fax: +852 2816 2863
Article first published online: 7 APR 2011
Copyright © 2011 American Association for the Study of Liver Diseases
Volume 53, Issue 4, pages 1148–1153, April 2011
How to Cite
Fung, J., Lai, C.-L., Yuen, J., Cheng, C., Wu, R., Ka-Ho Wong, D., Seto, W.-K., Hung, I. F.-N. and Yuen, M.-F. (2011), Randomized trial of lamivudine versus entecavir in entecavir-treated patients with undetectable hepatitis B virus DNA: Outcome at 2 Years. Hepatology, 53: 1148–1153. doi: 10.1002/hep.24192
Potential conflict of interest: Nothing to report.
Supply of lamivudine was kindly provided by GlaxoSmithKline (Hong Kong).
- Issue published online: 7 APR 2011
- Article first published online: 7 APR 2011
- Accepted manuscript online: 14 FEB 2011 12:11PM EST
- Manuscript Accepted: 6 JAN 2011
- Manuscript Received: 27 SEP 2010
- GlaxoSmithKline (Hong Kong)
We aimed to determine the 2-year outcomes of entecavir followed by lamivudine in patients with undetectable viral load (<12 IU/mL) and normal alanine aminotransferase (ALT) after initial entecavir treatment for at least 6 months. Patients were randomly assigned 1:1 to continue with entecavir or switch to lamivudine. Liver biochemistry and hepatitis B virus (HBV) DNA were determined at weeks 0, 4, 12, 24, 48, 72, and 96. Mutational analysis using line-probe assay were performed at weeks 0, 24, 48, and 96 and at the time of HBV DNA relapse. There was no elevation of ALT observed in any patients up to 96 weeks. At 96 weeks of follow-up, 19/25 (76%) patients in the lamivudine arm had persistently undetectable HBV DNA, compared with 25/25 (100%) patients in the entecavir arm. Six patients in the lamivudine arm had HBV DNA >20 IU/mL, occurring at a range of 12 to 96 weeks. Of these, four patients had HBV DNA of less than 100 IU/mL during rebound (three had undetectable HBV DNA after switching back to entecavir), and the remaining two patients had HBV DNA levels of 7,973 and 699 IU/mL. Three patients (12%) had evidence of drug-resistant mutations, of which two patients had rtM204I mutation and one patient had rtM204V mutation. One of these three patients had previous lamivudine exposure before entecavir treatment and one patient had questionable drug compliance. Conclusion: Sequential therapy using entecavir followed by lamivudine resulted in virological rebound in 24% of patients after 96 weeks. Prior optimal viral suppression with entecavir did not confer any significant advantage in patients who switched to lamivudine. (HEPATOLOGY 2011;)
Chronic hepatitis B (CHB) affects ≈400 million people worldwide, constituting an important health burden in areas where hepatitis B virus (HBV) infection is endemic. Up to 40% of CHB patients may develop cirrhosis and its complications, including hepatocellular carcinoma (HCC).1 Lamivudine (LAM) was the first oral nucleoside analog available for the treatment of CHB, and is effective in retarding disease progression, including liver decompensation and HCC in both patients with cirrhosis and without cirrhosis.2, 3 Unfortunately, long-term LAM therapy is associated with high rates of drug-resistance, with ≈76% after 8 years of treatment.3
Previous studies have identified early viral suppression to be a predictor of long-term outcomes including the development of drug resistance. In a study of 159 hepatitis B e-antigen (HBeAg)-positive CHB patients treated with LAM, viral suppression at 24 weeks was predictive of long-term resistance risks with a median follow-up of 29 months.4 In those patients who achieved undetectable HBV DNA by real-time polymerase chain reaction (PCR) assay at week 24, the long-term resistance risk was only 8% compared to 64% for those with HBV DNA of >4 log copies/mL. Similar findings have also been shown in other antiviral agents including adefovir (ADV) and telbivudine.5, 6 A recent study also showed that rapid viral suppression at week 4 with LAM therapy is associated with no drug resistance for up to 5 years of therapy.7 Therefore, achieving optimal early viral suppression is associated with a lower risk of drug resistance.
