4 Consideration of the evidence
4.1 The Appraisal Committee reviewed the data available on the clinical and cost effectiveness of telbivudine for the treatment of chronic hepatitis B, having considered evidence on the nature of the condition and the value placed on the benefits of telbivudine by people with chronic hepatitis B, those who represent them, and clinical specialists. It was also mindful of the need to take account of the effective use of NHS resources.
4.2 The Committee discussed the decision problem and evidence presented in the manufacturer's submission. It discussed with the clinical specialists the importance and relevance of the various possible surrogate markers of disease expression and response to treatment. The Committee heard from the patient experts about the impact of hepatitis B on quality of life and the importance of having a variety of treatments available. The Committee was also mindful of the long-term risk of progression to cirrhosis or hepatocellular carcinoma associated with chronic hepatitis B infection and the resulting impact in terms of costs, mortality and health-related quality of life. The Committee agreed that avoiding these adverse consequences was the most important goal in the treatment of chronic hepatitis B. It noted that the relationship between any surrogate endpoints measured in clinical studies and these final health outcomes should be taken fully into consideration.
4.3 The Committee was advised by the clinical specialists of the differences between HBeAg-positive and HBeAg-negative disease. It acknowledged that, in the main, rather than different infections, HBeAg-positive and -negative disease represent different stages of infection. This is because HBeAg-negative chronic hepatitis B most commonly develops when the virus that was originally suppressed following HBeAg/antibody seroconversion mutates and the infection re‑emerges from immune control. The Committee also understood that hepatitis in the HBeAg-negative phase of the disease carries a high risk of progression to cirrhosis or hepatocellular carcinoma. Therefore, it is important to maintain a low viral load in patients with HBeAg-negative disease.
4.4 The Committee was advised by the clinical specialists of the relative importance of different tests in the diagnosis and management of chronic hepatitis B. It was persuaded that measurement of viral load (HBV DNA) is an important predictor of future liver damage, and can be used to identify patterns of viral resistance to treatment. The clinical specialists confirmed that, in HBeAg-positive disease, reductions in HBV DNA levels by antiviral treatment may accelerate seroconversion. They also stated that in HBeAg-positive disease, seroconversion could indicate that treatment could be stopped, although current clinical practice is to continue for 6 months after seroconversion. The Committee understood that this endpoint of treatment did not apply to HBeAg-negative disease and that assessment of when to stop treatment was more difficult. In many cases treatment would need to be lifelong. The clinical specialists stated that serum ALT levels were usually correlated with HBV DNA levels and serum ALT levels would be expected to normalise with a reduction in HBV DNA levels. The Committee also heard from the specialists that, in HBeAg-positive disease, spontaneous HBeAg/antibody seroconversion is associated with high serum ALT levels and that high serum ALT alone, without histological evidence of liver disease, was not an indication for treatment. The Committee heard that it is current clinical practice to start antiviral treatment only on the basis of liver inflammation (confirmed by biopsy) irrespective of serum ALT levels. This is reflected in telbivudine's marketing authorisation.
4.5 The Committee considered the treatments available for patients with chronic hepatitis B in the UK. It discussed with the patient experts and clinical specialists the relevance of previous NICE guidance on chronic hepatitis B and where in the treatment pathway telbivudine should be used. The clinical specialists stated that telbivudine monotherapy could be used in place of lamivudine monotherapy. However, they also stated that lamivudine monotherapy was not a preferred option; in particular it was not considered suitable in highly replicative disease because of the associated high rate of emergence of viral resistance. Combination therapy was considered more appropriate in these instances.
4.6 The Committee considered evidence of telbivudine's efficacy in the subgroup of patients with serum ALT levels greater than or equal to twice the upper limit of normal (identified from the GLOBE trial population in the manufacturer's submission). The Committee was advised by the clinical specialists that estimates of telbivudine's efficacy in this subgroup were uncertain because they were based on a post-hoc analysis. The Committee expressed concerns over the relevance of the GLOBE trial population to UK practice, but it was persuaded by the clinical specialists that the ethnic mixes of the trial and UK patient populations were similar.
4.7 The Committee discussed the ERG critique of the efficacy results from the GLOBE trial; in particular, concerns that health-related quality of life data were not reported. However it concluded, on the basis of the clinical evidence from the GLOBE trial and testimonies from the clinical specialists and patient experts, that telbivudine was likely to be more effective than lamivudine for several of the outcomes measured, notably the primary endpoint (suppression of HBV DNA to less than 5 log10 copies/ml plus either clearance of detectable HBeAg or ALT normalisation). The Committee also noted that based on 2-year data there was a lower rate of viral resistance to treatment than was seen with lamivudine. However, it noted that resistance to telbivudine was likely to be problematic in the long term and that comparisons with treatment strategies involving the addition of other antivirals, such as adefovir dipivoxil, were the most appropriate for the evaluation of cost effectiveness.
