Aliment Pharmacol Ther 2011; 34: 767–774
Background An increasing number of patients with chronic hepatitis B (CHB) have experienced treatment failure to adefovir (ADV) and their management poses a growing challenge. Very limited data are available on the efficacy of entecavir (ETV) in patients previously treated with ADV.
Aim To examine the effect of ETV monotherapy on HBV DNA and ALT levels in CHB patients previously treated with ADV, but switched to ETV due to suboptimal response.
Methods Study candidates were enrolled from five community gastroenterology clinics in the U.S. Each completed at least 12 months of ETV treatment after being previously treated with ADV and experiencing suboptimal response. Primary and secondary outcome measurements were complete viral suppression (CVS, HBV DNA <100 IU/mL) and biochemical response (BR, ALT <40 U/L), respectively.
Results A total of 60 patients were included in this analysis. Twelve were lamivudine (LAM)-experienced and none were LAM-resistant. At time of switch to ETV, no patients had experienced CVS. The CVS rate was 68% after 12 months of ETV therapy. The BR rate was 67% at switch to ETV and 80% after 12 months. There was no significant difference in response rates between LAM-experienced and naïve patients. Among the eight patients with ADV resistance, each achieved CVS after 12 months of ETV therapy and seven achieved BR.
Conclusions In patients with suboptimal response to adefovir, complete viral suppression and biochemical response can be achieved in the majority by 12 months after switching to entecavir, including patients with prior exposure to lamivudine and those with adefovir resistance.
One of the primary objectives of anti-viral therapy for chronic hepatitis B (CHB) is complete suppression of viral replication.1–3 The importance of suppressing serum hepatitis B virus (HBV) DNA levels has been extensively studied. The degree of HBV viremia correlates with reduced risk for disease progression, cirrhosis and hepatocellular carcinoma.4–6 Rapid complete virological response after starting nucleos(t)ide analogue therapy is further associated with a lower rate of resistance development to anti-viral drugs over the long-term.7, 8
Adefovir (ADV) is the second oral drug approved for treatment of chronic hepatitis B. It is an acyclic phosphonate nucleotide analogue of adenosine monophosphate. Originally developed for the treatment of HIV, the FDA decided in 1999 against awarding approval for this indication due to concerns about renal toxicity. ADV was approved for the treatment of HBV in 2002 after demonstrating effectiveness at a much lower dose. Studies have demonstrated that after 48 weeks of treatment, ADV yields virological, biochemical and histological improvement in a significant number of both HBeAg positive and HBeAg negative HBV patients.9, 10 ADV has demonstrated limited resistance during the first 1–2 years of treatment. It has also been a valuable alternative for lamivudine (LAM)-resistant HBV infection with its absence of cross-resistance with L-nucleosides.11, 12 The durability of HBeAg seroconversion is similar to that of other oral anti-virals.6 However, ADV has significant limitations. It has a high rate of primary nonresponse, defined as failure to achieve a 1–log10 reduction in viral load after 12 weeks of therapy 13–15 possibly because of the low dosage needed to avoid renal toxicity.2 In addition, though they may appear to be low at first, resistance rates can approach up to 30% after five years of treatment 6, 16, 17
Entecavir (ETV) is a selective cyclopentyl guanosine analogue with potent activity against HBV. It is potent because of its unique mechanism of inhibiting the three functions of the HBV DNA polymerase: priming, reverse transcription and synthesis of positive-strand HBV DNA.13 ETV also has a long intracellular half-life enabling accumulation of intracellular entecavir triphosphate.14 In two large phase III double-blind trials, ETV demonstrated higher rates of histological, virological and biochemical improvement than LAM after 48 weeks of treatment in both nucleoside-naïve HBeAg positive and HBeAg negative patients.14, 15 The high potency of ETV is complemented by a very low resistance rate (≤1%) after four years of treatment. This may be due to rapid and sustained suppression of serum HBV DNA and high genetic barrier to resistance requiring three or more substitutions.16, 17 ETV also has an excellent safety profile, equivalent to that of LAM, which has a similar safety profile as placebo in clinical trials.14, 15
Due to the low potency of ADV, a significant population of patients treated with ADV have only experienced suboptimal response with incomplete viral suppression, which predisposes them to the consequences of unsuppressed HBV viral load and future anti-viral resistance. Alternative therapy with a more potent agent such as ETV or combination therapy is recommended in this setting.1–3 There are very limited data available on the treatment efficacy of ETV in patients previously treated with ADV. This study examines the effect of ETV monotherapy on HBV DNA and ALT levels in CHB patients previously treated with ADV, but switched to ETV due to suboptimal response to ADV.
