Potential conflict of interest: Jurrien Reijnders has received speaker's honoraria from Novartis and Bristol-Myers Squibb. Fabien Zoulim is consultant for Bristol-Myers Squibb, Gilead, Roche, and Novartis. David Mutimer has received honoraria and grants from Bristol-Myers Squibb. Jörg Petersen is a consultant for Bristol-Myers Squibb and received research support. Maria Buti has received speaker's honoraria and an advisory board fee from Bristol-Myers Squibb. Florian van Bommel has received travel grants from Roche and Gilead. Thomas Berg is a consultant for, advises, is on the speakers' bureau of, and has received grants from Gilead, Roche, Novartis, Merck, Tibotec, and Vertex. Heiner Wedemeyer received research grants and speaker's honoraria from Bristol-Myers Squibb, Gilead, Roche, and Novartis. Harry Janssen received grants from and is a consultant for Bristol-Myers Squibb, Gilead Sciences, Novartis, Roche, and Merck.
Supported by the Foundation for Liver and Gastrointestinal Research Rotterdam. The study was conducted with support from the European Network of Excellence for Vigilance against Viral Resistance and a research grant from Bristol-Myers Squibb. (Bristol-Myers Squibb was not involved in data collection, data analysis, writing of the manuscript, or decision to publish.)
Entecavir (ETV) is a potent inhibitor of viral replication in nucleos(t)ide analogue (NA)-naïve chronic hepatitis B (CHB) patients. The aim of this study was to investigate the long term efficacy and safety of ETV in NA-naïve CHB patients, particularly in those with detectable hepatitis B virus (HBV) DNA after 48 weeks, in whom treatment adaptation is suggested by current guidelines. In a multicenter cohort study, we investigated 333 CHB patients treated with entecavir monotherapy. The NA-naïve population consisted of 243 patients, whereas 90 were NA-experienced. Virological response (VR) (HBV DNA <80 IU/mL) was achieved in 48%, 76%, and 90% of hepatitis B e antigen (HBeAg)-positive and in 89%, 98%, and 99% of HBeAg-negative NA-naïve patients at weeks 48, 96, and 144, respectively. Thirty-six of 175 (21%) NA-naïve patients with at least 48 weeks of follow-up had a detectable load at week 48 (partial virological response [PVR]). Twenty-nine (81%) patients with PVR reached VR during prolonged ETV monotherapy, and none of them developed ETV-resistance. Among 22 patients with HBV DNA <1,000 IU/mL at week 48, VR was achieved in 21 (95%) patients, compared with eight of 14 (57%) patients with HBV DNA ≥1,000 IU/mL. Continuous HBV DNA decline was observed in most patients without VR during follow-up, and in three patients adherence was suboptimal according to the treating physician. ETV was safe and did not affect renal function or cause lactic acidosis. Conclusion: ETV monotherapy can be continued in NA-naïve patients with detectable HBV DNA at week 48, particularly in those with a low viral load because long-term ETV leads to a virological response in the vast majority of patients. (HEPATOLOGY 2011;)
Current treatment guidelines consider nucleos(t)ide analogues (NAs) and pegylated interferon as first-line treatment for chronic hepatitis B (CHB). The ultimate goal of treatment is prevention of cirrhosis, hepatic decompensation, and hepatocellular carcinoma.1 Entecavir (ETV) is a cyclopentyl guanosine analogue that has shown superior biochemical, virological, and histological efficacy compared with lamivudine (LAM) in large phase III trials.2, 3 Moreover, genotypic resistance to ETV is rare in NA-naïve patients through 5 years of continuous therapy.4 However, the efficacy of ETV is seriously compromised in LAM-refractory chronic hepatitis B virus (HBV) patients with increasing rates of genotypic ETV resistance and patients experiencing a virological breakthrough.5 Recently, we translated these previous findings to clinical practice in a large European multicenter study, because ETV was very effective in NA-naïve patients during the first year of therapy, but less effective in patients with LAM resistance at baseline.6
Avoiding viral resistance is a cornerstone of CHB treatment, because resistance is associated with a worsened outcome.7 Moreover, persistent viremia has been identified as a risk factor for a dismal outcome after 2 years of treatment with telbivudine (LdT).8 Therefore, current European guidelines have focused on patients with a partial virological response (PVR), defined as a >1 log IU/mL decline in HBV DNA from baseline but a detectable load at week 24 (LAM and LdT) or week 48 (adefovir [ADV], ETV, and tenofovir [TDF]). It is suggested that these patients could be at risk for developing genotypic resistance and that treatment adaptation in patients with a persisting viral load after 48 weeks should thus be considered.9 However, evidence supporting these guidelines is scarce and is based on data from studies with less potent NAs.8, 10 It is thus unclear whether treatment adaptation is necessary for naïve patients treated with the more potent drug ETV. Therefore, the aims of this cohort study were to (1) investigate the efficacy of ETV in clinical practice beyond 1 year for NA-naïve and NA-experienced chronic hepatitis B patients, (2) explore baseline factors associated with a PVR to ETV in NA-naïve patients, and (3) investigate whether a PVR compromises long-term ETV treatment success.
