Professional Documents
Culture Documents
Long-Term Outcomes of Hepatitis B Virus-Related Cirrhosis Treated With Nucleos (T) Ide Analogs
Long-Term Outcomes of Hepatitis B Virus-Related Cirrhosis Treated With Nucleos (T) Ide Analogs
ScienceDirect
ORIGINAL ARTICLE
Received 23 May 2016; received in revised form 19 August 2016; accepted 22 August 2016
KEYWORDS Background/purpose: This study aimed to evaluate the outcomes of chronic hepatitis B
cirrhosis; patients with cirrhosis who received long-term nucleos(t)ide analog therapy.
hepatitis B virus; Methods: A total of 546 consecutive cirrhotic patients treated with entecavir (n Z 359), tel-
hepatocellular bivudine (n Z 104), or tenofovir (n Z 83) for chronic hepatitis B were enrolled. The incidence
carcinoma; of hepatocellular carcinoma (HCC) and overall survival were evaluated.
nucleos(t)ide Results: During a median follow-up of 39 months, 56 (10.3%) patients developed HCC and 14
analogs; (2.6%) patients died. These outcomes were not associated with different antiviral use. Cox pro-
survival portional hazard analysis showed that old age (60 years) [hazard ratio (HR), 1.74; p Z 0.046],
statin use (HR, 2.42; p Z 0.017), low platelet count (<100,000/mL; HR, 2.00; p Z 0.039), and
variceal bleeding history (HR, 5.12; p < 0.001) were independent factors for HCC development.
With regard to survival, ChildePugh B/C (HR, 3.78; p Z 0.039) and low platelet count (<105/mL;
HR, 7.82; p Z 0.049) were independent factors. The estimated glomerular filtration rate signif-
icantly increased in patients receiving telbivudine (p Z 0.047), but decreased in those receiving
tenofovir (p < 0.001) at Year 2. Tenofovir use (HR, 1.98; p Z 0.005) was one of the independent
factors associated with the progression of chronic kidney disease stage.
Conclusion: Long-term nucleos(t)ide analog therapy does not guarantee against the HCC devel-
opment and mortality in chronic hepatitis B-related cirrhotic patients. Careful HCC surveillance
is necessary in patients with old age, statin use, low platelet count, and variceal bleeding his-
tory.
Copyright 2016, Formosan Medical Association. Published by Elsevier Taiwan LLC. This is an
open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/
by-nc-nd/4.0/).
Conflicts of interest: The authors have no conflicts of interest relevant to this article.
* Corresponding author. Division of Hepatogastroenterology, Department of Internal Medicine, Kaohsiung Chang Gung Memorial Hospital,
123 Ta Pei Road, Niao Sung, 833 Kaohsiung, Taiwan.
E-mail address: chh4366@yahoo.com.tw (C.-H. Hung).
http://dx.doi.org/10.1016/j.jfma.2016.08.006
0929-6646/Copyright 2016, Formosan Medical Association. Published by Elsevier Taiwan LLC. This is an open access article under the CC
BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Please cite this article in press as: Tsai M-C, et al., Long-term outcomes of hepatitis B virus-related cirrhosis treated with nucleos(t)ide
analogs, Journal of the Formosan Medical Association (2016), http://dx.doi.org/10.1016/j.jfma.2016.08.006
+ MODEL
2 M.-C. Tsai et al.
Please cite this article in press as: Tsai M-C, et al., Long-term outcomes of hepatitis B virus-related cirrhosis treated with nucleos(t)ide
analogs, Journal of the Formosan Medical Association (2016), http://dx.doi.org/10.1016/j.jfma.2016.08.006
+ MODEL
NUCs for HBV cirrhosis 3
Diagnostics, Indianapolis, IN, USA) according to the manu- log rank test. Progression of CKD stage was defined as an
facturers instructions. increase of at least one stage in the follow-up. Univariate
and multivariate analyses were carried out to identify in-
dependent factors using the Cox proportional hazard
Statistical analysis regression models. All analyses were carried out using SPSS
software version 15.0 (SPSS, Inc., Chicago, IL, USA). All
Continuous data are expressed as mean standard devia- statistical tests were two tailed, and a p value < 0.05 was
tion, and categorical data are expressed as numbers (per- considered statistically significant.