Patients and Methods
A total of 103 patients (Figure 1) from five community gastroenterology clinics, each previously treated with ADV and then switched to ETV, were screened for study inclusion. Each patient initiated ETV therapy after being previously treated with ADV for at least 12 weeks and experiencing suboptimal response, defined as achieving <2–log10 reduction in HBV DNA at six months or having detectable HBV DNA at 12 months. Each patient had pre-treatment HBV DNA >10 000 copies/mL or 2000 IU/mL. Each was 18 years of age or older. Patients were excluded if they were co-infected with HCV, HDV, or HIV, transplant recipients, receiving immunosuppressive therapy, LAM-resistant, or previously exposed to another nucleoside for more than two weeks and there was no genotypic mutation analysis performed to rule out viral resistance.
A total of 86 patients met the inclusion criteria. Of these 86 patients, 60 patients had completed at least 12 months of ETV therapy and included in the analysis. Twelve of these 60 patients were LAM-experienced and none were LAM-resistant by genotypic mutation analysis (Hepatitis B Virus Drug Resistance, Genotype, and BCP/Precore Mutations Test, test code 10529x, Quest Diagnostics, San Juan Capistrano, CA, USA).
Patients who did not complete up to 12 months of ETV had similar laboratory characteristics as those who completed 12 or more months of therapy. Prior to treatment with ADV, both groups had similar median ALT [31 (16–162) vs. 51 (16–356) U/L, P = 0.35] and HBV DNA levels [5.7 (4.1–8.7) vs. 6.0 (2.3–9.4) log10 IU/mL, P = 0.50]. Prior to treatment with ETV, both groups had similar median ALT [25 (15–164) vs. 35 (11–121) U/L, P = 0.43] and HBV DNA levels [4.9 (1.5–7.1]) vs. 4.0 (2.1–9.0) log10 IU/mL, P = 0.71]. At switch to ETV, both groups had a similar proportion of men (50% vs. 69%, P = 0.17) and HBeAg-positive patients (50% vs. 52%, P = 0.91), as well as similar median age [50 (30–81) vs. 45 (18–73) years, P = 0.19].
The primary outcome measure was complete viral suppression (CVS), defined as HBV DNA <100 IU/mL. The secondary outcome measure was biochemical response (BR), defined as ALT normalisation (ALT < 40 U/L). All patients provided written consent and the study was approved by the Western Institutional Review Board (WIRB, Olympia, WA, USA).
The following data were recorded for each patient at baseline: CBC, liver function tests, creatinine, HBeAg, HBsAg, anti-HBe and HBV viral load. Laboratory tests were performed at local laboratories operated by either Quest Diagnostics (San Juan Capistrano, CA, USA) or Lab Corp (Burlington, NC, USA). Serum HBV DNA levels were determined using HBV DNA quantitative polymerase chain reaction (PCR). Resistance status and genotype were determined using the HBV genotypic mutation panel (Hepatitis B Virus Drug Resistance, Genotype, and BCP/Precore Mutations Tests) performed by Quest Diagnostics (San Juan Capistrano, CA, USA).
All except 11 patients were treated with ETV 1.0 mg daily. Patients were evaluated approximately every 3–6 months. At every visit virological, haematological and biochemical data were recorded including HBV viral load and liver function tests. Adverse events were recorded and categorised according to modifications of the World Health Organization grading system. All patients completed at least 12 months of ETV treatment after suboptimal response to ADV.