In this investigator-initiated cohort study within the European Network of Excellence for Vigilance against Viral Resistance (VIRGIL), all consecutive adult CHB patients treated with ETV monotherapy between 2005 and May 2010 in 10 large European referral centers were included. Further eligibility criteria were: a viral load of at least 2,000 IU/mL at the initiation of ETV monotherapy, and duration of ETV monotherapy for at least 3 months. Patients were excluded if they had viral coinfections (human inmmunodeficiency virus, hepatitis C virus, hepatitis D virus) and if they had a liver transplantation before start of ETV therapy. All 220 patients from the published study cohort were included in this study, and 198 new patients were enrolled.6 In total, 418 chronic HBV patients treated with ETV monotherapy were identified. Eighty-five patients did not fulfill the entry criteria and were excluded from analysis: 33 patients had been treated with ETV monotherapy for less than 3 months, 58 patients had a baseline HBV DNA of less than 2,000 IU/mL, seven patients were coinfected with HCV, one patient was coinfected with HDV, and one patient underwent liver transplantation before start of treatment. A total of 333 patients were thus eligible for this analysis. Twenty (6%) of these patients were lost to follow-up. The study was conducted in accordance with the guidelines of the Declaration of Helsinki and the principles of Good Clinical Practice. Patients gave written informed consent according to standards of the local ethics committees.
Follow-up of Participants.
All patients were prospectively monitored every 3 months. At every visit, routine biochemical (alanine aminotransferase [ALT], bilirubin, albumin) and virological (HBV DNA level, hepatitis B e antigen [HBeAg], antibody against HBeAg) examinations were performed. Genotypic analysis was conducted (1) at baseline in all NA-experienced HBV patients; (2) in case of virological breakthrough, defined as an increase in serum HBV DNA level >1 log10 (10-fold) above nadir on at least two occasions after initial virological response; or (3) in case of serum HBV DNA >200 IU/mL at the end of follow-up. If ETV-resistant mutations were detected during follow-up, genotypic analysis was performed at baseline in NA-naïve patients. In NA-experienced patients, genotypic resistance was also assessed in stored serum samples obtained at the end of all previous NA-treatment regimes. HBV genotype was determined at start of ETV therapy. The diagnosis of cirrhosis was based on histology or ultrasound examinations.
The primary outcome was virological response (VR), defined as serum HBV DNA levels <80 IU/mL (≈400 copies/mL) during the on-treatment follow-up period. Secondary endpoints were HBeAg loss and seroconversion (in HBeAg-positive patients), hepatitis B surface antigen (HBsAg) loss and seroconversion, emergence of ETV-related mutations, and ALT normalization. Renal function was assessed by calculation of the estimated glomerular filtration rate in mL/minute/1.73 m2 using the modification of diet in renal disease equation, based on serum creatinine level, age, sex, and race.
Serum ALT, bilirubin, and albumin levels and international ratio of prothrombin time were measured locally using automated techniques. HBsAg, antibody against HBsAg, hepatitis B e antigen (HBeAg), and antibody against HBeAg were determined using commercially available enzyme immunoassays in all centers. Serum HBV DNA levels were measured using a quantitative real-time polymerase chain reaction assay, the COBAS AmpliPrep-COBAS TaqMan HBV test (CAP-CTM; Roche Molecular Systems, Inc., Branchburg, NJ), with a lower limit of detection of 12 IU/mL, in nine of 10 centers. In one center, serum HBV DNA was measured using Roche Amplicor (linear dynamic range, 400 to 200,000 copies/mL; Roche Diagnostic Systems, Branchburg, NJ). A conversion factor of 5.26 copies/IU was used for conversion of copies/mL to IU/mL. HBV genotypes and detection of HBV polymerase gene mutations was determined by direct sequencing or using the INNO-LiPA assay (Innogenetics, Gent, Belgium).