centage). Comparisons of differences in the categorical
data between groups were performed using the chi-square
test or Fisher exact test. Distributions of continuous vari- Results
ables were analyzed by the ManneWhitney U test or one-
way analysis of variance test with the least significant dif- Baseline characteristics
ference post hoc correction between groups where appro-
priate. Paired t test was performed to compare the A total of 546 CHB-related cirrhotic patients receiving ETV
estimated glomerular filtration rate (GFR) in serial (n Z 359), LdT (n Z 104), or TDF (n Z 83) were enrolled.
measurements. The baseline characteristics are shown in Table 1. Patients
Cumulative incidences of HCC development, overall in the TDF group were younger and had a higher platelet
mortality, and progression of chronic kidney disease (CKD) count compared with those in the ETV and LdT groups. The
stage were analyzed by the KaplaneMeier method with a mean HBV DNA was lower in the LdT group than in the ETV
Please cite this article in press as: Tsai M-C, et al., Long-term outcomes of hepatitis B virus-related cirrhosis treated with nucleos(t)ide
analogs, Journal of the Formosan Medical Association (2016), http://dx.doi.org/10.1016/j.jfma.2016.08.006
+ MODEL
4 M.-C. Tsai et al.
Figure 1 Cumulative incidences of HCC stratified by risk factors at baseline: (A) age, (B) statin, (C) ChildePugh classification, (D)
platelet count, (E) variceal bleeding history, and (F) NUCs. EVB Z esophageal variceal bleeding; HCC Z hepatocellular carcinoma;
NUC Z nucleos(t)ide analog.
Please cite this article in press as: Tsai M-C, et al., Long-term outcomes of hepatitis B virus-related cirrhosis treated with nucleos(t)ide
analogs, Journal of the Formosan Medical Association (2016), http://dx.doi.org/10.1016/j.jfma.2016.08.006
+ MODEL
NUCs for HBV cirrhosis 5
and TDF groups. The mean duration of follow-up in the TDF two groups but the rates appeared to be lower in statin
group was shorter than that in the ETV and LdT groups, users. Patients with statin use also demonstrated a lower
since TDF became available only 3 years ago in Taiwan. trend of normalization rate of alanine transaminase,
Of them, 34 (6.2%) patients had received statin therapy although no significance was found at any time point.
for more than 3 months during the observation period,
including simvastatin (n Z 4), rosuvastatin (n Z 10), pit-
avastatin (n Z 2), fluvastatin (n Z 6), and atorvastatin Factors associated with HCC development
(n Z 12). The comparison between statin users and non-
users is shown in Table S1. A total of 56 patients developed HCC after a mean follow-
up period of 39.5 17.8 months. As shown in Figure 1, the
KaplaneMeier method showed that old age (60 years) at
Treatment response entry (p Z 0.043), statin use (p Z 0.002), Child B/C
cirrhosis (p Z 0.013), low platelet count (<105/mL)
The numbers of patients with undetectable HBV DNA in the (p Z 0.048), and a history of gastroesophageal variceal
ETV, LdT, and TDF groups were 268 of 311 (86%), 79 of 97 bleeding (p < 0.001) were associated with HCC develop-
(81%), and 65 of 79 (82%) at Year 1 (p Z 0.436), and 205 of ment. By contrast, antiviral medication did not affect this
221 (93%), 56 of 77 (73%), and 23 of 23 (100%) at Year 2 outcome. The cumulative DDD was not significantly
(p < 0.001), respectively. Treatment schedule was modi- different between HCC and non-HCC cases (834 552 vs.
fied in 49 patients due to viral breakthrough (n Z 20), 602 577, p Z 0.289). There were no significant differ-
suboptimal response (n Z 18), or side effect (n Z 11). ences in HCC occurrence between different statins.
Viral breakthrough and genotypic resistance were found in Based on stepwise Cox proportional hazard analysis, old
16 (15%) LdT-treated and four (1%) ETV-treated patients. age (60 years) [hazard ratio (HR), 1.74; 95% confidence
None of the patients changed NUC because of renal interval (CI), 1.01e2.99; p Z 0.046], statin use (HR, 2.42;
dysfunction. 95% CI, 1.17e5.01; p Z 0.017), low platelet count (<105/
As shown in Figure S1, patients with statin use had a mL) (HR, 2.00; 95% CI, 1.04e3.88; p Z 0.039), and gastro-
lower rate of undetectable HBV DNA at Week 24 than esophageal variceal bleeding history (HR, 5.12; 95% CI,
nonstatin users (50% vs. 71%, p Z 0.041). There were no 2.62e10.01; p < 0.001) were independent risk factors for
significant differences at Years 1, 2, and 3 between these HCC development (Table 2).
