4 Evidence and interpretation

The Appraisal Committee (appendix A) considered evidence from a number of sources (see appendix B).

4.1 Clinical effectiveness

4.1.1 Three trials of peginterferon alfa-2a and five trials of interferon alfa-2b that included people with chronic hepatitis C, at least 70% of whom had mild disease, were included in the assessment report. All studies included the combination with ribavirin in at least one arm. The comparators in the trials varied. Two of the three peginterferon alfa-2a studies compared longer courses (48 weeks) with shorter courses (24 weeks); the third compared peginterferon alfa-2a with non-pegylated interferon alfa-2a.

4.1.2 No studies compared the strategies of early treatment of mild disease with a strategy of watchful waiting and treatment for those who progress to moderate or severe disease. Because this is the primary comparison of interest in this appraisal, results for the comparator arms in the studies in section 4.1.4 are not reported here.

Peginterferon alfa-2a plus ribavirin

4.1.3 HCV genotypes 2 and 3

4.1.3.1 The three trials of peginterferon alfa-2a plus ribavirin in people with genotypes 2 and 3 yielded sustained virological response rates of 72–84% after 24 weeks of treatment and rates of 78–80% after 48 weeks of treatment. For people infected with these genotypes, there was no statistically significant difference between the rates after 24 weeks of treatment or after 48 weeks. These results are consistent with those for studies conducted in people with moderate or severe chronic hepatitis C; in the previous appraisal (TA 75), the corresponding rate of sustained virological response in different trials that included people with both mild and moderate or severe disease was 79%.

4.1.4 HCV genotype 1

4.1.4.1 In people infected with genotype 1, peginterferon alfa-2a plus ribavirin yielded sustained virological response rates of 13–42% after 24 weeks of treatment and 40–52% after 48 weeks. In this group, the additional 24 weeks of treatment statistically significantly increased the rate of sustained response.

Peginterferon alfa-2b plus ribavirin

4.1.5 Although there was no trial of peginterferon alfa-2b combination therapy that met the Assessment Group's criteria for inclusion of trials of people with mild chronic hepatitis C, one study included 1014 (of 1530) participants who had been documented as having no or minimal fibrosis. In this study, a sustained virological response occurred in 57% of participants with no or minimal fibrosis (that is, with mild disease) who received 'high-dose' peginterferon alfa‑2b (1.5 micrograms/kg/week for 48 weeks) plus ribavirin. This compares with a rate of 44% among those who had bridging fibrosis or cirrhosis (that is, moderate or severe disease) who received high-dose peginterferon alfa-2b. In people who received 'low-dose' peginterferon alfa-2b (1.5 micrograms/kg/week for 4 weeks, then 0.5 micrograms/kg/week for 48 weeks) plus ribavirin, a sustained virological response rate of 51% was recorded among those with mild disease compared with 43% among those with moderate or severe disease. Results by genotype were not reported. The rate of sustained virological response was statistically significantly higher among participants on high-dose (but not low-dose) peginterferon alfa-2b combination therapy than among patients on non-pegylated interferon alfa-2b plus ribavirin.

Monotherapy trials

4.1.6 People who are unable to take ribavirin and are treated with monotherapy with peginterferon alfa or non-pegylated interferon alfa have much lower response rates than people treated with combination therapy. One trial of 159 people compared response rates with three different regimens of peginterferon alfa-2a with one of interferon alfa-2a. Rates of sustained virological response were 3% for interferon alfa-2a and between 10% and 29% for peginterferon alfa-2a, depending on the dose. Altogether, 82% of participants were classified as having mild chronic hepatitis C.

4.1.7 Another trial of 1219 people compared response rates for three different regimens of peginterferon alfa-2b and one of interferon alfa-2b. Altogether, 83% of participants in this study were classified as having mild disease. Rates of sustained virological response were 12% for those treated with interferon alfa-2b and between 18 and 23% for those treated with peginterferon alfa-2b, depending on the dose.

Summary

4.1.8 Taken as a whole, the evidence for combination therapy and monotherapy, and for pegylated and non-pegylated interferon alfa-2a and pegylated and non-pegylated interferon alfa-2b, suggests that rates of sustained virological response among patients with mild disease are about the same as those among patients with moderate or severe disease. Combination therapy with peginterferon alfa (2a or 2b) and ribavirin produces higher rates of sustained virological response than combination therapy with non-pegylated interferon alfa (2a or 2b). Monotherapy with pegylated interferon alfa-2a or alfa-2b produces higher response rates than monotherapy with non-pegylated interferon alfa.