HBV DNA levels were logarithmically transformed for analysis. ALT levels are expressed as values representing a ratio to the local upper limit of normal. Continuous variables were expressed as the mean ± SD or median (interquartile range) where appropriate. Follow-up times were calculated from the date of ETV treatment initiation to the date of event or censorship. The cumulative probability of achieving virological response was estimated by way of Kaplan-Meier analysis. Cox regression analysis was used to study which of the following baseline factors were associated with virological response to ETV monotherapy: Age, sex, race, body mass index, HBV genotype, HBeAg status, viral load, ALT level, presence of cirrhosis, prior treatment with LAM, history of LAM resistance, presence of LAM resistance at baseline, duration of LAM therapy, prior treatment with ADV, history of ADV resistance, prior treatment with pegylated interferon, ETV dosage, and treatment center. Factors that correlated strongly (i.e., presence of colinearity) were compared in separate models with each collinear variable using the Akaike information criterion method. A Cox model was used to estimate the influence of prior treatment with LAM and prior treatment with ADV on the virological response to ETV, adjusted for the confounding effects of HBeAg status, viral load, and prior treatment with LAM. The covariate ETV dosage was not included, because it was not associated with virological response in the univariate proportional hazards analysis and, when included in the model, did not improve model fit. All statistical tests were two-sided, and P < 0.05 was considered statistically significant. SPSS version 15.0 was used for all statistical analysis (SPSS Inc., Chicago, IL).
Baseline characteristics of the study population are shown in Table 1. One hundred forty-three (43%) patients were HBeAg-positive, the median ALT level was 1.7 (1.0-3.3) times the upper limit of normal, and the mean HBV DNA was 6.2 ± 1.7 log IU/mL at baseline. NA-experienced patients were more often HBeAg-positive (P < 0.001) had more often cirrhosis (P = 0.04), had a higher MELD score (P = 0.02) and were more often treated with 1 mg ETV (P < 0.001) compared with NA-naïve patients. In addition, 87% of LAM-resistant patients and 86% of ADV-resistant patients were treated with 1 mg. Overall median follow-up was 20 months (interquartile range, 11-32 months; range, 3-51 months).
Table 1. Baseline Characteristics of the Study Population
All Patients (N = 333)
NA-Naïve Patients (n = 243)
NA-Experienced Patients (n = 90)
Abbreviations: MELD, model of end-stage liver disease; PEG-IFN, pegylated interferon; ULN, upper limit of normal.
43 ± 14
43 ± 14
41 ± 13
Body mass index
25 ± 4.1
25 ± 3.5
25 ± 5.0
ALT (×ULN) (interquartile range)
HBV DNA (log10 IU/mL)
6.2 ± 1.7
6.2 ± 1.7
6.2 ± 1.8
Genotype (n = 265)
ETV dosage (0.5 mg)
Presence of cirrhosis
7.8 ± 2.1
7.5 ± 1.9
8.4 ± 2.4
Previous treatment with PEG-IFN
Previous treatment with LAM
Prior history of LAM resistance
LAM resistance at baseline
Previous treatment with ADV
Prior history of ADV resistance
ADV resistance at baseline
Previous treatment with TDF
Previous treatment with LdT
Efficacy of Entecavir in NA-Naïve Patients.