Table 2 Univariate and multivariate Cox regression analyses of factors associated with HCC development.
Comparison Univariate analyses Stepwise multivariate analyses
Hazard ratio (95% CI) p Hazard ratio (95% CI) p
Age (y) 60 vs. <60 1.73 (1.01e2.95) 0.046 1.74 (1.01e2.99) 0.046
Sex Male vs. female 1.11 (0.61e2.03) 0.738 d d
Body mass index Per 1 kg/m2 increase 1.02 (0.95e1.10) 0.572 d d
Diabetes mellitus Yes vs. no 1.39 (0.78e2.48) 0.272 d d
Metformin vs. no 1.05 (0.48e2.32) 0.902 d d
Hypertension Yes vs. no 1.70 (0.98e2.95) 0.061 d d
Statin use Yes vs. no 3.00 (1.45e6.12) 0.003 2.42 (1.17e5.01) 0.017
Antiplatelet agent use Yes vs. no 1.79 (0.65e4.97) 0.261 d d
Antiviral drug Telbivudine vs. entecavir 0.68 (0.32e1.45) 0.315 d d
Tenofovir vs. entecavir 0.52 (0.12e2.22) 0.378 d d
Baseline HBV-DNA Per log10 copies/mL increase 1.00 (0.82e1.22) 0.996 d d
HBV DNA undetectable Year 1 Yes vs. no 0.68 (0.33e1.41) 0.294 d d
HBV DNA undetectable Year 2 Yes vs. no 0.74 (0.23e2.42) 0.620 d d
HBV genotype B vs. C 0.78 (0.44e1.41) 0.416 d d
HBsAg titer Per 1 IU/mL increase 1.00 (1.00e1.00) 0.438 d d
HBeAg positive Yes vs. no 1.44 (0.81e2.57) 0.217 d d
ChildePugh class B or C vs. A 2.10 (1.16e3.80) 0.014 d d
Ascites Yes vs. no 3.03 (1.65e5.56) <0.001 d d
Total bilirubin Per 1 mg/dL increase 0.98 (0.89e1.07) 0.607 d d
Albumin Per 1 g/dL increase 0.53 (0.32e0.89) 0.016 d d
Creatinine Per 1 mg/dL increase 1.01 (0.84e1.21) 0.915 d d
MELD score Per 1 increase 1.03 (0.96e1.09) 0.450 d d
Alpha-fetoprotein Per 1 ng/mL increase 1.00 (0.99e1.00) 0.874 d d
Platelet (103/mL) <100 vs. 100 1.83 (1.08e3.10) 0.025 2.00 (1.04e3.88) 0.039
Variceal bleeding history Yes vs. no 7.47 (3.93e14.20) <0.001 5.12 (2.62e10.01) <0.001
CI Z confidence interval; HBeAg Z hepatitis B e antigen; HBsAg Z hepatitis B surface antigen; HBV Z hepatitis B virus;
HCC Z hepatocellular carcinoma; MELD Z model for end-stage liver disease.
Please cite this article in press as: Tsai M-C, et al., Long-term outcomes of hepatitis B virus-related cirrhosis treated with nucleos(t)ide
analogs, Journal of the Formosan Medical Association (2016), http://dx.doi.org/10.1016/j.jfma.2016.08.006
+ MODEL
6 M.-C. Tsai et al.
Factors associated with overall survival cirrhosis (HR, 3.78; 95% CI, 1.32e10.84, p Z 0.013) and low
platelet count (<105/mL; HR, 7.82; 95% CI, 1.01e60.41,
A total of 14 patients died during the follow-up period; 10 p Z 0.049) were independent factors (Table 3).
(71%) of them suffered from liver-related death: seven died
of HCC and three of complications associated with cirrhosis.