Over the past few years, more potent oral antiviral therapy has become available for the treatment of CHB. Entecavir (ETV), a carboxylic analog of guanosine, is found to be superior to LAM in both HBeAg-positive and HBeAg-negative patients in reducing HBV DNA and inducing histological improvements.8, 9 In addition, resistance to ETV is rare, with a rate of 1.2% after 5 years of ETV treatment.10 It is conceptually not known whether the initial profound HBV DNA suppression to undetectable levels by a potent antiviral agent such as ETV can be maintained by continuing treatment with a relatively less potent antiviral agent such as LAM. In addition, this kind of therapy has not been examined with regard to the chance of development of drug resistance. It is also not known whether resuming ETV in patients with virological rebound after this kind of switch therapy to LAM will continue to respond. All these questions are of clinical and economic significance given that most patients with CHB will require long-term therapy and treatment with an antiviral agent may be a cost concern in many countries. There are still patients who opt to start therapy with LAM because of cost-savings issues.
The implementation of switching therapy from ETV to LAM can only be recommended with two essential requirements. First, the viral suppression can be maintained after switching to LAM, and second, the risk of LAM-resistance after switching should be low. The aim of the present prospective study was to determine the efficacy and drug-resistance profile of ETV-LAM switching therapy in patients with undetectable viral load and normal alanine aminotransferase (ALT) after initial ETV treatment.
Patients and Methods
This was a prospective randomized trial comparing patients treated with ETV versus patients treated with ETV followed by LAM therapy. A total of 50 patients were recruited from the Hepatitis and Liver Clinic, Queen Mary Hospital, Hong Kong, between the period of September 2007 and July 2008. The sample size was calculated at 38 using a 2-sided significance level of 0.05, a desired power of 0.80, and an estimated ETV and LAM resistance rate at 2 years of 1% and 35%, respectively.
All patients were treated with ETV 0.5 mg for at least 6 months prior to study entry, with normal ALT and undetectable HBV DNA (<60 copies/mL). The upper limit of normal for ALT as defined by our laboratory ranges was 53 U/L for males and 31 U/L for females. Patients were then randomly assigned by the primary investigator at a ratio of 1:1 to two arms using computer-generated numbers preassigned to either arm. The first arm (ETV-ETV) continued on ETV 0.5 mg daily, whereas in the second arm (ETV-LAM), patients were switched from ETV to LAM 100 mg daily. Patients with elevated ALT or detectable viral load were not considered for the study. Other exclusion criteria included evidence of HCC and a history of decompensated liver cirrhosis. Written informed consent was obtained from all patients. This study was approved by the Ethics Committee / Institutional Review Board of the University of Hong Kong, and registered with ClinicalTrials.gov (NCT01013272).
The primary outcome of this study was the virological rebound. The secondary outcome was the development of drug-resistant mutation and biochemical flare. Patients had scheduled visits at weeks 0, 4, 12, 24, 48, 72, and 96. At each visit, routine liver biochemistry, hepatitis B serology, and serum HBV DNA measurements were performed. HBV DNA measurements were performed using the Cobas TaqMan (Roche Diagnostics, Branchburg, NJ) with a lower limit of detection of 60 copies/mL (12 IU/mL). Mutational analysis examining the presence of drug-resistant mutations was performed at weeks 0, 24, 48, 96, and at the time of virological rebound. This was determined by line-probe assay using INNO-LiPA HBV DR v2 and v3 (Innogenetics, Ghent, Belgium), allowing simultaneous detection of HBV polymerase wildtype and drug-induced mutations associated with LAM and ETV resistance at codons rt169, 173, 180, 184, 202, 204, and 250 of the HBV polymerase region. The presence of cirrhosis was defined by the appearance of a small-sized liver with nodular contour, or a coarse liver with evidence of portal hypertension on ultrasound.
