Eliglustat for treating type 1 Gaucher disease

The committee understood that type 1 Gaucher disease is chronic, and that it needs lifelong treatment and causes symptoms such as fatigue, bone pain and reduced mobility. The committee noted comments from the patient experts that there is considerable impact on bones. This leads to varying forms of disability and, even with current treatments, people can experience symptoms such as fatigue, bruising, bone pain and, in those with severe disease, fractures. The committee heard from the patient experts about the profound impact the disease has on patients' and carers' quality of life and emotional wellbeing. The committee concluded that type 1 Gaucher disease is a debilitating condition that has a significant impact on quality of life.

The committee discussed the current treatment options and management of type 1 Gaucher disease. The committee heard that the main treatment options available are imiglucerase and velaglucerase alfa, both of which are recommended by the Lysosomal Storage Disorder Expert Advisory Group and nationally commissioned. The committee heard that the 2 treatments are considered equivalent in terms of efficacy, but velaglucerase alfa is preferred because it has a lower cost. The committee heard that miglustat, a substrate reduction therapy, may be offered to people for whom ERT is not suitable. However, the clinical experts highlighted that its efficacy is modest and that it is not well tolerated. Supportive therapy (for example, blood products, bisphosphonates, analgesics) may be offered to patients not having ERT or miglustat, or alongside these treatments in patients with complications. The committee understood from the clinical experts that, for this reason, the most relevant comparators for eliglustat are velaglucerase alfa and imiglucerase. The committee heard that ERT was an established and effective treatment option that had changed the treatment landscape for type 1 Gaucher disease. However, patient experts highlighted that they were administered intravenously and that this could be burdensome for patients, resulting in poor quality of life and mental wellbeing. The advantages of an oral treatment were emphasised, that is, more freedom to travel and attend university, and to live a more normal life without regular transfusions. The committee concluded that intravenous ERT, such as velaglucerase alfa and imiglucerase, were established treatments in the NHS, but that an oral treatment option would be of significant value to patients.

The committee considered the clinical-effectiveness evidence presented by the company. It noted the evidence review group's (ERG's) comments that the trials were of reasonable quality. It heard from the clinical experts that the populations were generalisable to patients in clinical practice in England. However, the ERG highlighted that the non-inferiority margin of 25% for the ENCORE trial primary composite outcome was wider than normal. The committee was aware that the European Medicines Agency's Committee for Medicinal Products for Human Use noted that the trial did not comprehensively show that the usual regulatory standard of −20% had been achieved. The committee noted the company's explanation that the European Medicines Agency accepted a broader margin because of the rare nature of the disease, meaning that a larger trial could not feasibly be conducted. The committee understood the challenges in developing a clinical trial programme for a rare condition, and concluded that the ENCORE trial was sufficiently robust for its decision-making.

The committee discussed the appropriate dose for the ERT. It was aware that the dosages specified in the summary of product characteristics for imiglucerase and velaglucerase alfa (starting dosages of 60 U/kg every 2 weeks) were higher than those recommended in the NHS England standard operating procedure (maintenance dose of 15–30 U/kg). In the ENCORE trial, 58% of people had dosages of imiglucerase of at least 35 U/kg every 2 weeks. The committee questioned which dose reflects clinical practice in England. It heard from clinical experts that the approach in practice is to titrate the dose of ERT and use the lowest effective dose. It heard that patients generally start on 30 U/kg, followed by close monitoring for the first 12 months, with further dose reductions depending on response. The clinical experts stated that some people with newly diagnosed type 1 Gaucher disease occasionally have very severe disease and may need a higher starting dose. The committee concluded that the dose recommended in the standard operating procedure was reflective of clinical practice.