Of the four patients with non-liver-related death, two died Factors predicting progression of CKD stage
of malignancies other than HCC, one of acute stroke, and
one of an unknown cause. The progression of CKD stage was recorded from baseline to
As shown in Figure 2, the KaplaneMeier method showed follow-up. After excluding 11 patients (9 with end-stage
that Child B/C cirrhosis (p Z 0.001) and low platelet count renal disease and 2 with CKD Stage IV), univariate analyses
(<105/mL) (p Z 0.013) were associated with overall survival. showed that hypertension (p Z 0.002), TDF use (p < 0.001),
Univariate analyses showed that Child B/C cirrhosis and pretreatment CKD Stage I (p < 0.001) were significant
(p Z 0.002), ascites (p Z 0.001), albumin level (p Z 0.006), factors associated with the progression of CKD stage. Based
model for end-stage liver disease score ({10 * [(0.957 * ln on stepwise logistic regression analyses, hypertension (HR,
(creatinine)] [0.378 * ln (bilirubin)] [1.12 * ln 2.58; 95% CI, 1.65e4.04, p < 0.001), TDF use (HR, 1.98; 95%
(INR)]} 6.43) (p Z 0.002), and low platelet count (<105/mL) CI, 1.23e3.21, p Z 0.005), and pretreatment CKD Stage I
(p Z 0.016) were significant factors for overall survival. By (HR, 6.45; 95% CI, 0.71e11.19, p < 0.001) were indepen-
the stepwise Cox proportional hazard model, Child B/C dent predictors (Table 4).
Figure 2 Overall survival stratified by risk factors at baseline: (A) ChildePugh classification, (B) platelet count, (C) MELD score,
and (D) NUCs. MELD Z model for end-stage liver disease; NUC Z nucleos(t)ide analog.
Please cite this article in press as: Tsai M-C, et al., Long-term outcomes of hepatitis B virus-related cirrhosis treated with nucleos(t)ide
analogs, Journal of the Formosan Medical Association (2016), http://dx.doi.org/10.1016/j.jfma.2016.08.006
+ MODEL
NUCs for HBV cirrhosis 7
Table 3 Univariate and multivariate Cox regression analyses of factors associated with overall survival.
Comparison Univariate analyses Stepwise multivariate analyses
Hazard ratio (95% CI) p Hazard ratio (95% CI) p
Age (y) 60 vs. <60 1.87 (0.63e5.58) 0.263 d d
Sex Male vs. female 1.37 (0.38e4.90) 0.631 d d
Body mass index Per 1 kg/m2 increase 1.02 (0.89e1.18) 0.789 d d
Diabetes mellitus Yes vs. no 0.59 (0.13e2.64) 0.489 d d
Hypertension Yes vs. no 0.52 (0.12e2.34) 0.396 d d
Statin use Yes vs. no 0.05 (0.00e622.5) 0.524 d d
Antiplatelet agent use Yes vs. no 0.05 (0.00e3997) 0.597 d d
Antiviral drug Telbivudine vs. entecavir 1.09 (0.30e3.96) 0.893 d d
Tenofovir vs. entecavir 0.04 (0.00e128037) 0.675 d d
Baseline HBV DNA Per log10 copies/mL increase 1.16 (0.77e1.76) 0.470 d d
HBV genotype B vs. C 2.93 (0.61e14.13) 0.180 d d
HBsAg titer Per 1 IU/mL increase 1.00 (0.99e1.00) 0.380 d d
HBeAg positive Yes vs. no 0.56 (0.13e2.52) 0.452 d d
ChildePugh class B or C vs. A 5.12 (1.79e14.62) 0.002 3.78 (1.32e10.84) 0.013
Ascites Yes vs. no 5.80 (2.01e16.77) 0.001 d d
Total bilirubin Per 1 mg/dL increase 1.06 (0.99e1.13) 0.078 d d
Albumin Per 1 g/dL increase 0.30 (0.12e0.71) 0.006 d d
Creatinine Per 1 mg/dL increase 1.01 (0.84e1.21) 0.915 d d
MELD score Per 1 increase 1.14 (1.05e1.23) 0.002 d d
Alpha-fetoprotein Per 1 ng/mL increase 1.00 (0.99e1.00) 0.283
Platelet (103/mL) 100 vs. <100 3.82 (1.28e11.42) 0.016 7.82 (1.01e60.41) 0.049
Variceal bleeding history Yes vs. no 3.00 (0.67e13.42) 0.151 d d
CI Z confidence interval; HBeAg Z hepatitis B e antigen; HBsAg Z hepatitis B surface antigen; HBV Z hepatitis B virus; MELD Z model
for end-stage liver disease.