We planned to have a total sample size of approximately 80 patients. At n = 80, assuming a two-sided alpha of 0.05, the power of our proposed comparison between a hypothesised proportion of 45% and alternative proportions of 70%, 65% and 60%, were 99.7%, 95.6% and 77.3%, respectively.
Descriptive statistics were used to summarise all variables. Categorical variables were expressed as proportions while continuous variables were expressed as medians or means. HBV DNA levels were logarithmically transformed. The Pearson’s Chi-square test was used to compare categorical variables and the t-test was used to compare continuous variables. The primary outcome was complete viral suppression at 12 months. Cox proportional hazard regression models were used to estimate univariate and multivariate hazard ratios (HR) and associated confidence intervals relating potential predictors to development of CVS at month 12. A two-tailed P-value < 0.05 was considered statistically significant. All analyses were performed using stata version 10.0 (Stata Corporation, College Station, TX, USA).
The baseline characteristics of the 60 CHB patients included in this study are reported in Table 1. All patients were Asians. The majority were men and middle-aged with mean BMI in the upper range of normal. The median duration of prior ADV treatment was 70 (15–394) weeks. Almost half of all patients were HBeAg positive before ADV treatment. Four patients developed genotypic resistance to ADV during ADV therapy and four patients demonstrated phenotypic resistance, defined as experiencing an increase of >1 log10 HBV DNA from nadir on at least two consecutive measurements. By the time of switch to ETV treatment, slightly more than half of all patients were HBeAg positive.
|All patients (n = 60)||LAM-naïve (n = 48)||LAM-experienced (n = 12)|
|Age (years)||44 ± 12||44 ± 13||42 ± 8|
|Gender (male)||42 (70)||33 (69)||9 (75)|
|BMI (kg/m2)||24.7 ± 2.7||24.5 ± 2.4||25.6 ± 3.6|
|Positive HBeAg||29 (48)||20 (42)||9 (75)|
|ALT (U/L)||51 (16–356)||53 (16–356)||51 (16–314)|
|HBV DNA (log10 IU/mL)||6.0 (2.3–9.4)||5.8 (3.4–9.4)||7.2 (2.3–9.3)|
|Adefovir treatment duration (weeks)||70 (15–394)||67 (15–394)||74 (28–198)|
Treatment Response to Entecavir
Outcomes of ETV therapy are summarised for all patients in Figures 2a, 2b, 3a and 3b. At the time of switch to ETV, no patient had experienced CVS. The median decline in HBV viral load (DNA log10 IU/mL) after 12 months of ETV therapy was 3.2 (−0.67–9.0). The complete viral suppression rate after 12 months of therapy with ETV was 68% (55–80). At the time of the switch to ETV, the BR rate was 67% (53–78) and this increased to 80% (68–89) after 12 months of therapy with ETV. No patient experienced any significant adverse events during this study.
Responses to ETV therapy within LAM-naïve and experienced subgroups are summarised in Table 2. The median ALT and HBV DNA levels before ADV treatment and at time of switch to ETV are presented. There was no significant difference between LAM-naïve and LAM-experienced patients across these characteristics. Complete viral suppression and biochemical response rates were also similar between LAM-naive and LAM-experienced patients.
|LAM-naïve n = 48||LAM-experienced n = 12||P-value|
|Pre ADV||53 (16–356)||51 (16–314)||0.51|
|At switch||35 (11–121)||33 (13–93)||0.83|
|Month 12||29 (11–146)||27 (10–41)||0.19|
|At switch||31 (65)||9 (75)||0.77|
|Month 12||37 (77)||11 (92)||0.26|
|HBV DNA (log10 IU/mL)|
|Pre ADV||5.9 (3.4–9.4)||7.2 (2.3–9.3)||0.24|
|At switch||3.8 (2.1–9.0)||5.1 (2.2–8.7)||0.27|
|Month 12||0 (0.0–6.2)||0 (0.0–4.4)||0.25|
|HBV DNA decline (log10 IU/mL)|
|At switch||2.0 (−3.4–4.9)||1.6 (−1.2–5.8)||0.99|
|Month 12||3.2(−0.7–9.0)||3.2 (0.3–8.7)||0.99|
|Complete viral suppression|
|At switch||0(0)||0 (0)||–|
|Month 12||34 (71)||7 (58)||0.41|
Factors Associated with Complete Viral Suppression
On univariate analysis (Table 3), HBeAg positivity before ADV therapy and higher HBV DNA levels at time of switch to ETV were associated with lower odds of achieving CVS. There was a trend for higher ALT levels at time of switch to ETV to be associated with a lower probability of achieving CVS while there was a trend for older age being associated with a higher probability of CVS.