The committee discussed the dose of imiglucerase in the ENCORE trial. The committee noted that the dosage in ENCORE was between 30 U/kg and 130 U/kg every month and these efficacy data were used in the model. However, the dose in practice is lower (see section 5.4) and the company questioned if the efficacy data would need to be adjusted accordingly. The committee heard from the ERG that that the data showed that, in people having lower doses of ERT, their condition continues to respond to treatment. The clinical expert confirmed that, because treatment is individualised, the dose is titrated to the lowest effective dose. Higher doses of ERT might result in a more rapid response, but longer-term individualised therapy results in similar levels of response. The committee was satisfied that using the efficacy data for ERT from ENCORE was appropriate.

The committee discussed the dose of eliglustat in the ENCORE trial. About 48% of patients in this trial had a dosage of eliglustat (150 mg twice daily) higher than that recommended in the summary of product characteristics. The committee was aware that efficacy data from ENCORE were used in the model, and was concerned that this reflected response to a higher dosage than in the marketing authorisation for eliglustat. The company stated that their pharmacokinetic/pharmacodynamic modelling suggested only minor differences in plasma levels with the higher dosage, and that it would be associated with a negligible difference in clinical response. The committee understood, however, that the basis for this modelling was the blood concentration data from the trials in which dose adjustments had been made in response to blood concentration measurements. Therefore, the predictions from the model could be subject to bias. However, the clinical expert confirmed that experience in practice has shown the continued efficacy of eliglustat at a dosage of 100 mg twice daily. The committee was satisfied that using the efficacy data for eliglustat from the ENCORE trial would not introduce major bias to the results.

The committee discussed the remaining uncertainties within the evidence base. It noted that the placebo-controlled ENGAGE study, which included a treatment-naive population, also allowed inclusion of people who had previously had ERT provided they had not had ERT within 9 months of recruitment to the trial. Additionally, there were no comparative data with ERT for patients who had not had previous treatment. The committee also noted that there were few data on patients with poor metaboliser status; most patients in the trials were intermediate and extensive metabolisers. The company submission stated that up to 7% of the Gaucher population are poor metabolisers. Following consultation, the company stated that its pharmacokinetic/pharmacodynamic modelling suggested that similar clinical outcomes are expected for poor metabolisers having the lower dose of eliglustat. The committee concluded that it would need to take these uncertainties into account in its decision-making.

The committee discussed the results from the key clinical trials. It noted that the ENCORE trial achieved the pre-specified non-inferiority measure for eliglustat compared with imiglucerase based on the composite primary endpoint (encompassing haemoglobin levels, platelet counts, spleen volume and liver volume). The committee noted that there was no direct comparison of eliglustat with velaglucerase alfa but recalled that it was considered to be equivalent to imiglucerase (see section 5.2). Also, the results from the ENGAGE study showed a statistically significant and clinically meaningful improvement in spleen volume with eliglustat. The committee heard from the clinical experts that they considered eliglustat to be equivalent, or very nearly equivalent, to ERT based on clinical measures such as haemoglobin levels and platelet counts, as well as in terms of how patients felt while having eliglustat. The patient experts stated that the option of an oral treatment with eliglustat was invaluable and most patients would consider treatment with eliglustat if it was available. Following consultation, the company presented 4‑year data from ENCORE showing that the outcomes remained stable. The clinical expert stated that this showed that the efficacy of eliglustat was independent of the residual effects of prior long-term ERT. The committee noted that eliglustat is a lifelong treatment and long-term benefits remained uncertain because these data are not based on a comparison with ERT. However, the committee accepted that the data were consistent with the possibility of long-term benefit. The committee concluded that eliglustat is an effective treatment for type 1 Gaucher disease, but remained concerned about the uncertainty of effectiveness in comparison with ERT in the long term.

The committee considered the adverse effects associated with eliglustat. It noted that headache, nausea, diarrhoea, flatulence and fatigue were common adverse reactions highlighted in the summary of product characteristics. The committee heard from the clinical experts that the stopping rate of about 2 to 3% seen in the trials was similar to that seen in clinical practice. It highlighted that stopping treatment was generally in response to lifestyle changes such as wanting to start a family. The committee understood that the adverse effects associated with eliglustat were acceptable to patients, especially in the context of the advantages of oral administration.