As shown in Figure 3, there was no significant difference bleeding represented the major risk factors. These results
in mean estimated GFR between baseline and Years 1, 2, 3, were consistent with those of previous reports.24e30 Not
and 4 among patients treated with ETV. By contrast, esti- only in natural history, but also in the treatment of CHB
mated GFR significantly increased in patients receiving LdT patients, old age is absolutely the independent risk factor
at Year 1 (p Z 0.045) and Year 2 (p Z 0.047), but decreased for HCC development. In particular, baseline platelet count
in those receiving TDF at Year 1 (p < 0.001) and Year 2 is an independent factor predicting both HCC development
(p < 0.001) compared with baseline. and overall survival, suggesting that platelet count may be
a useful surrogate marker for the severity of cirrhosis in
CHB.28 Previous studies have shown that patients with a
Discussion history of gastroesophageal variceal bleeding had a high
risk of developing HCC in long-term follow-up. Importantly,
This large cohort study indicated that long-term NUC this association is independent of the degree of liver
therapy does not guarantee against the development of dysfunction and the duration of liver disease.29,30
HCC and mortality in CHB-related cirrhotic patients. Age, Another intriguing finding is that statin use is an inde-
statin use, platelet count, and variceal bleeding history pendent risk factor for HCC development in our CHB-
were independent risk factors for HCC development. In related cirrhotic patients. Statins, 3-hydroxy-
addition, ChildePugh B/C and platelet count were inde- 3methylglutaryl-coenzyme A (HMG-CoA) reductase
pendent factors for overall survival. Our data were similar inhibitors, a class of cholesterol-lowering drugs, are the
to those in a previous report showing that the annual inci- second most prescribed therapeutic medication following
dence of HCC was 3.9% in LAM-treated HBV patients with painkillers.31 Long-term use of statin drugs has been shown
cirrhosis or advanced fibrosis.7 Another large multicenter to increase risk of breast,32 thyroid,33 and prostate can-
study in Caucasian patients showed 3.5% and 7.3% of annual cer,34 whereas many studies have reported the inverse as-
HCC rates, respectively, for patients with compensated and sociation between statin use and cancer risk.35e37 As
decompensated cirrhosis.24 Taken together, it is obvious regards HCC, recent observational studies have demon-
that hepatocarcinogenesis in cirrhotic patients persists strated that statin use may be associated with a lower risk
even after long-term NUC therapy. However, although our of HCC,37,38 although others have shown no association.39
study was not a prospective randomized control trial, the However, these population-based studies were built on
annual incidence of HCC in our treated patients appeared health insurance research database plus medical diagnosed
to be lower than that in the untreated controls, based on codes,37,38 thus there was no detailed clinical information
historical data (7.4e22.4%).7,8,10,24 such as virological data and cirrhosis. In contrast, our study
In analysis for HCC development, our findings confirmed group consisted of CHB patients with cirrhosis who received
that old age, low platelet counts, and a history of variceal long-term NUC treatment and regular follow-up. After
Please cite this article in press as: Tsai M-C, et al., Long-term outcomes of hepatitis B virus-related cirrhosis treated with nucleos(t)ide
analogs, Journal of the Formosan Medical Association (2016), http://dx.doi.org/10.1016/j.jfma.2016.08.006
+ MODEL
8 M.-C. Tsai et al.
Table 4 Univariate and multivariate Cox regression analyses of factors associated with progression of CKD stage.a
Comparison Univariate analyses Stepwise multivariate analyses
Hazard ratio (95% CI) p Hazard ratio (95% CI) p
Age (y) 60 vs. <60 1.09 (0.62e1.92) 0.770 d d
Sex Male vs. female 0.73 (0.47e1.12) 0.149 d d
Body mass index Per 1 kg/m2 increase 1.02 (0.97e1.08) 0.444 d d
Diabetes mellitus Yes vs. no 1.08 (0.66e1.76) 0.775 d d
Hypertension Yes vs. no 1.93 (1.26e2.95) 0.002 2.58 (1.65e4.04) <0.001
Antiviral drug Tenofovir vs. nontenofovir 2.42 (1.51e3.89) <0.001 1.98 (1.23e3.21) 0.005
Baseline HBV DNA Per log10 copies/mL increase 1.16 (0.99e1.35) 0.069 d d
HBeAg positive Yes vs. no 0.95 (0.57e1.57) 0.840 d d
HBV genotype B vs. C 0.80 (0.52e1.23) 0.306 d d
HBsAg titer Per 1 IU/mL increase 1.00 (1.00e1.00) 0.786 d d
ChildePugh class B or C vs. A 1.32 (0.78e2.23) 0.305 d d
Ascites Yes vs. no 1.01 (0.51e2.00) 0.983 d d
CKD stage I vs. II/III 5.68 (3.30e9.79) <0.001 6.45 (3.71e11.19) <0.001
MELD score Per 1 increase 1.02 (0.966e1.08) 0.498 d d
Platelet (103/mL) 100 vs. <100 0.92 (0.55e1.55) 0.754 d d
Variceal bleeding history Yes vs. no 1.45 (0.63e3.31) 0.381 d d
CI Z confidence interval; CKD Z chronic kidney disease; HBeAg Z hepatitis B e antigen; HBsAg Z hepatitis B surface antigen;
HBV Z hepatitis B virus; MELD Z model for end-stage liver disease.
