to the homogenization of the study. On the other hand, we believe that it provides significant contributions and can inspire subsequent studies, as it is the longest follow-up study in this field in the literature.

5. Conclusions

After at least 10 years of treatment, HBV-DNA negativity was achieved in almost all patients, and HBsAg sero-clearance was rarely achieved. Treatment-related side effects were more common in TDF. However, the overall rate of treatment-related adverse events and discontinuation of medication were low. The overall rate of treatment-related adverse events was 18.5% (n = 32). The overall rate of patients requiring medication change due to adverse events was 10.4% (n = 18). Nearly half of cirrhotic patients regressed to the non-cirrhotic stage. Very few patients developed HCC and the long-term mortality rate was similar to the general population.

Author contributions

Conceptualization: Zülal İstemihan, Sezen Genç Uluçeçen, Aynure Rüstemzade, Kanan Nuriyev, Filiz Akyüz, Sabahattin Kaymakoğlu.

Data curation: Zülal İstemihan, Gamze Kemeç, Timurhan Cebeci, Okan Çetin, Sezen Genç Uluçeçen, Aynure Rüstemzade, Kanan Nuriyev, Filiz Akyüz, Kadir Demir, Fatih Beşişik.

Formal analysis: Zülal İstemihan, Bilger Çavuş, Kadir Demir, Fatih Beşişik, Sabahattin Kaymakoğlu.

Investigation: Zülal İstemihan, Gamze Kemeç, Timurhan Cebeci, Okan Çetin, Asli Çifcibaşi Örmeci, Filiz Akyüz, Kadir Demir, Fatih Beşişik.

Methodology: Zülal İstemihan, Gamze Kemeç, Timurhan Cebeci, Okan Çetin, Bilger Çavuş, Asli Çifcibaşi Örmeci, Kadir Demir, Fatih Beşişik, Sabahattin Kaymakoğlu.

Resources: Zülal İstemihan, Gamze Kemeç, Timurhan Cebeci, Okan Çetin, Sezen Genç Uluçeçen, Bilger Çavuş, Asli Çifcibaşi Örmeci, Filiz Akyüz, Kadir Demir, Fatih Beşişik, Sabahattin Kaymakoğlu.

Supervision: Sabahattin Kaymakoğlu.

Validation: Zülal İstemihan, Bilger Çavuş.

Visualization: Asli Çifcibaşi Örmeci.

Writing – original draft: Zülal İstemihan, Sabahattin Kaymakoğlu.

Writing – review & editing: Zülal İstemihan, Sabahattin Kaymakoğlu.