|HR (95% CI)||P||HR (95% CI)||P|
|Age||1.02 (0.99–1.04)||0.13||1.01 (0.98–1.04)||0.58|
|Male gender||0.73 (0.38–1.37)||0.33||1.27 (0.56–2.88)||0.57|
|Positive HBeAg (before ADV)||0.38 (0.19–0.75)||0.005||0.55 (0.21–1.43)||0.22|
|ALT at switch||0.99 (0.97–1.00)||0.14||0.99 (0.98–1.01)||0.51|
|Viral load at switch||0.69 (0.55–0.87)||0.002||0.77 (0.58–1.03)||0.08|
|Prior lamivudine exposure||0.76 (0.34–1.72)||0.51||0.87 (0.32–2.42)||0.80|
Upon multivariate analysis (Table 3), none of these variables remained associated with CVS. There was a trend for an association between higher HBV DNA levels at time of switch to ETV and lower probability of achieving CVS at 12 months following ETV therapy. Prior LAM experience was not a negative predictor for CVS under ETV monotherapy in patients with suboptimal response to prior ADV therapy.
Treatment Response to Entecavir in Patients with Adefovir Resistance
There were a total of eight patients with ADV resistance in this study (increase of >1 log10 HBV DNA from nadir on at least two consecutive labs while on ADV treatment after initial virological response). ADV resistance was also confirmed by genotypic mutation analysis in five of these patients. Each of these eight patients experienced CVS after 12 months of ETV therapy and seven achieved BR.
An increasing number of CHB patients have experienced treatment failure to ADV and their management poses a growing challenge. Treatment failure leads to resurgence of viraemia, selection for resistant HBV strains and increased risk for disease progression, cirrhosis and hepatocellular carcinoma. This study demonstrates that in patients with only suboptimal response to ADV, CVS can be achieved in the majority (68%) by 12 months after switching to ETV. There was no significant difference in CVS or BR rates between LAM-experienced and naïve patients and this was confirmed by multivariate regression analysis. Each of the eight patients in this study with resistance to ADV also achieved CVS by month 12.
Very limited data are available on the efficacy of ETV therapy in HBV patients who have experienced suboptimal response to ADV. There are however, some data on the efficacy of combination therapies. Berg et al. compared the efficacy and safety of tenofovir (TDF) monotherapy against Truvada® (fixed dose combination of TDF/emtricitabine, Gilead Sciences, Foster City, CA, USA) in CHB patients with suboptimal response to ADV. After 96 weeks, ADV suboptimal responders switched to TDF monotherapy also achieved a complete viral suppression rate of 68%. Those switched to Truvada® realised complete viral suppression rates of 75% at 48 and 96 weeks. The treatment outcomes we observed after 12 months of ETV therapy were comparable to these results and suggest that for the treatment of some ADV-experienced patients, ETV monotherapy alone may be sufficient.