a
Progression of CKD stage was defined as an increase of at least one stage in the follow-up.
Please cite this article in press as: Tsai M-C, et al., Long-term outcomes of hepatitis B virus-related cirrhosis treated with nucleos(t)ide
analogs, Journal of the Formosan Medical Association (2016), http://dx.doi.org/10.1016/j.jfma.2016.08.006
+ MODEL
NUCs for HBV cirrhosis 9
hyperlipidemia and statin use remains to be clarified. 13. Liaw YF, Gane E, Leung N, Zeuzem S, Wang Y, Lai CL, et al. 2-
Third, the median follow-up time of TDF-treated patients Year GLOBE trial results: telbivudine is superior to lamivudine
was relatively short, since this drug became available only in patients with chronic hepatitis B. Gastroenterology 2009;
3 years ago in Taiwan. A longer follow-up period in the TDF 136:486e95.
14. Tenney DJ, Rose RE, Baldick CJ, Pokornowski KA, Eggers BJ,
group should be needed in the future.
Fang J, et al. Long-term monitoring shows hepatitis B virus
In conclusion, CHB-related cirrhotic patients still have a resistance to entecavir in nucleoside-naive patients is rare
high risk of HCC despite long-term treatment with NUCs. through 5 years of therapy. Hepatology 2009;49:1503e14.
Statin use is associated with a significant increase in the risk 15. Menne S, Cote PJ, Korba BE, Butler SD, George AL, Tochkov IA,
of HCC among these patients. Although considerable un- et al. Antiviral effect of oral administration of tenofovir disoproxil
certainty is involved in the carcinogenesis associated with fumarate in woodchucks with chronic woodchuck hepatitis virus
statins, our observation suggests that careful HCC surveil- infection. Antimicrob Agents Chemother 2005;49:2720e8.
lance is necessary in patients with old age, statin use, low 16. Marcellin P, Asselah T, Boyer N. Fibrosis and disease progres-
platelet count, and variceal bleeding history. sion in hepatitis C. Hepatology 2002;36:S47e56.
17. Hung CH, Lu SN, Wang JH, Lee CM, Chen TM, Tung HD, et al.
Correlation between ultrasonographic and pathologic di-
Acknowledgments agnoses of hepatitis B and C virus-related cirrhosis. J Gastro-
enterol 2003;38:153e7.
18. Wu CY, Chen YJ, Ho HJ, Hsu YC, Kuo KN, Wu MS, et al. Asso-
This study was supported in part by contract grant ciation between nucleoside analogues and risk of hepatitis B
CMRPG8C0961 from Chang Gung Memorial Hospital, Taiwan. virus-related hepatocellular carcinoma recurrence following
liver resection. JAMA 2012;308:1906e14.
19. WHO Collaborating Center for Drugs Statistics Methodology.
Appendix A. Supplementary data
ATC index with DDDs 2003. Oslo: WHO; 2003.
20. Chen TH, Chen CJ, Yen MF, Lu SN, Sun CA, Huang GT, et al.
Supplementary data related to this article can be found at Ultrasound screening and risk factors for death from hepato-
http://dx.doi.org/10.1016/j.jfma.2016.08.006. cellular carcinoma in a high risk group in Taiwan. Int J Cancer
2002;98:257e61.
21. Bruix J, Sherman M, Practice Guidelines Committee, American
References Association for the Study of Liver Diseases. Management of
hepatocellular carcinoma. Hepatology 2005;42:1208e36.