4.2 Cost effectiveness

4.2.1 The assessment report found six studies examining the cost effectiveness of treatment for people with mild disease. Three of these studies compared interferon combination therapy with no treatment rather than with delayed treatment. These three studies showed that interferon combination therapy was cost effective when compared with standard care (all estimated mean incremental cost-effectiveness ratios [ICERs] were less than £10,000 per quality-adjusted life year [QALY]). Two studies compared early treatment with peginterferon alfa combination therapy with delayed treatment. They showed that, for genotypes 2 and 3, early treatment is apparently cost effective when compared with delayed treatment, but the case for early treatment for genotype 1 is less clear.

4.2.2 The model employed in the Roche submission determined the cost effectiveness of peginterferon alfa-2a plus ribavirin against no treatment. The estimated mean cost per QALY for treating people with mild disease was £1000 for genotypes 2 and 3, and £4000 for genotype 1.

4.2.3 The model employed in the Schering-Plough submission determined the cost effectiveness of peginterferon alfa-2b plus ribavirin against no treatment. The model is academic-in-confidence. The estimated mean cost per QALY for treating mild chronic hepatitis C was £1000 for genotypes 2 and 3, and £3000 for genotype 1.

4.2.4 The model developed by the Assessment Group incorporated seven health states: remission, mild disease, moderate disease, compensated cirrhosis, decompensated cirrhosis, liver cancer and liver transplantation. It used rates of sustained virological response from the manufacturers' submissions and transition rates between the seven health states from a number of sources. For example, the estimated transition rate from mild to moderate disease as used in the model was 2.5% per year, which was obtained from observational data relating to 373 cases from a routine practice in a hospital in London between 1990 and 2001. Health-state utilities and costs were estimated from the UK mild hepatitis C trial. For comparability with the previous review, benefits were discounted at 1.5% per year and costs at 6%, with a sensitivity analysis at 3.5% for both costs and benefits.

4.2.5 Non-1 genotype HCV

4.2.5.1 For people with non-1 genotype infection, the base case estimated mean ICER for deferring treatment with peginterferon alfa‑2a combination therapy until the disease reaches the moderate or severe stage relative to best supportive care (no treatment) is £1300 per QALY. The corresponding ICER for peginterferon alfa-2b plus ribavirin is £1400. This analysis confirms that the strategy of treating people with moderate or severe disease is cost effective compared with not treating them at all.

4.2.5.2 The estimated mean ICERs for early treatment with peginterferon alfa in combination with ribavirin are £3700 for peginterferon alfa-2a and £4300 for peginterferon alfa-2b, when compared with deferring treatment until the disease reaches the moderate-to-severe stage in people with non-1 genotype infection. This analysis shows that treatment of mild disease is cost effective compared with waiting until the patient reaches the moderate stage of the disease. These are the key cost-effectiveness estimates for this appraisal.

4.2.5.3 Further analysis shows that combination therapy with pegylated interferon is cost effective when compared with the corresponding non-pegylated therapy for treating people with mild disease.

4.2.6 Genotype 1 HCV

4.2.6.1 For people with genotype 1 infection, the base case estimated mean ICER is £6900 per QALY for watchful waiting followed by treatment with peginterferon alfa-2a combination therapy when the disease reaches the moderate or severe stage, relative to best supportive care (no treatment). The corresponding ICER for peginterferon alfa-2b plus ribavirin is £4700. This analysis confirms that the strategy of treating only moderate or severe disease is cost effective compared with not treating the infection at all.

4.2.6.2 For people with genotype 1 infection, the estimated mean ICERs for early treatment with peginterferon alfa in combination with ribavirin are £10,300 for peginterferon alfa-2a and £8300 for peginterferon alfa-2b when compared with deferring treatment until the disease reaches the moderate-to-severe stage, and reflecting the early stopping rules recommended in the summary of product characteristics for peginterferon alfa-2a. This analysis shows that treating mild disease is very likely to be cost effective compared with waiting until the patient reaches the moderate stage. These are the key cost-effectiveness estimates for this appraisal.

