Entecavir versus tenofovir disoproxil fumarate on the reduction of incidence of hepatocellular carcinoma in patients with chronic hepatitis B‐related liver cirrhosis

This study aimed to compare the effect of long‐term continuous entecavir (ETV) compared with tenofovir disoproxil fumarate (TDF) on the reduction of hepatocellular carcinoma (HCC) development in patients with chronic hepatitis B (CHB) related liver cirrhosis. This study recruited patients who had CHB‐related liver cirrhosis and received ETV or TDF treatment for more than 6 months. Regular assessments of ultrasonography and alpha‐fetoprotein test were arranged every 3 months for HCC detection. Five‐year cumulative incidence of HCC and risk factors for HCC development were analyzed. A total of 286 consecutive cirrhotic patients were included, 198 in the ETV group and 88 in the TDF group. During a median follow‐up of 57.5 months, 25 (12.6%) patients in the ETV group and 12 (13.6%) patients in the TDF group developed HCC. The 5‐year cumulative incidence of HCC was comparable between the ETV and TDF groups (6.57% vs. 9.09%, log‐rank p = .242). Multivariate Cox proportional hazard analysis revealed that male, old age, diabetes, and low platelet count were independent risk factors for HCC development. This study observed that long‐term ETV or TDF provided comparable preventive effects on HCC development in patients with CHB‐related liver cirrhosis.


| INTRODUCTION
Hepatitis B virus (HBV) infection is a global health problem which causes a wide spectrum of liver diseases, including acute or fulminant hepatitis, inactive carrier state, chronic hepatitis, cirrhosis, and hepatocellular carcinoma (HCC). 1 It is estimated that HBV infection will account for 3.3 million new cases 796 000 deaths in 2020 due to HBV-related liver cirrhosis, HCC, or both. 2 Current guideline recommends nucleotide analogues (NAs) as one of the first-line therapy for patients with chronic hepatitis B (CHB) and the preferred initial NAs includes entecavir (ETV), tenofovir disoproxil fumarate (TDF), and tenofovir alafenamide. 35][6] However, most of the included patients in these meta-analyses were non-cirrhotic.For patients with CHB-related cirrhosis, evidence regarding the association of TDF versus ETV with risk of HCC development was inconsistent.The meta-analysis by Jeong et al. reported that TDF significantly lowered the risk of HCC development compared with ETV in CHB patients with cirrhosis (risk ratio = 0.69, 95% confidence interval [CI] = 0.56-0.84), 6whereas the metaanalysis by Dave et al. reported no difference between ETV-and TDF-treated groups (adjusted hazard ratio [HR] = 0.90, 95% CI = 0.66-1.23). 5Since the conclusion remains controversial, this study aimed to evaluate the effectiveness of continuous ETV versus TDF therapy on the development of HCC for patients with CHB-related liver cirrhosis.

| Patients
This observational study recruited the patients who received ETV or TDF treatment for CHB for more than 6 months between December 2007 and May 2020 at Tainan Municipal Hospital.All patients had positive hepatitis B surface antigen (HBsAg) for more than 6 months, serum HBV DNA level ≥2000 IU/mL, and liver cirrhosis.Liver cirrhosis was diagnosed by (a) histopathological exams (Metavir score F4 or Ishak score F5-6) or by (b) ultrasonographic findings of liver cirrhosis combined with splenomegaly or ultrasonographic findings of liver cirrhosis combined with endoscopic findings of gastric or esophageal varices.Patients would be excluded if they had prior history of HCC, previously treated with pegylated interferon-alpha, lamivudine or telbivudine, co-infection of human immunodeficiency, hepatitis A virus, hepatitis C virus, hepatitis D virus, or hepatitis E virus.A flowchart summarized the patient enrollment in Figure 1.This study was approved by the Institutional Review Board of Tainan Municipal Hospital (IRB Number: 1110101), and was conducted in accordance with the Declaration of Helsinki.Informed consent was waived due to the retrospective nature of the study and the analysis used anonymous clinical data.