1. Liaw YF, Leung N, Kao JH, Piratvisuth T, Gane E, Han KH, et al., 22. Bruix J, Sherman M, American Association for the Study of Liver
Chronic Hepatitis B Guideline Working Party of the Asian- Diseases. Management of hepatocellular carcinoma: an up-
Pacific Association for the Study of the Liver. Asian-Pacific date. Hepatology 2011;53:1020e2.
consensus statement on the management of chronic hepatitis 23. Hung CH, Chen CH, Lu SN, Wang JH, Hu TH, Huang CM, et al.
B: a 2008 update. Hepatol Int 2008;2:263e83. Precore/core promoter mutations and hepatitis B virus geno-
2. McMahon BJ. Epidemiology and natural history of hepatitis B. type in hepatitis B and C dually infected patients treated with
Semin Liver Dis 2005;25(Suppl. 1):3e8. interferon-based therapy. Antiviral Res 2012;93:55e63.
3. El-Serag HB. Hepatocellular carcinoma. N Engl J Med 2011; 24. Papatheodoridis GV, Dalekos GN, Yurdaydin C, Buti M, Goulis J,
365:1118e27. Arends P, et al. Incidence and predictors of hepatocellular
4. Liaw YF, Chu CM. Hepatitis B virus infection. Lancet 2009;373: carcinoma in Caucasian chronic hepatitis B patients receiving
582e92. entecavir or tenofovir. J Hepatol 2015;62:363e70.
5. Chen CJ, Yang HI, Su J, Jen CL, You SL, Lu SN, et al. Risk of 25. Kumada T, Toyoda H, Tada T, Kiriyama S, Tanikawa M,
hepatocellular carcinoma across a biological gradient of serum Hisanaga Y, et al. Effect of nucleos(t)ide analogue therapy on
hepatitis B virus DNA level. JAMA 2006;295:65e73. hepatocarcinogenesis in chronic hepatitis B patients: a pro-
6. Chen CJ, Yang HI, Iloeje UH. Hepatitis B virus DNA levels and pensity score analysis. J Hepatol 2013;58:427e33.
outcomes in chronic hepatitis B. Hepatology 2009;49(5 Suppl.): 26. Hsu YC, Wu CY, Lane HY, Chang CY, Tai CM, Tseng CH, et al.
S72e84. Determinants of hepatocellular carcinoma in cirrhotic patients
7. Liaw YF, Sung JJ, Chow WC, Farrell G, Lee CZ, Yuen H, et al. treated with nucleos(t)ide analogues for chronic hepatitis B. J
Lamivudine for patients with chronic hepatitis B and advanced Antimicrob Chemother 2014;69:1920e7.
liver disease. N Engl J Med 2004;351:1521e31. 27. Chen CJ, Yang HI. Natural history of chronic hepatitis B
8. Lai CL, Yuen MF. Prevention of hepatitis B virus-related hepa- revealed. J Gastroenterol Hepatol 2011;26:628e38.
tocellular carcinoma with antiviral therapy. Hepatology 2013; 28. Lu SN, Wang JH, Liu SL, Hung CH, Chen CH, Tung HD, et al.
57:399e408. Thrombocytopenia as a surrogate for cirrhosis and a marker for
9. Niro GA, Ippolito AM, Fontana R, Valvano MR, Gioffreda D, the identification of patients at high-risk for hepatocellular
Iacobellis A, et al. Long-term outcome of hepatitis B virus- carcinoma. Cancer 2006;107:2212e22.
related chronic hepatitis under protracted nucleos(t)ide ana- 29. Chen WC, Lo GH, Lai KH, Cheng JS, Hsu PI, Lin CK. Develop-
logues. J Viral Hepatol 2013;20:502e7. ment of hepatocellular carcinoma after successful manage-
10. Sung JJ, Tsoi KK, Wong VW, Li KC, Chan HL. Meta-analysis: ment of esophageal variceal bleeding. J Chin Med Assoc 2004;
treatment of hepatitis B infection reduces risk of hepatocel- 67:557e64.
lular carcinoma. Aliment Pharmacol Ther 2008;28:1067e77. 30. Ripoll C, Groszmann RJ, Garcia-Tsao G, Bosch J, Grace N,
11. Cho JY, Paik YH, Sohn W, Cho HC, Gwak GY, Choi MS, et al. Burroughs A, et al. Hepatic venous pressure gradient predicts
Patients with chronic hepatitis B treated with oral antiviral development of hepatocellular carcinoma independently of
therapy retain a higher risk for HCC compared with patients severity of cirrhosis. J Hepatol 2009;50:923e8.
