Recommendations
In persons with confirmed or suspected antiviral resistance
a,b,c (i.e. history of prior exposure or primary non-response) to lamivudine, entecavir, adefovir
d or telbivudine, a switch to tenofovir
e is recommended. (
Strong recommendation, low quality of evidence)
- a
Treatment failure: May be primary or secondary.
In settings with access to HBV DNA testing: Primary antiviral therapy failure may be defined as failure of a drug to reduce HBV DNA levels by ≥1×log10 IU/mL within 3 months following initiation of therapy. Secondary antiviral treatment failure may be defined as a rebound of HBV DNA levels of ≥1×log10 IU/mL from the nadir in persons with an initial antiviral treatment effect (≥1×log10 IU/mL decrease in serum HBV DNA).
In settings without access to HBV DNA testing: Treatment failure and drug resistance may be suspected based on the following features: receiving antiviral drugs with a low barrier to resistance together with documented or suspected poor adherence, laboratory measures such as an increase in serum aminotransferases, and/or evidence of progressive liver disease. Note: Elevation in ALT level tends to occur late and is a relatively poor predictive marker of resistance. Confirmation of antiviral drug failure can be established by sequencing the HBV DNA polymerase and identifying specific genetic markers of antiviral drug resistance.
- b
Treatment adherence should be reinforced in all persons with confirmed or suspected antiviral resistance. See also
Chapter 6, Box 6.2, Monitoring adherence to antiviral therapy.
- c
Some countries and health-care providers may consider switching persons to tenofovir from existing antiviral regimens with a low barrier to resistance before evidence of treatment failure, but no formal recommendation has been made in these guidelines.
- d
For adefovir resistance, a switch to either tenofovir or entecavir can be considered.
- e
To date, there has been no reported resistance with tenofovir. If there is primary non-response, then treatment adherence should be reinforced and monitored. At present, there is therefore no indication to switch to an alternative drug regimen.
7.1. Background
A major concern with long-term NA therapy is the selection of drug-resistance mutations. HBV has a high rate of replication with up to 1010–12 mutations generated every day. Higher rates of resistance are observed in persons with high baseline HBV DNA levels, longer duration of treatment, and a slower treatment-related decline in HBV DNA levels (1,2). Several drug-resistance mutations in the HBV polymerase reduce efficacy to more than one NA, resulting in cross-resistance to several agents, which limits future options for treatment. This is a particular risk in persons treated sequentially with NAs with a low barrier to resistance (lamivudine, adefovir and telbivudine) as monotherapy (3–8). Once drug-resistance mutations have developed, they are archived within the virus population and are rapidly selected if the same, or a cross-reacting antiviral agent, is reintroduced. The emergence of antiviral resistance usually leads to an increase in HBV DNA levels or viral rebound after initial response during therapy, which is likely to be followed by biochemical breakthrough with a rise in the ALT levels and, in some cases, hepatitis flares and progression to hepatic decompensation (6). In general, the management of such persons previously treated with lamivudine, adefovir or telbivudine is based on the established in vitro and in vivo efficacy of the potent NAs tenofovir and entecavir, and knowledge of the patterns of cross-resistance across different NAs (1,7,8).
Of the six approved NAs (lamivudine, adefovir, entecavir, telbivudine, tenofovir, emtricitabine), lamivudine is associated with the highest rate of drug resistance, entecavir with very low rates of resistance (except in persons previously exposed to lamivudine and adefovir), and currently none with tenofovir. The widespread use of lamivudine for persons with CHB and high HBV DNA levels in some countries has led to a high burden of lamivudine-resistant hepatitis B. Lamivudine resistance is of particular importance in the Asia–Pacific region where the prevalence of HBV infection is high, the infection is mainly acquired perinatally or in early childhood, and lamivudine and adefovir have been widely used without access to appropriate second-line regimens (1,2,9–15).
7.2. Summary of the evidence
Question: The purpose of the evidence review was to assess the most effective treatment regimen for the management of treatment failure due to resistance in persons previously treated with single agents with a low barrier to resistance (lamivudine, telbivudine or adefovir) (see
Web appendix 2:SRs7, 6b and 6d). The interventions analysed include switching to treatment with agents with a high barrier to resistance (tenofovir or entecavir) compared to adding in a second agent (combination therapy), or continuing regimens with a low barrier to resistance (lamivudine, telbivudine or adefovir). Key outcomes were rates of ALT normalization, undetectable HBV DNA, HBeAg seroconversion, HBsAg loss, reversion of fibrosis stage, mortality, severe adverse effects and antiviral resistance.