All patients at the time of study enrolment had undetectable serum HBV DNA measurements (<60 copies/mL). Patients in the ETV-LAM group were switched back to ETV 0.5 mg therapy under protocol if there was evidence of virological rebound with LAM therapy. The criteria for virological rebound were strictly defined as a single HBV DNA level of >100 copies/mL (20 IU/mL), or persistent HBV DNA levels of 60-100 copies/mL (12-20 IU/mL) on three consecutive samples taken 2 weeks apart. Patients with virological rebound were switched back to ETV and continued on ETV for the remaining follow-up period. Additional mutational analyses using the methods described above were performed in these patients at the time of virological rebound. Patients who developed LAM resistance were treated with a combination of LAM and ADV.
All statistical analyses were performed using SPSS v. 16.0 (Chicago, IL). A chi-square test was used for categorical variables and Fisher's exact test when appropriate. Continuous variables with skewed distribution were analyzed using the Mann-Whitney test. P < 0.05 was considered statistically significant.
A total of 50 patients were recruited into either the ETV-ETV group (n = 25) or the ETV-LAM group (n = 25), with a follow-up period of 96 weeks in both groups. The baseline demographic and laboratory data are summarized in Table 1. These two groups were comparable for age, gender, HBeAg status, and liver biochemistry.
|Parameters||Entecavir Group n=25||Lamivudine Group n=25||P-value|
|Age (Years)||49 (23-56)||50 (22-62)||0.71|
|Sex (Male)||20 (80%)||16 (64%)||0.21|
|Cirrhosis||3 (12%)||1 (4%)||0.60|
|HBeAg-positive||5 (20%)||4 (16%)||1.00|
|Bilirubin (umol/L)||9 (2-46)||10 (7-16)||0.50|
|ALT (U/L)||27 (12-45)||22 (13-38)||0.41|
|Albumin (g/L)||45 (40-48)||46 (42-49)||0.09|
|Platelets (109/L)||187 (101-466)||193 (54-256)||0.37|
|HBV DNA (copies/mL)||<60||<60||0.66|
|Length of prior ETV (months)||11 (6-24)||13 (6-25)||0.33|
All patients had normal ALT at the time of study entry. During the scheduled visits at weeks 4, 12, 24, 48, 72, and 96, none of the patients had elevation of ALT above the upper limit of normal.
The patient flow and outcome are summarized in Fig. 1. In the ETV-ETV arm, none of the patients had detectable HBV DNA levels at weeks 4, 12, 24, 48, 72, and 96; all the patients maintained HBV DNA undetectability up to 96 weeks. In the ETV-LAM group, six patients developed evidence of virological rebound (with HBV DNA >100 copies/mL). All six patients maintained normal ALT at the time of virological rebound, and ALT remained normal through to 96 weeks, without evidence of liver decompensation. Two were HBeAg-positive; the remaining four patients were HBeAg-negative. The detailed demographic, biochemical, and virological profiles of the six patients (labeled patients 1 to 6) with virological rebound are summarized in Table 2. The flow and outcome of these six patients are shown in Fig. 2.
|Parameters||Patient 1||Patient 2||Patient 3||Patient 4||Patient 5||Patient 6|
|Rebound time (weeks)||12||12||24||72||96||96|
|Pre-ETV HBV DNA (copies/ml)||7.4 x 107||4.7 x 107||1.6 x 107||6.9 x 107||8.2 x 105||6.4 x 108|
|Length of ETV treatment (months)||15||10||18||10||19||18|
|ALT at time of rebound||Normal||Normal||Normal||Normal||Normal||Normal|
|HBV DNA at time of rebound (copies/mL)||115||467||430||235||39,867||3,497|
Of these six patients, three (patients 1, 2, and 3) developed rebound within the first 6 months at weeks 12, 12, and 24. These three patients had no genotypic resistance to LAM or ETV at the time of virological rebound. The remaining three patients (patients 4, 5, and 6) developed rebound at weeks 72, 96, and 96. For patients 1 to 4 who developed virological rebound before 96 weeks, only patient 1 did not have undetectable HBV DNA at week 96 after switching back to ETV. Patient 1 had an HBV DNA level of 522 copies/mL at week 96, which subsequently became undetectable with a repeat measurement outside the study period at week 108. Patients 5 and 6 developed virological rebound at week 96. Patient 5 had questionable drug compliance and declined either switching back to ETV or adding ADV. A repeat HBV DNA measurement at week 140 remained positive, albeit at a lower level of 9,253 copies/mL compared to 39,867 copies/mL at the time of rebound. Patient 6 had undetectable HBV DNA at week 108 after the addition of ADV.
