Ruxolitinib for treating disease-related splenomegaly or symptoms in adults with myelofibrosis

Clinical effectiveness

4.3 The committee considered the evidence presented by the company on the clinical effectiveness of ruxolitinib. It noted that the company had presented 2 randomised controlled trials (RCTs), COMFORT‑I and COMFORT‑II, which evaluated the efficacy of ruxolitinib in patients who had intermediate‑2 or high-risk myelofibrosis (its main source of evidence), and supportive evidence from 4 non-RCT studies of ruxolitinib in patients with intermediate‑1 risk myelofibrosis or a low platelet count (ROBUST, JUMP, study 258 and EXPAND). The committee was aware that the COMFORT trials had also been the main source of evidence for NICE's previous technology appraisal guidance on ruxolitinib for disease-related splenomegaly or symptoms in adults with myelofibrosis, but that longer-term data from these trials (COMFORT‑I median follow‑up 3 years, COMFORT‑II median follow‑up 3.5 years) had become available since the guidance was published.

4.4 The committee discussed the relationship between the marketing authorisation for ruxolitinib and the populations in the COMFORT trials and the 4 non-RCT studies. It noted that the COMFORT trials included only patients who had intermediate‑2 or high-risk myelofibrosis with platelet counts over 100×10⁹/litre, but that the marketing authorisation was not defined by risk categories or platelet count. The COMFORT trials only covered a subset of the population covered by the marketing authorisation. The committee noted that the 4 non-RCT studies included patients with intermediate‑1 and intermediate‑2 risk myelofibrosis or with platelet counts of 50–100×10⁹/litre, and noted that these studies provided some evidence for the use of ruxolitinib in a subgroup of patients who were not included in the COMFORT trials but are included in the marking authorisation for ruxolitinib. The committee concluded that data from the COMFORT trials and the 4 non-RCT studies should be considered, because the data were obtained from populations that are covered by the marketing authorisation for ruxolitinib and so are relevant for decision-making. However, it noted that the company had restricted its economic assessment to the population in the COMFORT‑II trial (see section 4.9) and so the committee would use the other studies principally as corroborative evidence.

4.5 The committee considered the generalisability of the results from the COMFORT trials and the 4 non-RCT studies. It heard from the clinical experts that ruxolitinib would mostly be used in higher-risk people who had splenomegaly or symptoms. The committee was also aware that the COMFORT trials did not include patients with low platelet counts (under 100×10⁹/litre) but that 2 of the non-RCT studies (study 258 and EXPAND) included patients with platelet counts of between 50 and 100×10⁹/litre. It heard from the clinical expert that clinicians would treat people with a platelet count of more than 100×10⁹/litre with ruxolitinib (which is reflective of the population in the COMFORT trials) and also people with a platelet count of 50–100×10⁹/litre (which is reflective of the population in study 258 and EXPAND), because this is consistent with the summary of product characteristics for ruxolitinib. The clinical experts stated that clinicians may occasionally treat people with platelet counts below 50×10⁹/litre after careful consideration and informed discussion with the person about the benefits and risks of ruxolitinib, because the summary of product characteristics for ruxolitinib does not provide dosing recommendations for this population. The committee concluded that the results from the COMFORT trials and the non-RCT studies were generalisable to the people who would have treatment with ruxolitinib in England; that is, people with intermediate‑2 or high-risk myelofibrosis.

4.6 The committee noted that COMFORT‑II was the only study included in the company's submission with an active treatment group and discussed whether the comparator group (best available therapy) was relevant to clinical practice in England. The committee noted the evidence review group's (ERG) concerns that the selection of treatments that made up best available therapy in the trial included lenalidomide (but was replaced with thalidomide in the revised model, based on the British Committee for Standards in Haematology guideline for investigation and management of myelofibrosis [2012]). The committee heard from the clinical experts that lenalidomide is rarely used in clinical practice in England. It heard that hydroxycarbamide was the main treatment currently used in clinical practice, but people with myelofibrosis are a heterogeneous group and therefore treatments would often be adapted to individual patient needs. The committee concluded that the treatments used in the best available therapy group (including thalidomide but without lenalidomide) in COMFORT‑II were clinically relevant.

4.7 The committee considered the clinical-effectiveness evidence for ruxolitinib on spleen size and spleen volume. It noted that the COMFORT trials showed that ruxolitinib provided significant benefits for reduction in spleen size and spleen volume. The committee also noted that the results from the 2 non-RCT studies (ROBUST and JUMP) were generally consistent with the results from the COMFORT trials and that the results were similar between patients with intermediate‑1 and high-risk myelofibrosis (although the number of patients in the different risk subgroups was low). The committee was aware that there was no direct association between spleen size and symptoms and that a person could have a modest size spleen with severe symptoms or a large spleen with minimal symptoms. It noted that COMFORT‑I also assessed symptom reduction, and that the results showed a clinically meaningful improvement in myelofibrosis-associated symptoms for patients who had treatment with ruxolitinib compared with a worsening of symptoms for patients who had placebo. The committee was aware that the results from 2 of the non-RCTs (ROBUST and JUMP) also showed symptom reduction with ruxolitinib and that the results were similar between patients with intermediate‑1 risk and high-risk disease (although the number of patients in the different risk subgroups was low). The committee was aware of the emphasis that the patient experts placed on symptoms in myelofibrosis (see section 4.1) and concluded that symptoms (especially itch and fatigue) and spleen size were both important outcomes to consider and that ruxolitinib was effective in reducing spleen size and relieving symptoms in people with intermediate‑1, intermediate‑2 and high-risk myelofibrosis. It therefore concluded that ruxolitinib was a clinically effective treatment for disease-related splenomegaly or symptoms in adults with myelofibrosis.