4.8 The Committee discussed the economic analysis presented in the manufacturer's submission, the ERG's critique of the submission and the exploratory analyses undertaken by the ERG. In particular, it discussed the complexity and lack of transparency of the viral load model. With regard to transparency, the Committee was impeded by the lack of detail provided in the manufacturer's submission about which parameters were used. With regard to complexity, the Committee acknowledged that the natural history of the disease required a number of health states to be defined in the economic models. However, the Committee noted that the large number of health states meant that the data available from clinical studies were not sufficient to support clearly the transition probabilities indicated. In addition, the methods used to deal with the sparseness of the data had led to uncertainty about the outputs of the manufacturer's economic models. The Committee considered that both this complexity and lack of transparency undermined the credibility of the economic results. The Committee noted that the manufacturer did not consider alternative approaches that might have reduced the complexity of the viral load model. This, together with appropriate risk modelling, might have reduced the data requirements for populating the viral load model. The Committee noted that the economic results generated by the viral load model appeared sensitive to the choice of 'priors' and noted that the manufacturer did not present any univariate sensitivity analyses that identified the key drivers of cost effectiveness in either model. The Committee accepted that the sensitivity analyses presented by the ERG for both of the economic models showed a reduction in the probability of telbivudine being cost effective at willingness to pay thresholds of £20,000 and £30,000 per additional QALY gained. This resulted in lamivudine being the preferred option in the range of cost-effectiveness estimates that are usually seen to represent a cost-effective use of NHS resources.
4.9 The Committee considered that the transparency of the viral load model for assessing the cost effectiveness of telbivudine for the treatment of HBeAg-negative patients was reduced by the lack of detail in the manufacturer's submission about which parameters were used. Additionally, the Committee was mindful that the manufacturer had commented during consultation that the concerns raised about the viral load model could not be rectified within the time constraints of the appraisal. The Committee noted the manufacturer's acceptance of the fact that this made it difficult to judge the cost effectiveness of the use of telbivudine in HBeAg-negative patients. The Committee concluded that, in light of the uncertainty about the cost effectiveness of telbivudine in HBeAg-negative chronic hepatitis B and the sensitivity analyses presented by the ERG, telbivudine could not be recommended as a cost-effective use of NHS resources.
4.10 In considering the cost effectiveness of telbivudine for treating HBeAg-positive chronic hepatitis B, the Committee was able to proceed further by considering the seroconversion model provided by the manufacturer. However, the Committee noted that although it was based on a model used in a previous appraisal, the seroconversion model focused solely on a subset of the GLOBE trial population (specifically only people with serum ALT levels greater than or equal to twice the upper limit of normal). The Committee discussed the validity of clinical efficacy analysis based on this subgroup; it considered that the validity of the results was dependent on the statistical integrity of the subgroup as well as its biological plausibility and clinical relevance. The manufacturer provided partial reassurance on the statistical integrity of this subgroup: randomisation of the GLOBE trial had been stratified into treatment-eligible groups with serum ALT levels greater than two and a half times the upper limit of normal. The Committee was further mindful of comments from the manufacturer that the efficacy analyses of telbivudine in the subgroup of patients with serum ALT levels greater than or equal to twice the upper limit of normal was exploratory and the study was not powered to demonstrate differences between these subgroups. With regard to the clinical relevance of the subgroup, the Committee was mindful of comments from the clinical specialists that antiviral treatment of chronic hepatitis B was initiated principally on the basis of histological confirmation of liver inflammation, irrespective of serum ALT levels.
4.11 The Committee discussed the updated economic analysis based on the seroconversion model presented by the manufacturer. The Committee considered that although the adjustments made addressed some of the criticisms made by the ERG, the economic analysis was still based solely on the subgroup of patients with serum ALT levels greater than or equal to twice the upper limit of normal. Combined with the results of the sensitivity analyses presented by the ERG on the seroconversion model, the Committee considered that it therefore did not have a sufficient basis on which to recommend telbivudine as a cost-effective use of NHS resources in people with HBeAg-positive chronic hepatitis B. The Committee was also mindful that recommending a treatment that was somewhat more effective than lamivudine monotherapy would not necessarily be helpful in the context of highly replicative disease in which resistance was likely to develop rapidly, for which combination therapy was more appropriate.
4.12 Overall, the Committee agreed that there was evidence that telbivudine was likely to be more clinically effective and have a more favourable resistance profile than lamivudine monotherapy in patients with HBeAg-positive disease. However, it did not agree with the manufacturer that the evidence presented on the cost effectiveness of telbivudine in the subgroup of patients with serum ALT levels greater than or equal to twice the upper limit of normal could be used as a reliable basis for decision-making in patients with HBeAg-positive disease.
4.13 In light of the economic models and evidence presented, the Committee concluded that telbivudine, within its licensed indication, could not be recommended as a cost-effective use of NHS resources for the treatment of chronic hepatitis B.