| Treatment and evaluation
All patients were given of ETV (0.5 mg once daily) or TDF (300 mg once daily) and received regular follow-up.Regular assessments which were performed every 3 months included biochemical tests, complete blood count, ultrasonography, and alpha-fetoprotein test.Multiphase magnetic resonance imaging or computed tomography would be arranged for patients with a clinical suspicion of HCC, for example, a newly developed nodular lesion on ultrasonography or elevated alphafetoprotein level.The diagnosis of HCC was based on noninvasive imaging criteria and/or histopathological exams.Imaging criteria was the combination of hypervascularity in the late arterial phase and washout in the portal venous phase and/or delayed phases.
Collected data included demographics, baseline characteristics, Child-Pugh score, and HBV e-antigen (HBeAg) status.Duration of follow-up was determined by the interval from the initiation of ETV or TDF treatment to the date of HCC diagnosis or the end of data collection (i.e., May 31, 2020).Patients with a death date in the admission file were censored.A predictive score, risk estimation for hepatocellular carcinoma in chronic hepatitis B (REACH-B), was calculated for each patient. 7V group TDF group  F I G U R E 2 Cumulative incidence of HCC in patients who received entecavir or tenofovir disoproxil fumarate for chronic hepatitis B with compensated cirrhosis.models.Results were presented as HR with corresponding 95% CI.Standardized incidence ratio (SIR) was calculated to compare the observed and predicted incidence of HCC.A two-tail p < .05 was considered statistical significance.Statistical analyses were performed using SPSS 15.0 (SPSS Inc., Chicago, IL).