with inactive stage disease. Gut 2014;63:1943e50. 31. Gutt R, Tonlaar N, Kunnavakkam R, Karrison T,
12. Lok AS, Lai CL, Leung N, Yao GB, Cui ZY, Schiff ER, et al. Long- Weichselbaum R, Liauw S. Statin use and risk of prostate
term safety of lamivudine treatment in patients with chronic cancer recurrence in men treated with radiation therapy. J
hepatitis B. Gastroenterology 2003;125:1714e22. Clin Oncol 2010;28:2653e9.
Please cite this article in press as: Tsai M-C, et al., Long-term outcomes of hepatitis B virus-related cirrhosis treated with nucleos(t)ide
analogs, Journal of the Formosan Medical Association (2016), http://dx.doi.org/10.1016/j.jfma.2016.08.006
+ MODEL
10 M.-C. Tsai et al.
32. McDougall JA, Malone KE, Daling JR, Cushing-Haugen KL, 38. Tsan YT, Lee CH, Wang JD, Chen PC. Statins and the risk of
Porter PL, Li CI. Long-term statin use and risk of ductal hepatocellular carcinoma in patients with hepatitis B virus
and lobular breast cancer among women 55 to 74 years infection. J Clin Oncol 2012;30:623e30.
of age. Cancer Epidemiol Biomarkers Prev 2013;22: 39. Yang SC, Lee CM, Hu TH, Wang JH, Lu SN, Hung CH, et al.
1529e37. Virological response to entecavir reduces the risk of liver dis-
33. Hung SH, Lin HC, Chung SD. Statin use and thyroid cancer: a ease progression in nucleos(t)ide analogue-experienced HBV-
population-based case-control study. Clin Endocrinol (Oxf) infected patients with prior resistant mutants. J Antimicrob
2015;83:111e6. http://dx.doi.org/10.1111/cen.12570. Chemother 2013;68:2154e63.
34. Chang CC, Ho SC, Chiu HF, Yang CY. Statins increase the risk of 40. Fung J, Seto WK, Lai CL, Yuen MF. Extrahepatic effects of
prostate cancer: a population-based case-control study. Pros- nucleoside and nucleotide analogues in chronic hepatitis B
tate 2011;71:1818e24. treatment. J Gastroenterol Hepatol 2014;29:428e34.
35. Liu Y, Tang W, Wang J, Xie L, Li T, He Y, et al. Association 41. Gane EJ, Deray G, Liaw YF, Lim SG, Lai CL, Rasenack J, et al.
between statin use and colorectal cancer risk: a meta-analysis Telbivudine improves renal function in patients with chronic
of 42 studies. Cancer Causes Control 2014;25:237e49. hepatitis B. Gastroenterology 2014;146. 138e146 e5.
36. Wu XD, Zeng K, Xue FQ, Chen JH, Chen YQ. Statins are asso- 42. Tsai MC, Lee CM, Chiu KW, Hung CH, Tung WC, Chen CH, et al.
ciated with reduced risk of gastric cancer: a meta-analysis. Eur A comparison of telbivudine and entecavir for chronic hepatitis
J Clin Pharmacol 2013;69:1855e60. B in real-world clinical practice. J Antimicrob Chemother
37. Wu CY, Lin JT, Ho HJ, Su CW, Lee TY, Wang SY, et al. Associ- 2012;67:696e9.
ation of nucleos(t)ide analogue therapy with reduced risk of 43. Tsai MC, Yu HC, Hung CH, Lee CM, Chiu KW, Lin MT, et al.
hepatocellular carcinoma in patients with chronic hepatitis B: Comparing the efficacy and clinical outcome of telbivudine and
a nationwide cohort study. Gastroenterology 2014;147. entecavir naive patients with hepatitis B virus-related compen-
143e151 e5. sated cirrhosis. J Gastroenterol Hepatol 2014;29:568e75.
Please cite this article in press as: Tsai M-C, et al., Long-term outcomes of hepatitis B virus-related cirrhosis treated with nucleos(t)ide
analogs, Journal of the Formosan Medical Association (2016), http://dx.doi.org/10.1016/j.jfma.2016.08.006