4.2.6.3 Further analysis shows that pegylated interferon combination therapy is cost effective when compared with the corresponding non-pegylated therapy for treating people with mild disease.

4.2.7 Monotherapy: all genotypes

4.2.7.1 In people unable to take ribavirin, monotherapy with peginterferon may be used. The estimated mean ICER for early treatment with peginterferon alfa monotherapy is £3000 per QALY for peginterferon alfa-2a and £2300 for peginterferon alfa-2b against the same treatment deferred until a
later stage.

4.2.8 Sensitivity analyses

4.2.8.1 Estimated mean ICERs are even lower using current stopping rules, whereby treatment is stopped at 12 weeks if a 100-fold reduction in viral load has not occurred.

4.2.8.2 Sensitivity analyses conducted in the assessment report do not lead to ICERs of more than £20,000 except when the average age of patients is increased by 15 years.

4.3 Consideration of the evidence

4.3.1 The Committee reviewed the data on the clinical effectiveness and cost effectiveness of interferon alfa (pegylated and non-pegylated) and ribavirin for the treatment of mild chronic hepatitis C, having considered evidence on the nature of the condition and the value placed on the benefits of this treatment by people with the condition (or who have had the condition), those who represent them, and clinical experts. It was also mindful of the need to take account of the effective use of NHS resources.

4.3.2 The Committee heard from clinical and patient experts that studies show the quality of life of people with mild disease, even in the absence of histological evidence of definitive liver disease, is on average lower than that of people who have had mild disease and who have been cleared of the virus. The experts advised the Committee that such a difference in quality of life is an important contributor to the benefit of early treatment of mild disease.

4.3.3 The Committee also heard from the experts that sustained virological response, when the virus is undetectable 6 months after treatment has finished, is maintained for 10 years in more than 90% of people. Additionally, it is unusual for re-infection with HCV to occur after a sustained virological response has been achieved. These observations were consistent with the assumptions of the cost-effectiveness model used in the assessment report. Therefore, the Committee concluded that the model inputs were appropriate in these respects.

4.3.4 The Committee understood that about half of the approximately 3000 people in England and Wales who are treated each year have a sustained virological response and are effectively cured, but that the number of new cases is greater than the number of people being cured. Therefore, the number of people with the disease is still rising.

4.3.5 The Committee heard that early treatment of people with chronic hepatitis C in the general population could potentially reduce the likelihood of the spread of infection and that this would lead to a better estimate of cost effectiveness than that demonstrated in the models, in which the effect of the spread of infection had not been considered.

4.3.6 The Committee considered that, although the predicted effectiveness of the treatment of mild disease in a clinical setting was not as high as the average efficacy seen in the clinical trials, this difference was not likely to affect the importance of the therapy or overall cost effectiveness.

4.3.7 The Committee discussed a recent change in the UK marketing authorisation for peginterferon alfa-2b for people with genotype 1 (and by extension, genotype 4) and low viral loads (less than 600,000 IU/ml). This change allows combination therapy to be stopped at 24 weeks rather than continue to 48 weeks. The Committee has not reviewed the clinical effectiveness evidence for this modification of the marketing authorisation, so is not able to comment specifically on it. However, the Committee did note the caution carried in the marketing authorisation regarding the possibility that there may be a lower rate of sustained viral response if treatment is limited to 24 weeks. Nevertheless, if the success rate for sustained viral response were not to be affected by the shortening of the treatment period, the Committee recognised that this regimen would be cost effective compared with the 48-week treatment.

4.3.8 The Committee considered that an important question was whether it is cost effective to treat immediately people who have mild chronic hepatitis C with peginterferon alfa plus ribavirin or to wait until their disease reaches the moderate stage ('watchful waiting'). The Committee disregarded the manufacturers' models, which they considered had not made the relevant comparison. The Committee understood that, if the rate of progression to moderate disease were sufficiently low, it might be better not to subject all people with mild disease to the side effects of combination therapy because few might progress to moderate disease or beyond.