| RESULTS
This study recruited a total of 286 consecutive cirrhotic patients, 198 (69.2%) patients in the ETV group and 88 (30.8%) patients in the TDF group.Mean age was 60.7 (10.5) years and 205 (71.7%) patients were male.The median duration of follow-ups was 57.5 (range 3-149) months.There was no significant difference in demographics and baseline characteristics between the ETV and TDF groups (Table 1).Twenty-five (12.6%) patients in the ETV group and 12 (13.6%)patients in the TDF group developed HCC during follow-ups, two patients in the ETV group died; one died of HCC, and one died of complication associated with cirrhosis.The 5-year cumulative incidence of HCC was 6.57% in the ETV group and 9.09% in the TDF group (log-rank, p = .242).(Figure 2).
As presented in Table 3, the last observed HCC case in the ETV group was at week 120.At the end of follow-up, 25 HCC cases were observed and 34.08 HCC cases were predicted by the REACH-B model with an SIR of 0.73 (95% CI = 0.47-1.08,p = .96).For the TDF group, the last HCC case was at week 72.Twelve HCC cases were observed in the TDF group, whereas REACH-B model predicted 9.13 HCC cases with an SIR of 1.31 (95% CI = 0.67-2.29,p = .21).Figures 3 and 4, respectively, showed the graphical comparisons of the observed and predicted HCC cases in the ETV and TDF groups.Although several studies have demonstrated the benefits of TDF versus ETV on reduced risk of HCC development for CHB patients, evidence remains inconsistent for cirrhotic patients.This study found that the 5-year cumulative incidence of HCC was not significantly different between the ETV and TDF groups (6.57% vs. 9.09%, p = .242).Male, old age, presence of diabetes, and low platelet count at baseline were independent risk factors for HCC development for patients with CHB-related liver cirrhosis.][6] However, a multicenter study including 1843 CHB patients reported that liver cirrhosis was significantly associated with higher risk of HCC development, whereas antiviral treatments was not. 8As real-world studies usually recruited the CHB patients with heterogeneous HBV features, specific evidence for cirrhotic patients is necessary.In this study, only the patients with CHB-related liver cirrhosis were included and the results showed no difference in HCC development between the ETV and TDF groups.Our results were comparable with several previous cohort studies.A retrospective cohort study from Southern Taiwan enrolled 546 cirrhotic patients who were treated with NAs for CHB.This study found that there were no significant differences in HCC occurrence between ETV-and TDF-treated groups. 9nother study from Southern Taiwan recruited 894 treatment-naïve patients with CHB-related compensated cirrhosis and used several adjusting and matching models to analyze the association of ETV compared with TDF with HCC incidence.The results showed that TDF was associated with numerical lower risk of HCC development compared with ETV, but the difference did not reach a statistical significance. 10In a cohort study from Central and Southern Taiwan including 1560 patients with CHB-related liver cirrhosis, multivariate analysis showed that TDF was significantly associated with a lower risk of HCC compared with ETV in all patients.Antiviral therapy remained as a significant factor associated with HCC development in patient with compensated cirrhosis (n = 1320), but it was not a significant factor after adjustment with propensity score matching or inverse probability of treatment weighting. 11In a Korean cohort study recruiting 970 patients with CHB-related compensated cirrhosis, multivariate analysis showed that the HCC risk was not significantly different between the ETV and TDF group. 12Notably, a large population-based cohort study from Korea reported different conclusion from ours.The subgroup analysis of 2914 propensitymatched pairs of CHB-related cirrhosis patients revealed that TDF was associated with a significantly lower risk of HCC by 33% compared with ETV. 13 These contradictory findings might be explained by the different inclusion and exclusion criteria, patient characteristics, and confounders in these studies.Although Choi et al. provided two possible mechanisms by which TDF outperforms ETV in reducing the HCC risk, supporting human data for the mechanism is limited. 14,15While current observations remain contradictory, more surveillance is needed.
The risk factors for HCC development in patients with CHB-related liver cirrhosis identified in the present study included male, age, diabetes, and platelet count, which were comparable with previous reports.However, this study failed to identify some variables which also have been shown to be risk factors for HCC development in this population, for example, albumin, 11 renal function,, 11 HCC family history, 10 gastrointestinal varices, 9,10 alpha-fetoprotein, 10 and medications for comorbidity. 9These scattered results reflected the heterogeneous characteristics of patients with HBV infection.
There were several limitations in this study.The first came from the non-randomized and retrospective study design.Potential selecting bias and reporting bias could not be avoided.Secondly, the number of patients in the TDF group (n = 88) was much smaller than that in the ETV group (n = 198) because ETV was approved by the Taiwan Food and Drug Administration years before TDF.A larger sample with long-term follow-up was required for further comparisons between ETV and TDF.Thirdly, this study did not analyze quantitative HBsAg level and HBV genotype, and also did not consider treatment responses and prior antiviral therapies.Neither quantitative HBsAg level nor HBV genotype was found to be an independent predictor of HCC development, 16 and the patients with HBeAg seroconversion had significantly lower risk of HCC than those with persistently positive HBeAg. 17Last but not the least, the magnitude of underestimation was even dramatic in patient with cirrhosis since the REACH-B model based on patient without evidence of cirrhosis.Therefore, these potential confounders should be considered in the future studies.
In conclusion, for patients with CHB-related liver cirrhosis, this observational study did not find significant difference in cumulative incidence of HCC between the patients treated with long-term continuous ETV and those with TDF.Although the results suggested comparable preventive effects of ETV versus TDF on HCC development, further comparisons of antiviral agents are warranted for the optimization of CHB management.

No− 2 − 3 − 5 − 2 − 6
History of HCC at enrollment 5 Co-infection with other hepatitis virus 6 Follow-up duration less than 6 months 4 Child-Pugh class C 3 Pegylated interferon, lamivudine, telbivudine experienced 8 At baseline, a total of 340 CHB cirrhotic patients from Tainan Municipal Hospital were enrolled according to inclusion criteria.Finally, a total of 286 consecutive cirrhotic patients were analyzed; ETV group (n = 198), TDF group (n = 88).F I G U R E 1 Flowchart summarizing the patient enrollment.T A B L E 1 Demographics and baseline characteristics of the included patients.
Cox proportional hazards models for risk factors of hepatocellular carcinoma development in patients who received entecavir or tenofovir disoproxil fumarate for chronic hepatitis B with compensated cirrhosis.
Continuous data are expressed as mean (standard deviation) and categorical data as count (percentage).Kaplan-Meier method with log-rank test was used to compare the cumulative incidence of HCC between the ETV and TDF groups.Risk factors for HCC development were assessed by Cox proportional hazards models.Demographics and variables with p < .2 in univariate models were selected into the multivariateT A B L E 2