In total, 243 (73%) patients were NA-naïve and treated for a median of 19 months (interquartile range, 11-32 months; range, 3-45 months) (Table 2). For HBeAg-positive patients (n = 86), the cumulative probability of achieving VR at week 48, 96, and 144 was 48% (95% confidence interval [CI], 36-60), 76% (66-86), and 90% (81-99), respectively (Fig. 1). HBeAg loss rates were 10% at week 48, 21% at week 96, and 34% at week 144. Corresponding rates for HBeAg seroconversion were 8%, 16%, and 24%. ETV therapy was not stopped in any patient achieving HBeAg seroconversion. HBsAg loss occurred in one (1%) of the HBeAg-positive patients. For HBeAg-negative patients (n = 157), the cumulative probability of achieving VR at week 48, 96, and 144 was 89% (95% CI, 84-93), 98% (95% CI, 95-100), and 99% (95% CI, 97-100), respectively (Fig. 1). HBsAg loss occurred in two (1%) HBeAg-negative patients. There were no significant differences in virological response rates per center. Five patients experienced a virological breakthrough, but no genotypic resistance to ETV was detected.
Table 2. Virological and Biochemical Response to ETV, Number of Observed Events
Data are presented as mean ± SD or n (%) unless stated othewise.
The antiviral effect of ETV is described for 43 patients who were directly switched to ETV.
Baseline HBV DNA, log10 IU/mL
6.2 ± 1.7
6.4 ± 1.8
6.3 ± 1.5
5.6 ± 1.8
5.9 ± 1.8
5.4 ± 1.8
Median follow-up, months (interquartile range)
Partial Virological Response in NA-Naïve Patients.
PVR at week 48 occurred in 36 (21%) of 175 NA-naïve patients with at least 48 weeks of follow-up. High baseline HBV DNA (odds ratio, 0.67; 95% CI, 0.50-0.89; P = 0.005) and HBeAg positivity (odds ratio, 0.25; 95% CI, 0.10-0.60; P = 0.002) were the only independent risk factors for having a PVR. Overall follow-up of patients with a PVR lasted for 27 months (IQR, 19-35 months). Twenty-nine (81%) of 36 patients achieved a VR beyond week 48 (Table 3). Moreover, 10 patients needed more than 96 weeks of continuous ETV therapy to achieve a VR. Patients achieving a VR had a lower HBV DNA at week 48 than those who did not achieve a VR during prolonged ETV monotherapy. Patients with a PVR were stratified according to their viral load at week 48 (Fig. 2). Twenty-one (95%) of 22 patients with HBV DNA <1,000 IU/mL and 8 (57%) of 14 patients with HBV DNA ≥1,000 IU/mL achieved a VR without treatment adaptation during prolonged treatment beyond week 48. Overall, a continuous HBV DNA decline was observed in six of seven (86%) patients without VR at the end of follow-up. In three (43%) patients, noncompliance was suspected by the local physician. Five patients were switched to a TDF-containing regimen (due to virological breakthrough in two patients, one of these two patients first achieved VR); all patients achieved a VR during follow-up. In two patients, ETV dosage was changed from 0.5 mg to 1 mg, both of whom achieved a VR during follow-up. However, no ETV resistance was detected during follow-up, neither at virological breakthrough nor at the time of treatment adaptation.
Table 3. NA-Naïve Patients with a PVR at Week 48
All Patients (n = 36)
Patients with PVR (n = 29)
Patients with No PVR (n = 7)
Characteristics are shown for 36 patients with a detectable HBV DNA after 48 weeks of continuous ETV monotherapy. Data are presented as mean ± SD or n (%).
41 ± 14
43 ± 14
33 ± 13
Presence of cirrhosis
Follow-up, months (interquartile range)
Baseline HBV DNA (log10 IU/mL)
7.7 ± 1.4
7.5 ± 1.5
8.2 ± 1.1
HBV DNA week 48 (log10 IU/mL)
2.7 ± 1.3
2.3 ± 1.2
3.8 ± 0.8
HBV DNA decline at week 48 (log10 IU/mL)
5.0 ± 1.3
5.1 ± 1.3
4.3 ± 0.8
HBV DNA last visit (log10 IU/mL)
3.0 ± 0.9
Load >1,000 IU/mL at week 48
Efficacy of ETV in NA-Experienced Patients.