Systematic review and network meta-analysis
The systematic review (see
Web appendix 2: SR7) was based on data from one existing systematic review (16), comprising five RCTs and three non-randomized studies in China and South Korea, together with several randomized trials in persons with lamivudine resistance or a partial response to lamivudine (17–23). Included studies compared the effects of entecavir with either continuation of lamivudine, or a combination of lamivudine plus adefovir, or use of lamivudine plus adefovir versus continuation of lamivudine plus adefovir.
A switch to entecavir (compared with continuation of lamivudine) significantly improved virological and biochemical outcomes over 96 weeks (17–20). However, high rates of entecavir resistance were observed at 5 years. The quality of evidence for these outcomes was moderate due to imprecision. In the systematic review comparing entecavir with lamivudine plus adefovir, there were no differences in any of the assessed outcomes (undetectable HBV DNA, ALT normalization, and HBeAg seroconversion) after 48 weeks (16). The quality of evidence for these outcomes was low or very low.
Network meta-analysis: As tenofovir and entecavir have not been compared directly in an RCT, an NMA (see
Web appendix 2: SR6b) was also undertaken to enable a direct comparison and estimation of the relative efficacy and ranking of different antiviral therapies, based on another systematic review of all the relevant RCT data (both indirect and direct treatment comparisons of single, combination and sequential therapy) (18,24–32) used in the development of the UK NICE CHB guidelines (33). The treatments evaluated were a switch to an NA with a high barrier to resistance or continuation with or add-on therapy, and included the following agents: tenofovir, entecavir, adefovir, lamivudine, telbivudine and emtricitabine (in combination with tenofovir).
Seven RCTs of pair-wise comparisons based on 919 HBeAg-positive, lamivudine-resistant persons were included for the outcome of undetectable HBV DNA (<300 copies/mL [i.e. 60 IU/mL]), and six studies based on 771 persons for the outcome of HBeAg seroconversion (33). Tenofovir followed by entecavir plus adefovir combination therapy had the highest probability of achieving undetectable HBV DNA (66.2% and 33.8%, respectively) and HBeAg seroconversion (39.8% and 31.2%, respectively) at the end of 1 year of treatment among all the evaluated treatments. After 1 year of tenofovir treatment, 89% (95% CI: 51.8–98.2%) of lamivudine-resistant persons would be expected to achieve undetectable HBV DNA and 17.6% (95% CI: 1.4–74.9%) HBeAg seroconversion. No NMA was conducted for lamivudine-resistant, HBeAg-negative persons. The quality of the direct evidence (pair-wise comparisons) was rated as moderate to very low.
7.3. Rationale for the recommendations
Balance of benefits and harms
The Guidelines Development Group recognized that, in some countries, the widespread use of lamivudine and other NAs with a low barrier to resistance as first-line therapy for CHB has led to a high burden of resistant CHB. Overall, the Guidelines Development Group endorsed the principle that the most potent agent, and one which does not share cross-resistance, should be used to treat resistant CHB.
The Guidelines Development Group therefore recommended switching to tenofovir monotherapy as the most effective antiviral therapy for persons with confirmed or suspected lamivudine resistance for several reasons, which are listed below.
Despite the lack of direct evidence from RCTs on evaluation of tenofovir in persons with HBV drug resistance, evidence from the NMA showed that of all the antivirals considered, tenofovir is associated with the highest probability at 1 year of achieving low or undetectable HBV DNA levels in persons with lamivudine-resistant HBV. The Guidelines Development Group considered that the same tenofovir switch strategy would also apply to HBeAg-negative persons, although no NMA was available for this group. This would have a beneficial effect on disease progression and also reduce possible transmission of resistance.
There are deleterious consequences of continuing treatment with an ineffective antiviral agent, and ongoing HBV replication confers an increased risk of disease progression to cirrhosis, end-stage liver disease and HCC.