The committee noted that the main comparator for this evaluation was ERT. It also noted that, because NICE has not evaluated ERT, there was uncertainty about its benefits and value for money and, by extension, the benefits and value for money of eliglustat. The committee noted the statement from NHS England that the risks around value for money offered by ERT were lower for Gaucher disease compared with the risks for conditions such as Fabry disease. This is because it believed, in Gaucher disease, the effectiveness of ERT is well established and because the dose of ERT can be titrated to the lowest effective dose and the number of patients is lower. However, the committee was mindful that the benefits and value for money of ERT has not been formally considered. The committee noted that its considerations on the value for money of eliglustat were based on the current evidence and clinical practice, but that they would need to be reconsidered if ERT was no longer available in routine practice. The committee also encouraged the company, NHS England and treatment centres to collect more evidence, particularly on the longer-term benefits of eliglustat and ERT for treating type 1 Gaucher disease.

The committee discussed the company's cost–consequence model and the assumptions on which it was based. It noted that the model structure was complex but reflected the important health states. The committee discussed the key assumptions included in the company's economic model:

• In the absence of direct evidence comparing eliglustat with ERT in patients who had not previously had treatment, the company assumed that eliglustat and ERT have equal efficacy in such patients. The ERG stated that evidence from the ENCORE trial would have been more appropriate. Following consultation, the company stated that the mean treatment duration with ERT before entering ENCORE was about 10 years, so these data could not be generalised to people who had not previously had treatment. The company stated that its assumption of equivalence was supported by an indirect comparison (Ibrahim et al., 2016) on the basis of which the European Medicines Agency's Committee for Medicinal Products for Human Use stated that comparable results can be expected. The ERG agreed that using data from ENCORE was not ideal, but considered that it was superior to the company's approach. The company used data from ENGAGE to estimate transition probabilities for patients having eliglustat, and applied these to both treatment arms in the first cycle of the model. The ERG stated that this did not capture any potential differences between eliglustat and ERT. The committee agreed that both approaches had limitations. It heard that, because these transition probabilities were applied to the first cycle only, it had a very small impact on the results.

• The company assumed long-term equivalence of eliglustat and ERT, and the ERG highlighted that this had a considerable impact on estimated incremental quality-adjusted life years (QALYs). The committee agreed with the ERG that non-inferiority was not the same as equivalence, and that non-inferiority in the short term does not imply non-inferiority in the long term. The committee considered the 4‑year data presented by the company following consultation (see section 5.8) and also noted that the company presented varied approaches to transition within the model, resulting in a negligible impact on total QALYs gained. The ERG, however, clarified that the assumption of long-term equivalence was not underpinned by how transition probabilities are calculated, but by using the same probabilities in the long term across both arms of the model. The committee maintained that there was uncertainty around the assumption of equivalence in the long term.

• The dosage of ERT used in the model was 42.4 U/kg every 2 weeks, based on the mean dose of imiglucerase patients had in the ENCORE study. The committee recalled (see section 5.4) that a dose of between 15 U/kg and 30 U/kg was considered most reflective of clinical practice. The committee was aware that the dose of ERT was a key driver of costs and that the ERG had explored the impact of including a dose of 25 U/kg. The committee considered that the ERG exploratory analysis that included a dose of 25 U/kg was appropriate. Following consultation, the company stated that real world weight should also be factored into estimating the total administered dose (see section 4.58). The ERG clarified that that dose of ERT in the ERG analyses was obtained from English prescribing data reporting average units per month, so the average weight in the model was not relevant. However, the ERG presented exploratory analyses using estimates based on real world weight.

• The company assumed that the mortality risk does not increase with disease severity. The committee considered that this was an unrealistic assumption. It noted that the ERG explored the impact of increased mortality risk for patients in the 'marked' and 'severe' health states.