Nineteen patients (76%) in the LAM arm had undetectable HBV DNA and normal ALT levels throughout the 96 weeks of switching to LAM therapy. There were no significant differences with respect to age, gender, HBeAg status, ALT, pretreatment HBV DNA levels, and length of prior ETV treatment between those with and without virological rebound in the LAM group (Table 3).
|Parameters||No Virological Rebound (n=19)||Virological Rebound (n=6)||P-value|
|Age (years)||50 (22-62)||52 (41-54)||0.828|
|Pre-ETV HBV DNA (log copies/mL)||7.0 (4.8-8.8)||7.9 (5.9-8.8)||0.514|
|Length of prior ETV (months)||13 (6-25)||17 (10-19)||0.555|
|At the time of randomization||11%||33%||0.234|
|Bilirubin (umol/L)||11 (7-16)||8 (7-15)||0.198|
|ALT (U/L)||22 (13-38)||19 (14-31)||0.176|
|Albumin (g/L)||46 (42-49)||45 (42-46)||0.687|
No LAM- or ETV-signature mutations were detected at the time of study entry. There was no evidence of drug-resistant mutations in the ETV arm and in those without virological rebound in the LAM arm throughout the study period. Of the six ETV-LAM switched patients who developed virological rebound, three patients (patients 1, 5, and 6) had tyrosine-methionine-aspartate-aspartate (YMDD) mutations at rt204 of the HBV polymerase (Fig. 2). Patient 1 had previously switched back to ETV at week 12 after an initial virological rebound. There was no evidence of YMDD mutation at the time of the first virological rebound, and subsequent HBV DNA levels became undetectable after switching back to ETV. This patient had evidence of a second virological rebound, with the appearance of both rtM204I and rtM250V mutations at week 72. No evidence of mutation occurred at rt180, rt184, and rt202, which are the other signature mutations for ETV resistance. This patient was subsequently changed to combination LAM + ADV, with undetectable HBV DNA at week 108. On further questioning, the patient recalled being treated with a 6-month course of LAM several years prior to his ETV treatment, although this was not known at the time of study enrolment. Patients 5 and 6 had evidence of YMDD mutation at the time of virological rebound at week 96. Patient 5 developed rtM204V mutation (without mutation at rt180) and declined additional add-on therapy with ADV or switching back to ETV at a dose of 1 mg (this was the patient with questionable compliance), and his HBV DNA remained positive at a low level at week 140. Patient 6 developed rtM204I mutation (without mutation at rt180) and received add-on ADV therapy, and repeat HBV DNA measurement at week 108 showed undetectable HBV DNA.
There were no new symptoms and no serious adverse events observed in either arm up to 96 weeks of follow-up.
To date, there is a lack of data regarding switching antiviral therapy in the management of CHB. Most of the existing data on switching therapy have involved a combination of immunomodulatory agent along with an oral nucleoside/nucleotide analog.11, 12 The use of switching therapy involving two oral nucleoside/nucleotide analogs have mainly been observed in studies where a second drug replaces the initial drug when drug resistance develops.13-17 Even less data exist for oral switching therapy in the absence of drug resistance. A recent study reported the efficacy of LAM-to-ETV switching.18 This study concluded that those with suboptimal viral suppression with LAM may be at risk of developing subsequent ETV resistance.