Fifty-one (57%) NA-experienced patients had received prior treatment with ADV. To investigate the efficacy of ETV as salvage therapy for ADV-treated patients, the antiviral effect of ETV is given in Table 2 for 43 (84%) patients who were directly switched to ETV monotherapy. Twelve (28%) of these patients had a history of genotypic resistance to ADV (rtN236T/ rtA181V/T). Seventy-two (80%) NA-experienced patients had received prior treatment with LAM. Adjusted for baseline viral load and HBeAg status, antiviral response to ETV was influenced neither by prior ADV therapy (hazard ratio [HR], 0.92; 95% CI, 0.57-1.51; P = 0.75) nor by previous ADV resistance (HR, 1.23; 95% CI, 0.56-2.70; P = 0.61) (Fig. 3). In contrast, presence of LAM-resistant mutations at baseline (HR, 0.13; 95% CI, 0.04-0.42; P < 0.001) and a previous history of LAM resistance (HR, 0.40; 95% CI, 0.19-0.84; P = 0.015) were significantly associated with a reduced probability of achieving VR.
Eighteen patients experienced a virological breakthrough after a median follow-up of 20 months (range, 11-32 months). Five (2%) NA-naïve patients experienced a virological breakthrough; nonadherence was suspected in two of these patients, and three patients were switched to a TDF-containing regimen (TDF monotherapy in one, TDF add-on in one, and TDF plus emtricitabine in one). Six NA-experienced patients were switched to a TDF-containing regimen (TDF monotherapy in two, TDF add-on in three, and TDF plus emtricitabine in one), and one patient stopped ETV treatment without beginning another NA treatment. Genotypic mutations to ETV were detected in four NA-experienced patients; one of these patients was previously exposed only to ADV and achieved a VR before developing a virological breakthrough (Table 4). A TDF-containing treatment regimen was initiated in all four patients, and a subsequent decline in HBV DNA was observed in three of them. No mutations associated with decreased sensitivity to ETV were observed in any of the NA-naïve patients, including those with a viral load >200 IU/mL at the end of follow-up.
Table 4. Characteristics of Four Patients Developing Genotypic Mutations to ETV
At start of ETV therapy
HBV DNA (log10 IU/mL)
Prior LAM exposure
Prior LAM resistance
Baseline LAM resistance
Viral load at maximum viral suppression, log10 IU/mL
Month of resistance
At time of resistance
HBV DNA, log10 IU/mL
rtM204V rtL180M rtI169T L217R
L180M M204V V173L A181T N236T
rtM204V rtL180M V173L
rtM204V L180M T184I
Response to salvage therapy
Addition of tenofovir
Addition of tenofovir
HBV DNA at last follow-up, log10 IU/mL
Adverse events included dizziness, headache, and loss of appetite (in three different patients). None of the patients developed clinically evident lactic acidosis, but lactate was not routinely measured. One patient died from hepatocellular carcinoma (which was already present at the start of ETV treatment), one patient had a recurrence of hepatocellular carcinoma, and one patient died from non-Hodgkin lymphoma. To assess renal safety, we analyzed creatinine levels in a subset of 188 patients with an available baseline creatinine level. The mean estimated glomerular filtration rate (eGFR) at baseline was 95.8 mL/minute/1.73m2. The mean decrease in eGFR during follow-up was 1.2±18.1 mL/minute/1.73m2 (P = 0.38). None of the patients experienced an increase in serum creatinine > 0.5 mg/dL. In a subset of 9% of patients with an eGFR <70 mL/minute/1.73 m2, eGFR increased (1.4 ± 10.0 mL/minute/1.73 m2; P = 0.60). Age was the only risk factor significantly associated with developing an eGFR <60 mL/minute/1.73 m2.
The current multicenter study showed that ETV is effective up to 3 years in NA-naïve patients, irrespective of having a virological response at week 48. The vast majority of NA-naïve patients with a PVR achieved undetectable HBV DNA through prolonged therapy without treatment adaptation. Genotypic resistance to ETV was not detected in any of the patients with a PVR at week 48. We showed that ETV is a safe antiviral drug with a good renal tolerance and minimal adverse effects. Persistent viral replication and the development of resistance during treatment with NAs for CHB have been associated with adverse treatment outcome.8, 11 Therefore, European Association for the Study of the Liver guidelines suggest treatment adaptation in patients with a PVR to prevent treatment failure and the development of resistance.9 PVR is defined as a decline of >1 log IU/mL in HBV DNA but failure to achieve undetectable HBV DNA levels at week 48 in patients treated with continuous ETV monotherapy.