4.8 The committee considered the overall-survival data. It was aware that the long-term data (median follow‑up 3.5 years) from COMFORT‑II showed a statistically significant difference in overall survival for ruxolitinib compared with best available therapy, using both the intention-to-treat analysis and the analysis adjusting for crossover. It noted the hazard ratios, which after adjusting for crossover, strongly indicated a survival benefit for ruxolitinib (see section 3.2). The committee therefore concluded that there was sufficient evidence to show that ruxolitinib increased overall survival compared with best available therapy.

4.9 The committee considered the adverse events associated with ruxolitinib. It noted that the company had presented long-term data on adverse events from the COMFORT trials and supporting data from the 4 non-RCT studies. The committee accepted that ruxolitinib was generally well-tolerated and that haematological adverse events were common with ruxolitinib. It heard from the patient experts that the adverse events reported with ruxolitinib were considered manageable by patients. The committee heard from the clinical experts that haematological outcomes (for example, anaemia and thrombocytopenia) are important in managing myelofibrosis. It was aware that ruxolitinib dose reductions rather than transfusions were the main means of treating haematological problems and heard from the clinical experts that the rate of blood transfusions would be equivalent for ruxolitinib and other available treatments for myelofibrosis in clinical practice. The committee concluded that ruxolitinib did have a negative impact on haematological outcomes in the short term for people with myelofibrosis, but agreed that these were manageable.

Cost effectiveness

4.10 The committee discussed the company's general approach to developing its economic model. It noted that the ERG considered the company's approach to be well presented and appropriate. It also noted the ERG's comments that the data used in the model was obtained mainly from COMFORT‑II and therefore the cost-effectiveness estimates obtained from the model were specific to a population with intermediate‑2 or high-risk myelofibrosis. The committee acknowledged that the population in the company's economic model was only a subset of the population covered by the marking authorisation for ruxolitinib (see section 4.3), but agreed that the company's model was acceptable for assessing the cost effectiveness of ruxolitinib only for people with intermediate‑2 or high-risk myelofibrosis. It also acknowledged that people with intermediate‑1 risk myelofibrosis were not included in the model and therefore it was not able to consider this subgroup in its decision making. The committee also considered the revised model submitted by the company after consultation on the appraisal consultation document and noted that the revised assumptions (see section 3.52) were in line with the ERG's critique of the original model.

4.11 The committee discussed whether the company's assumption of no drug wastage for ruxolitinib was appropriate. It noted that the ERG had provided exploratory analyses on the original model, which allowed for 5%, 10% and 15% wastage of ruxolitinib. The committee heard from the clinical experts that the company's assumption of no drug wastage for ruxolitinib reflected drug usage in clinical practice. It agreed that the ERG's exploratory analyses allowing significant drug wastage for ruxolitinib were not representative of clinical practice. The committee discussed whether the drug costs for patients having treatment with ruxolitinib used in the economic model reflected the drug costs for ruxolitinib in clinical practice. It was aware that the drug costs for ruxolitinib were estimated from the starting doses as defined in the summary of product characteristics for ruxolitinib and the actual dose usage in COMFORT‑II. The committee heard from the clinical experts that it was difficult to estimate the drug costs for the 'average' patient seen in clinical practice because the dosage used varied between people and depended on several factors such as platelet count, response to treatment and adverse events. The committee agreed that there was some uncertainty over whether the drug costs for ruxolitinib used in the economic model reflected the drug costs for ruxolitinib in clinical practice, but agreed that the drug costs used were appropriate because they were based on the same trial data on which the effectiveness inputs were based.

4.12 The committee discussed the company's revised economic model that included an updated patient access scheme for ruxolitinib, and the ERG's critique of the submission. The committee noted the company's revised base-case incremental cost-effectiveness ratio (ICER) result for ruxolitinib compared with best available therapy of £31,200 per quality-adjusted life year (QALY) gained. It also considered the scenario analyses presented by the ERG, based on the revised cost-effectiveness model from the company. It considered that the assumptions of Scenario 1 and Scenario 2 (see section 3.55) are clinically plausible, but noted the ERG's concerns about the uncertain evidence on the upgrade of baseline utility value. The committee agreed that the estimated ICER for ruxolitinib was largely robust to a range of values and model assumptions (based on sensitivity analyses conducted on the original model). The committee did not, however, favour factoring in a modification of the ICERs to reflect carers' quality of life. Although it agreed that carers' health could be affected by caring, it did not consider the results robust; nor did it consider that myelofibrosis stood out amongst severe illnesses in having a more profound carer burden. Most importantly, the committee concluded that the scenario proposed by the company did not take into account the opportunity cost of carers' burden (that is, the carers' burden relieved by other treatments currently available in the NHS that ruxolitinib might displace).