Van Bommel et al. conducted a large-scale, retrospective study in the long-term efficacy of TDF monotherapy in HBV patients who had experienced either prior failure or genotypic resistance to ADV (n = 8), LAM (n = 18), sequential LAM-ADV therapy (n = 73), or add-on combination therapy with both drugs (n = 29). Among the 113 patients studied, 21 (19%) had genotypic resistance against ADV while 70 (62%) had genotypic resistance against LAM.18 The efficacy of TDF was not influenced by the presence of mutations associated with LAM resistance which is consistent with in vitro studies reporting that LAM-resistant HBV strains remain fully sensitive to TDF.19, 20 However, only 52% of patients with ADV resistance treated with TDF achieved CVS during the observation period. Known cross-resistance between ADV and TDF may have accounted for these observations.21, 22 Due to the poor efficacy of TDF in this regard, combination therapy with TDF and a second nucleos(t)ide agent without cross-resistance is recommended for the treatment of ADV resistance.23–25 The findings of the current study suggest that a simpler approach may be feasible. Each of the eight patients with resistance to ADV achieved CVS after 12 months of ETV therapy. In vitro studies have also demonstrated that ETV remains effective against ADV-resistant HBVstrains.22, 26–29
As noted earlier, ETV was effective in achieving high viral suppression rates in both LAM-naïve and experienced patients in the current study. There was no significant difference in CVS rates between the two groups, likely because no LAM-experienced patients had any history of LAM resistance. LAM resistance is in fact associated with cross-resistance to ETV. Development of ETV resistance requires multiple substitutions in HBV reverse transcriptase and two of these are specifically associated with LAM resistance (L180M and M204V substitutions).23, 24, 30, 31 Resistance to ETV is therefore rare in LAM-experienced patients unless they have one or more pre-existing mutations associated with LAM resistance.
To date, only two published studies have specifically investigated the efficacy of ETV therapy in ADV suboptimal responders. Reijinders et al. reported the effects of ETV rescue therapy in a small cohort of 14 patients with limited virological response to ADV.8 ETV therapy yielded a limited viral load decline in both LAM-experienced (n = 8) and naive (n = 6) patients with only one out of 14 patients achieving HBV DNA levels <373 copies/mL over a median follow-up duration of 15 months. The poor virological response was attributed to possible underdosing and host specific factors. In a larger follow-up study by the same group, responses to ETV therapy were similar to those observed in the current study25: ETV was effective in the treatment of ADV-experienced patients including those with ADV resistance and there was no difference in the efficacy of ETV between LAM-naïve and experienced patients without history of LAM resistance.
A limitation of our study was the observational design. Subgroup analysis was also limited by subgroup sizes because sample size calculation for the study was not planned according to these analyses. Yet, this study cohort is the largest to date of any published study specifically examining the effects of ETV salvage therapy on ADV suboptimal responders. The patients in this cohort were also more homogenous than in similar studies, particularly with regard to the inclusion of only LAM-experienced patients without prior history of LAM resistance. Collectively, these findings are consistent with those from in vitro data and the very limited prior in vivo studies which support their reliability. Additional data will be acquired as treatment responses through year two of this study are monitored. Patients in this cohort were all Asian and results may not be fully generalisable to CHB patients of other ethnicities.
In conclusion, our observations suggest that the response of ADV suboptimal responders to ETV monotherapy can be comparable to those achieved after rescue with combination therapies. Patients with ADV resistance can also be effectively treated with ETV. In addition, ETV monotherapy can yield virological responses in LAM-experienced ADV suboptimal responders comparable to those achieved by LAM-naïve ADV suboptimal responders, though ETV should not be used to treat patients with prior history of LAM resistance. Combination regimens may provide additional treatment options, but this study suggests an effective yet simpler and less costly regimen. Pharmacoeconomic analyses would also help to clarify the position of ETV monotherapy relative to other salvage regimens given the importance of thoughtful resource allocation in the contemporary health care environment.32
Declaration of personal interests: Edward Sheen, Huy A. Nguyen, Khanh K. Nguyen and Phuong Tran have no financial relationship with a commercial interest. Huy N. Trinh is associated with Gilead Sciences Inc and Bristol-Myers Squibb. Thuan T Nguyen is associated with Gilead Sciences Inc. Son T. Do is associated with Bristol-Myers Squibb, Gilead Sciences Inc, Novartis Pharmaceuticals, Roche, Centocor Inc and Abbott. Ruel T. Garcia is associated with Gilead Sciences Inc. Mindie H. Nguyen is associated with Gilead Sciences Inc, Bristol-Myers Squibb, Novartis and Bayer. Declaration of funding interests: This study was supported by an investigator-initiated research grant from Bristol-Myers Squibb to the Pacific Health Foundation. The grantor did not participate in data collection, analysis, drafting, or decision to publish this manuscript.