4.3.9 The Committee discussed the progression rates (from mild to moderate disease). The Committee was told that there is no way of knowing which patients are likely to progress from mild disease to moderate disease, but that progression is somewhat slower for younger people, for those who have been infected for a shorter time, for females and for certain ethnic groups. The Committee considered at length the progression rates assumed in the assessment report model, and noted that these were from clinical trials of people who presented for treatment. The Committee appreciated that this may represent a small proportion of the total number thought to be infected with HCV and that people who were asymptomatic and had not sought treatment may experience a different rate of progression, on average, from those who had sought treatment. The Committee accepted that low rates of progression would mean that treating people with mild disease may affect the overall cost effectiveness of early treatment when compared with watchful waiting. The Committee considered that although low progression rates may exist among the population of all people with mild disease, it was likely that the progression rates found in clinical studies and used in the cost-effectiveness analysis might only be applicable to people with mild disease who present for treatment. Therefore, the Committee concluded that the estimated ICERs in the clinical setting currently pertaining in England and Wales would be acceptable.

4.3.10 However, the Committee recognised that if a much higher proportion of people with mild disease were to be diagnosed (for example, from a screening programme), the average progression rate of such a group could be so low that it might no longer be optimal to offer early treatment. That is, the Committee considered that the cost effectiveness of treating all people with mild disease identified as a result of a screening programme has yet to be proven, and that this guidance to treat people with mild disease might not necessarily apply to a screened population.

4.3.11 The Committee considered estimates used in the economic models of improvements in quality of life for people receiving treatment for mild disease and whether these improvements were importantly different between mild and moderate disease. The Committee heard that many people infected with chronic hepatitis C believe that they are stigmatised and discriminated against, which further reduces their quality of life. Clearing the disease could thus be associated with improvements that may not be reflected in economic models, and this could lead to an underestimate of the benefits of treatment. The Committee was persuaded that the estimated ICERs for combination therapy or monotherapy with peginterferon alfa for mild disease remained below £26,000 per QALY even if only very small improvements in quality of life (1%) were considered. This was the same for all genotypes. The estimated ICER could be as high as £42,000 per QALY for genotype 1 (the genotype with the highest estimated ICER) only when there was assumed to be no improvement in quality of life when mild disease was cleared. The Committee was not persuaded that this latter scenario was appropriate, so accepted the range of predicted ICERs less than £26,000 per QALY.

4.3.12 The Committee was persuaded by the experts that the previous guidance (TA 75) on treating people with moderate chronic hepatitis C who continue to use intravenous drugs and/or misuse alcohol and people co-infected with HIV should be extended to members of all such groups who have mild disease. Thus, the Committee concluded that, with respect to those continuing to use intravenous drugs, in naturalistic settings, the rate of discontinuation of treatment would not be so great as to prevent the treatment being cost effective. In addition, with respect to people who continue consuming alcohol, the Committee considered that continued alcohol consumption is not in itself an absolute contraindication to therapy, but it should be emphasised as an important contributory factor to the development of liver disease and should be taken into account in advice and information given by the clinical team.

4.3.13 The Committee was mindful that the guidance on recommending treatment for mild disease caused by all genotypes would mean that liver biopsy will no longer be required to diagnose the severity of the disease before treatment can be initiated. It was felt that this would increase the uptake of treatment among people unwilling to undergo this procedure. Additionally, it would reduce the cost of disease management somewhat and avoid the pain and complications associated with liver biopsy. However, the Committee heard from a clinical expert that the number of biopsies carried out in people with chronic hepatitis C may not fall a great deal because biopsies would still be recommended for reasons not directly related to the decision to initiate therapy.

4.3.14 Treatment with combination peginterferon therapy has side effects, and the Committee appreciated that some people with mild disease may decide, in consultation with their clinician, to wait until they reach the moderate stage of the disease before beginning treatment. The Committee understood that, although people opting to begin treatment with combination therapy immediately may not need a liver biopsy, the situation may be different for people who opt to defer therapy. Clinicians will need to discuss the possible need for liver biopsy with people who opt to defer treatment. It should be explained that biopsy is used to determine whether a person has progressed to the moderate stage of the disease.

4.3.15 The Committee decided that the previous guidance on moderate or severe disease (TA 75) should be retained for mild disease with respect to: the length of treatment for different genotypes; people requiring monotherapy; and second courses of treatment. The Committee did not believe that there was sufficient evidence to recommend combination therapy or monotherapy with peginterferon alfa for people with mild chronic hepatitis C who are under the age of 18 years, or those who have had a liver transplant.