4.13 The committee considered the most plausible ICER for people with intermediate‑2 or high risk myelofibrosis. It also considered it important to evaluate the cost effectiveness of ruxolitinib separately for the different subgroups (intermediate‑2 and high-risk myelofibrosis) because of the different prognosis and response to therapy in these groups. It was aware of the uncertainty in the values in section 3.42 and noted the ERG's critique of the additional analyses in section 3.57. It considered that the subgroup analysis provided by the company was appropriate. The committee noted that the ICER for the intermediate‑2 subgroup was £26,000 per QALY gained, which is within the range normally considered as cost effective. The committee therefore considered ruxolitinib to be a cost-effective option for treating disease-related splenomegaly or symptoms in intermediate‑2 risk myelofibrosis. It also noted that the ICER for the high-risk subgroup was £38,000 per QALY gained.

4.14 Because the ICER for people with high-risk myelofibrosis was above £30,000 per QALY gained, the committee discussed whether ruxolitinib fulfilled the criteria for a life-extending, end-of-life treatment. The committee considered the supplementary advice from NICE that should be taken into account when appraising treatments that may extend the life of patients with a short life expectancy and that are licensed for indications that affect small numbers of people with incurable illnesses. For this advice to be applied, all the following criteria must be met:

4.15 The committee discussed the criterion of a small patient population. It accepted the estimates in the company's submission that 1,185 people are estimated to be living with myelofibrosis in England and would be eligible for treatment with ruxolitinib for disease-related splenomegaly or symptoms associated with myelofibrosis. The committee concluded that the eligible population for England did not exceed 7,000 and that ruxolitinib met the end-of-life criterion for a small patient population.

4.16 The committee discussed the criterion of extension to life of more than an average of 3 months. It noted that because median overall survival was not reached in the ruxolitinib group it was not possible to calculate the median (or mean) survival benefit associated with ruxolitinib compared with best available therapy in the COMFORT‑II trial. However, it noted the results of an indirect comparison analysis between the ruxolitinib treatment group of COMFORT‑II and the Dynamic International Prognostic Scoring System (DIPSS) cohort. It noted that this analysis produced estimates of median survival of 5 years from diagnosis on ruxolitinib compared with 3.5 years for the DIPSS cohort. The committee concluded that treatment with ruxolitinib provided an extension of life of more than an average of 3 months.

4.17 The committee discussed whether people with disease-related splenomegaly or symptoms associated with myelofibrosis would be expected to have a mean life expectancy of less than 24 months. It was aware that median overall survival in the best available therapy group of COMFORT‑II was 28 months in people with intermediate‑2 or high-risk disease. The committee gave further consideration to the range and relevance of the evidence available on the expected survival of people with high-risk disease from the various prognostic scoring systems (International Prognostic Scoring System [IPSS] for primary myelofibrosis, DIPSS and DIPSS‑plus). It noted that the company's submission reported that median survival using the various prognostic scoring systems varied from a median of 1.3 to 2.3 years for people with high-risk myelofibrosis. The committee acknowledged that there was some uncertainty about the life expectancy of people with myelofibrosis but agreed that the various prognostic scoring systems provided the best available evidence because the data were based on patients before they had any treatment. The committee considered whether the life expectancy of patients with high-risk myelofibrosis met the end-of-life criterion of less than 24 months and was persuaded that the life expectancy for people with high-risk myelofibrosis was likely to be less than 24 months. The committee therefore concluded that it had been provided with evidence that high-risk patients met all of the end-of-life criteria and considered ruxolitinib to be a cost-effective option for treating disease-related splenomegaly or symptoms in high-risk myelofibrosis.

4.18 The committee considered whether ruxolitinib is an innovative treatment. It agreed that ruxolitinib provided a step change in treating splenomegaly and symptoms in people with myelofibrosis. The committee acknowledged that ruxolitinib is a targeted treatment and manages symptoms for which there is currently no available treatment. Therefore, the committee agreed that ruxolitinib is innovative, but there were no additional gains in health-related quality of life over those already included in the QALY calculations.

4.19 The committee was aware of NICE's position statement on the Pharmaceutical Price Regulation Scheme (PPRS) 2014, and in particular the PPRS payment mechanism, 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 is any basis for taking a different view with regard to the relevance of the PPRS to this appraisal. It concluded that the PPRS payment mechanism was not relevant for its consideration of the cost effectiveness of any of the technologies in this appraisal.

Summary of appraisal committee's key conclusions