• The company assumed that there are no administration costs associated with eliglustat because it is an oral therapy. The ERG explored including a monthly dispensary cost for eliglustat but, following consultation, the company stated that eliglustat could be dispensed less frequently. The committee agreed with the ERG that there was uncertainty around the frequency and, because this had a minor impact on the results, the ERG's approach of including a monthly dispensary cost was pragmatic.

• The ERG highlighted that the administration costs for ERT were likely to be overestimated in the company's model because they were higher than the costs of hospital administration. The company stated that this would depend on the perspective of the costing analysis, but the ERG confirmed that all data available supported lower costs for home administration. The committee agreed that the ERG's exploration assuming equal cost was appropriate, and potentially overestimates the cost of ERT.
  
  The committee considered that these reflected important uncertainties in the model, but was satisfied that the ERG had presented results based on assumptions suitable for decision-making.

The committee discussed the utility increment used in the company's model for oral therapy, which it understood was the key driver of QALY benefits. It heard from the patient and clinical experts that the availability of an oral treatment would have a huge impact on health-related quality of life compared with an intravenous infusion. The committee took note of several patient testimonies describing the positive impact of an oral treatment and the potential this offered for them to return to a more normal life. The committee heard from the ERG that it agreed that oral therapy would provide a clear quality-of-life benefit but questioned the extent of the benefit assumed by the company, even though this was based on a vignette study. The ERG highlighted that an increment of 0.12 was substantial when compared with the decrements from significant adverse events and the benefits of other oral therapies estimated in previous NICE submissions. The committee was aware that the ERG explored an alternative utility increment of 0.05. The committee concluded that, although the true value was uncertain, the alternative value used by the ERG was more appropriate.

The committee noted the results of the company's cost–consequence model (see section 4.37). The committee agreed that there was considerable uncertainty around these estimates because of the assumptions discussed in sections 5.11 and 5.12. The committee considered the ERG's exploratory analysis around these assumptions (see section 4.52) represented more plausible incremental costs and benefits associated with eliglustat. The committee noted that incremental QALYs reduced from 2.28 to 1.06 for the treatment-experienced population and from 2.44 to 1.05 for the treatment-naive population when the ERG included all of its revised assumptions. When the ERG incorporated the confidential discounts available for eliglustat and ERT, eliglustat resulted in cost savings for both populations. Using real world weight in the estimation of the dose of ERT would increase the cost savings further. The committee was mindful of the uncertainty around the long-term impact of eliglustat but appreciated the important advantages of an oral treatment. The committee concluded that, taking into account the confidential discounts for ERT and eliglustat, eliglustat offered value for money compared with ERT for people with intermediate and extensive metaboliser status.

The committee noted the results of the company's cost–consequence model in the people with poor metaboliser status. The committee then discussed the results of the ERG's exploratory analysis for people with poor metaboliser status noting that these reflected the assumptions preferred by the committee. These ERG results indicated that eliglustat was cost saving and the cost savings increased further when the confidential discounts were included. The committee was concerned that the trials included very few people with poor metaboliser status, so questioned whether the results from the model could be generalised to this population. Following consultation, the company stated that its pharmacokinetic/pharmacodynamic modelling indicated that no differences in clinical outcomes were expected at a dosage of 100 mg twice daily for eliglustat. The committee was mindful that very few patients had poor metaboliser status and that an oral treatment option was valued by patients. The committee concluded that eliglustat offered value for money in people with poor metaboliser status.

The committee considered whether it should take into account the consequences of the Pharmaceutical Price Regulation Scheme (PPRS) 2014, and in particular the PPRS payment mechanism, when evaluating eliglustat. It noted NICE's position statement about this, and accepted the conclusion 'that the 2014 PPRS payment mechanism should not, as a matter of course, be regarded as a relevant consideration in its assessment of the cost effectiveness of branded medicines'. The committee heard nothing to suggest that there was any basis for taking a different view about the relevance of the PPRS to this evaluation of eliglustat. It therefore concluded that the PPRS payment mechanism was irrelevant in considering the value for money offered by eliglustat.