In our cohort of NA-naïve patients treated with ETV, 36 patients failed to achieve a VR at week 48. Among these patients with a PVR, 81% achieved a VR without treatment adaptation through 15 additional months of therapy. Cumulated probability of achieving a VR beyond week 48 was higher for patients with HBV DNA <1,000 IU/mL at week 48. Importantly, despite two patients experiencing a virological breakthrough, no resistance was detected in these NA-naïve patients. This is in accordance with the ETV phase III trial in which, albeit with incomplete follow-up, a substantial number of patients achieved a response beyond the first year of treatment, whereas genotypic resistance remained rare through 5 years of continuous monotherapy.4, 12, 13 Our findings are in contrast with previous studies on LdT/LAM and ADV in which persistent viral replication at week 24 and week 48 of therapy was identified as a predictor of the emergence of subsequent viral resistance.8, 10 This highlights that treatment paradigms based on data from studies investigating agents with a low barrier to resistance cannot be translated to newer and more potent drugs as ETV and TDF.
Nevertheless, not all ETV-treated patients with a PVR achieved a VR through prolonged treatment. After thorough examination, we determined noncompliance in three of these seven (43%) patients, which we believe explains the inability to achieve HBV DNA undetectability. The problem of nonadherence is supported by a previous study that suggested partial response to ADV is most likely due to noncompliance and host pharmacological factors.14 One of seven patients without a VR experienced a virological breakthrough, and treatment was adapted by the treating physician. However, it is important to note that six of seven (86%) patients who failed to achieve a VR during follow-up still had a declining load at the end of follow-up, which suggests that achieving a VR can probably be reached in the majority of cases. Patients with a PVR could therefore be considered slow responders instead of partial responders. Our study shows that continuing ETV treatment appears to be safe and effective in patients with detectable HBV DNA at week 48, especially in patients with a lower viral load at week 48, of whom 95% achieved a VR.
Decreased sensitivity to ETV for LAM-refractory patients was soon known after introduction of this agent.5, 15 Our study confirms these results, because the antiviral efficacy of ETV was seriously diminished in these patients, even after correction for possible confounders such as high baseline HBV DNA and HBeAg positivity. Moreover, our study underlines that even after resistance testing at baseline; the absence of LAM-associated mutations does not guarantee a susceptible virus during ETV treatment. This suggests that if there is a suspected history of LAM-resistance, TDF-containing regimens should be used instead of ETV monotherapy, because LAM-resistant strains remain susceptible to TDF monotherapy.16
Consistent with in vitro data, our study shows that antiviral efficacy of ETV treatment is not influenced by prior exposure or resistance to ADV.17-21 Until now, only small studies or studies with a relatively short follow-up have confirmed the in vitro efficacy of ETV in ADV-experienced or ADV-resistant patients in real-life practice.6, 22-24 Our findings are of particular interest, because both ETV and TDF can thus be used as salvage therapy for ADV-experienced patients.9, 25
Data from the large phase III trials with a selected population showed that entecavir has few adverse effects in patients with compensated liver disease.2, 3 However, a recent report indicates that patients with decompensated cirrhosis are at risk for developing lactic acidosis.26 We show that ETV is safe during prolonged therapy in this heterogeneous cohort, even in the presence of cirrhosis. Moreover, we show that ETV does not affect renal function, which might be a concern during TDF therapy.27
Limitations of our study include the observational design and the heterogeneous group of patients; however, we used Cox regression to correct for confounders as treatment duration, HBV DNA, HBeAg status, and previous LAM resistance. Nevertheless, this heterogeneous population is also representative for clinical practice and makes it possible to compare different groups of (NA-experienced) patients within one study.
In conclusion, in contrast to what is suggested in recently published European Association for the Study of the Liver guidelines on the management of chronic hepatitis B, adjustment of ETV monotherapy in NA-naïve patients with a PVR at week 48 is not necessary. We show that continuous therapy beyond week 48 is safe and effective and results in VR in the vast majority of patients, particularly those with HBV DNA <1,000 IU/mL at week 48. Furthermore, genotypic resistance to ETV was not observed in this subset of NA-naïve patients. For both NA-naïve and NA-experienced patients, ETV proved to have a favorable safety profile.
We thank M. Fasano, E. Verhey, I. Brings, and C. Boucher for assistance during data collection, study preparation, and retesting of samples.