The use of tenofovir, which does not share cross-resistance, would avoid the selection of further compensatory mutations and development of drug resistance, with reservoirs of resistant HBV mutants. Clinical and molecular evidence indicates that resistance to lamivudine (L180M + M204V/I) confers cross-resistance to telbivudine and entecavir, but not tenofovir. In addition, although treatment failure and development of resistance occurs rarely in naive persons treated with entecavir, resistance to entecavir is more common in persons with lamivudine resistance. The Guidelines Development Group therefore recommended that entecavir not be used as salvage therapy in persons with known or suspected lamivudine resistance (
34).
Primary non-response (defined as less than 1 log decrease in HBV DNA level after 3 months of treatment, in settings where HBV DNA testing is available) is rare in persons initiating and adherent to entecavir or tenofovir treatment, but can occur in persons treated with lamivudine, adefovir or telbivudine. Sequential treatment of persons with lamivudine-resistant CHB with adefovir or telbivudine or entecavir can lead to the selection of multidrug-resistant hepatitis B and should be avoided.
A switch to tenofovir monotherapy in persons who have developed resistance to lamivudine, adefovir, telbivudine or entecavir simplifies clinical management and drug procurement.
There was little evidence from the systematic review for an advantage of adding NAs or combined use of NAs conferring a benefit in cases of lamivudine resistance.
Tenofovir has the potential to be more widely available and affordable in LMICs through access to reduced prices via a range of mechanisms including license agreements negotiated with the Medicines Patent Pool for use in HIV infection (but also available for HBV infection).
The Guidelines Development Group also recognized that the most common reason for virological breakthrough is poor adherence, and therefore regular counselling should be offered on the importance of treatment adherence, especially in persons with evidence of virological breakthrough.
The Group also recognized that the most effective strategy to minimize the future burden of lamivudine resistance was the wider use of NAs with a high barrier to resistance in first-line therapy. The Guidelines Development Group considered that some countries and physicians may consider switching persons to tenofovir from existing antiviral regimens with a low barrier to resistance before evidence of treatment failure, but no formal recommendation was made.
Resource considerations
Drug costs: see
Chapter 6: First-line antiviral therapies: resource considerations.
Diagnosis of treatment failure: Measurement of HBV DNA levels and testing for drug resistance are fundamental to confirming treatment failure and genotypic HBV resistance, but there is extremely limited access to these in LMICs. In these settings, ascertainment of the development of resistance will largely be based on clinical suspicion and, in some instances, by an increase in serum aminotransferases. However, elevation in ALT tends to occur late and has been shown to be a relatively poor predictive marker of resistance (35). In countries where resistance testing is not available, a change to tenofovir would not incur added costs, although this may not be applicable in Asia.
BOX 7.1Diagnosing treatment failure
Objective monitoring of adherence to antiviral therapy is essential for effective long-term management of CHB. Each clinic visit is an opportunity for assessing and supporting treatment adherence, and may require a combination of approaches, depending on the local context.
Treatment adherence should be reinforced in all persons with confirmed or suspected antiviral resistance. See also
Chapter 6, Box 6.2, Monitoring adherence to antiviral therapy.
Definition of treatment failure
In settings where HBV DNA testing is available: Primary antiviral treatment failure may be defined as failure of an antiviral drug to reduce HBV DNA levels by ≥1×log10 IU/mL within 3 months. Secondary antiviral treatment failure may be defined as a rebound of HBV DNA levels of ≥1×log10 IU/mL from the nadir in persons with an initial antiviral treatment effect (≥1×log10 IU/mL decrease in serum HBV DNA).
In settings where HBV DNA testing is not available: Treatment failure and drug resistance may be suspected based on the following features: receiving antiviral drugs with a low barrier to resistance together with documented or suspected poor adherence, and laboratory measures such as an increase in serum aminotransferases, and/or evidence of progressive liver disease.
Note: Elevation in ALT level tends to occur late and is a relatively poor predictive marker of resistance.
Confirmation of antiviral drug failure can be established by sequencing the HBV DNA polymerase and identifying specific genetic markers of antiviral drug resistance.
Research gaps
Evaluate further the utility and predictive value of monitoring ALT levels and other markers to identify the development of genotypic or phenotypic resistance.
Evaluate the impact of treatment with NAs with a high genetic barrier to resistance in persons with treatment failure, and on other important outcomes, such as histological improvement, development of further drug resistance and adverse events.