The committee discussed the estimated uptake of eliglustat over a 5‑year period. It noted that the company had revised the estimates in its original submission based on its experiences in European countries. The company explained that the uptake was shown to be lower in the first year and the company had adjusted for this in its revised estimates. The estimates are deemed commercial in confidence by the company and cannot be reported here. However, the clinical experts confirmed that, while the uptake in England was expected to be higher in the first 2 years compared with the company's estimates, the overall 5‑year estimated uptake was reasonable. The committee was satisfied that the company's revised estimates sufficiently reflected the expectations in clinical practice in England.

The committee discussed the company's budget impact analysis. It was aware that it was based on estimates of total costs generated by the cost–consequence model, but the company also made some additional assumptions (see section 4.40). The committee considered the assumptions in the company's budget impact analysis. It noted that the company assumed that patients with a new diagnosis would start treatment on eliglustat rather than ERT, and that dosing for ERT was based on the ENCORE trial (that is, 42.4 U/kg every 2 weeks), which was not reflective of clinical practice (see section 5.4). The committee also noted that the model was based on people who were intermediate and extensive metabolisers, so excluded poor metabolisers, which would have overestimated the total costs of eliglustat for patients eligible for treatment. The ERG also highlighted issues related to incorporation of mortality and stopping treatment from the cost–consequence model (see sections 4.47 and 4.48), which the committee agreed would have underestimated the budget impact of eliglustat. The committee concluded that the company's estimates of budget impact were too uncertain, and so it considered the ERG's exploratory budget impact analyses in its decision-making.

The committee discussed the ERG's exploratory analyses of the budget impact analysis. It noted that the ERG revised several assumptions that were the same as its exploratory analysis of the company's cost–consequence model, with the additional assumptions of zero mortality, no treatment stopping, and that 4% of eliglustat patients were poor metabolisers. The committee was satisfied that these explorations reflected the committee's preferences. Following consultation, the company stated that it was inappropriate to exclude mortality because any deaths would mean the NHS is no longer paying for treatment. The ERG, however, considered that the company's approach potentially double counted mortality and preferred to exclude mortality and stopping treatment from the cost–consequence model and only include it in the budget impact model. The committee considered that, while approaches could differ, it was important that the approach used was internally consistent and did not double count the impact of mortality on budget impact. The committee was also aware that this had a negligible impact in the model and was content to consider the ERG's results. The committee understood that taking into account the confidential discounts available for eliglustat and ERT, eliglustat resulted in cost savings compared with ERT.

The committee noted that, because eliglustat is an oral therapy, it would give people the freedom to travel and attend university, and remove the need for people to take time off work for intravenous infusion appointments. It heard that the drug would be associated with important indirect mental health benefits because it allows people to live a more normal life. The committee concluded that eliglustat is likely to have a significant impact on people's lives beyond its direct health benefits.

The committee noted that, although eliglustat is an oral therapy, it will be important for people to have the drug started and to be monitored in expert centres. The committee understood from the company submission that no additional development or staff training above what is already in place for the provision of care will be needed in relation to eliglustat. The committee heard from the clinical experts that the availability of eliglustat will reduce the need for the nursing support that is often needed for home infusions of ERT, and patient experts highlighted the burden on specialist centres of running homecare services. The committee concluded that the impact of eliglustat on the delivery of specialised services is likely to be relatively negligible.

The committee understood that type 1 Gaucher disease can be a debilitating condition that has severe effects on the lives of people with the condition, and their families and carers. It agreed that there was uncertainty about the equivalence of eliglustat compared with ERT in the long term. However, the committee considered that, because it is an oral treatment, it could potentially provide important quality-of-life benefits for people currently having intravenous ERT, as well as for people who have not previously had treatment. Together with its consideration that eliglustat was cost saving compared with ERT, the committee concluded that it could recommend eliglustat, within its marketing authorisation, for treating type 1 Gaucher disease when ERT would otherwise be offered.