The current study aims to investigate the efficacy of oral ETV-LAM switching therapy after optimal viral suppression with ETV. This is a proof of concept for switching therapy in terms of the efficacy of viral suppression and the rate of drug resistance. The risk of LAM resistance should further be decreased by implementing a strict criteria for switching back to ETV (HBV DNA of >100 copies/mL). We hypothesized that the risk of developing LAM resistance in patients who already had undetectable HBV DNA levels should be low. This was based on previous studies showing that early viral suppression to PCR undetectability was associated with a more favorable outcome and less chance of developing drug resistance.4-7 In addition, those with lower pretreatment viral loads were associated with lower risk of resistance.19 This approach assumes that suppression of HBV DNA to undetectable levels using ETV is equivalent to the early virological suppression of LAM, which predicts a favorable outcome with LAM therapy. However, the risk of LAM resistance may still be present, particular in those with preexisting resistant mutations which may be selected out with the introduction of LAM monotherapy.
In the current study, none of the patients who remained on the ETV arm developed virological rebound, compared to six patients who switched to LAM, indicating suboptimal viral suppression with LAM. Of the six patients with virological rebound, three patients had evidence of YMDD mutations at the rt204 HBV polymerase region. One of these three patients had highly questionable compliance, which might have contributed to suboptimal viral suppression and subsequently increasing the risk of drug resistance. Another patient had prior exposure to LAM before ETV therapy; unfortunately, this history was not offered at the time of study enrollment. Switching to LAM and rechallenging again with LAM might predispose this patient to the development of M204I mutation. The presence of this mutation may subsequently increase the risk of developing ETV mutations after switching back to ETV. In patients with preexisting LAM-resistant mutations, the risk of developing ETV resistance is much higher.20 The remaining patient who developed YMDD mutation at 96 weeks had undetectable HBV DNA up until the last protocol follow-up, without evidence of drug-resistant mutation beforehand. This emphasizes the importance of on-treatment HBV DNA monitoring using a sensitive assay, even in patients with apparent good virological suppression. Furthermore, two patients developed virological rebound at 96 weeks, suggesting that the occurrence of virological rebound continues to increase over time. As the follow-up duration of the current study was relatively short, further increase in virological rebound rates is to be expected with longer follow-up. To confirm the long-term resistance rate, additional studies with longer duration of follow-up are needed, especially for those patients who relapsed after week 48.
Our hypothesis of continuation with antiviral agent of low potency with LAM after adequate viral suppression by ETV was not substantiated in 24% of patients. Twelve percent of patients developed LAM-resistant mutations. This rate is comparable to a previous study with LAM monotherapy which showed that despite optimal early viral suppression, resistance was observed in 8% after a median follow-up of 29 months.4
Therefore, prior HBV DNA suppression with ETV did not impart significant advantage in LAM-treated patients. The potential cost-saving benefit of switching to lamivudine is negated by the development of drug-resistant mutations, which would necessitate the addition of adefovir, or switching to tenofovir.
There are several limitations of the current study. First, the length of ETV therapy was not controlled, as patients were already taking ETV prior to enrollment into the study. However, this reflects the real-life clinical situation where patients will likely be on ETV for variable lengths of time before switching to LAM is considered. Moreover, there were no significant differences in the length of ETV therapy and pretreatment viral loads between those patients with and without virological rebound after switching to LAM. Second, the majority of the patients in the study were HBeAg-negative, and further studies may be needed to confirm these findings for HBeAg-positive patients. Third, neither the HBV genotypes nor the rate of achieving PCR negativity during the initial ETV treatment was determined, and therefore their effects on virological rebound could not be assessed.
To conclude, ETV-LAM switching resulted in a virological rebound rate of 24% and a resistance rate of 12% after 96 weeks. Prior optimal viral suppression with ETV did not confer any significant advantage in patients who switched to LAM.
- 10Entecavir at five years shows long-term maintenance of high genetic barrier to hepatitis B virus resistance. Hepatol Int 2008; 2: A88-89., , , , , , et al.