The Appraisal Committee reviewed the data available on the clinical and cost effectiveness of romiplostim, having considered evidence on the nature of chronic ITP and the value placed on the benefits of romiplostim by people with the condition, those who represent them, and clinical specialists. It also took into account the effective use of NHS resources.
The Committee considered the nature of the condition, and noted evidence submitted and presented by the patient experts and clinical specialists on the clinical signs and symptoms associated with chronic ITP. The Committee heard that the signs and symptoms associated with low platelet counts vary, and that bleeding and bruising can have a considerable impact on the daily activities of people with chronic ITP, may attract a social stigma associated with the appearance of bruises, and can limit lifestyle choices. The Committee heard that many people with ITP experience fatigue, but that there is no clear relationship between fatigue and platelet count or haemoglobin concentration. The Committee understood that anxiety about the risk of bleeding can affect a person's quality of life and the ability to work, travel and/or undertake leisure activities. The Committee understood from patient experts that a bleed could result in a person seeking medical care to receive rescue therapies, and if the bleeding was severe the person could need hospitalisation.
The Committee discussed the clinical management of chronic ITP. It noted that the pathway of care for chronic ITP varies depending on the person's circumstances, and that no single standard treatment pathway is used in routine practice. The Committee heard from the clinical specialists that, in the UK, first-line treatment for chronic ITP is considered to be corticosteroids (or intravenous immunoglobulin for people for whom corticosteroids are contraindicated), also referred to as rescue therapy. The clinical specialists estimated that approximately 30% of people would enter remission after such first-line treatment. The Committee heard from the clinical specialists that, for chronic ITP that does not respond to rescue therapy, active treatments are considered as second-line treatment including rituximab, immunosuppressive agents (azathioprine, mycophenolate mofetil, ciclosporin), danazol, dapsone, and cytotoxic agents (cyclophosphamide, vinca alkaloids). The clinical specialists explained that clinicians increasingly prescribe rituximab as the first choice of active treatment (for approximately 50% to 60% of patients who need active treatment), and this leads to remission in approximately 50% of people treated. However, the clinical specialists noted that these people will, in general, eventually relapse and need further treatment. The clinical specialists also stated that azathioprine would be used for people whose condition is refractory to rituximab or who are intolerant of rituximab, but that cyclophosphamide and ciclosporin were considered too toxic, and that people do not tolerate vinca alkaloids and danazol well and were considered unlikely to benefit from them. The Committee understood that people receiving active treatments for ITP would need monitoring and would still be likely to need rescue therapies from time to time.
The Committee understood that people who have certain treatments (intravenous immunoglobulin, corticosteroids and immunosuppressive agents) over long periods of time will experience adverse effects that can lead to chronic conditions, including infections, diabetes mellitus, an increased risk of gastrointestinal bleeds, and hypertension. The Committee heard that some people may need knee or hip replacements as a result of the side effects of long-term use of corticosteroids. The Committee understood from comments made by consultees and commentators in response to the first ACD that the consequences of treatment with alkylating agents (such as cyclophosphamide and vinca alkaloids) can include malignancies and fertility problems.
The Committee discussed when a person with chronic ITP would receive treatment and heard from the clinical specialists that a person's platelet count alone does not determine whether or not he or she receives treatment, and that clinicians take into account the person's symptoms (such as fatigue) and risk of bleeding. Overall, the Committee understood from the clinical specialists that the lower a person's platelet count, the more likely a clinician is to offer treatment, but that treatment is not usually offered unless absolutely necessary because of the side effects of many of the current treatments. Active treatments are more likely to be considered for people with severe symptoms or who are at high risk of bleeding, and in particular when rescue therapies are failing to produce satisfactory platelet counts or relief of symptoms. The Committee heard that people with a platelet count of between 20×109 and 30×109 per litre would rarely be offered an active treatment in the UK unless they had a significant bleed. The clinical specialists and patient experts informed the Committee that in clinical practice a platelet count of between 10×109 and 20×109 per litre would be the level at which clinicians would be likely to begin active treatment. A separate group of people who may be considered for active treatments at higher platelet counts would be those, for example, who needed aspirin therapy for cardiovascular disease. For people without severe symptoms and who are not perceived to be at high risk of bleeding, the preferred approach would be a strategy of 'watch and rescue' (with 'rescue' intervention dictated by the frequency of bleeding episodes) rather than an active treatment. The Committee heard from the clinical specialists that only people who are classed as having severe ITP would receive an active treatment, and this accounts for approximately 5% to 10% of people with chronic ITP. The Committee concluded that, in UK clinical practice, treatment for people with chronic ITP is dictated by the severity of symptoms, in particular bleeding.
The Committee discussed the place of romiplostim in the pathway of care for people with chronic ITP, and considered the appropriate comparators. It noted that the licensed indication for romiplostim for people who have undergone splenectomy is restricted to people who are refractory to other treatments such as corticosteroids and immunoglobulin, and also that romiplostim may be considered as a second-line treatment for people with a contraindication to splenectomy. The Committee understood that there are few treatments licensed for the treatment of chronic ITP. The Committee heard from the clinical specialists that the place of romiplostim in clinical practice would be for people whose condition is refractory to rituximab, or who are intolerant of rituximab.
The Committee considered the evidence presented by the manufacturer on the clinical effectiveness of romiplostim compared with placebo and standard care. It was mindful that the evidence was mainly derived from 2 small placebo-controlled RCTs. The Committee noted that romiplostim significantly improved platelet count (the response measure used in the trials) and reduced the frequency of bleeding – particularly the occurrence of moderate and severe bleeding episodes. However, the Committee was mindful that these 2 RCTs did not provide clear evidence about the relative effectiveness of romiplostim compared with the active comparator treatments listed in the scope for the appraisal. The Committee noted that the manufacturer took a pragmatic rather than a systematic approach when collecting evidence for comparator treatments because of heterogeneity among studies of the comparators and because many of the comparators do not have a marketing authorisation for the treatment of ITP. The Committee also noted that data related to certain outcomes, such as the time to failure and the mean response time for romiplostim, had been generated by the manufacturer from a non-comparative open-label study. The Committee acknowledged comments received from consultees and commentators in response to consultation on the first ACD that highlighted the difficulty of conducting comparative trials with romiplostim, mainly because ITP is a rare and heterogeneous disease. The Committee heard from the clinical specialists that the population in the trials had the most severe ITP, which is estimated as representing 1% to 4% of people with chronic ITP, and were at high risk of bleeding and therefore needed high levels of rescue therapies throughout the trial. The Committee was aware of the limitations of the evidence on clinical effectiveness provided by the small RCTs, but considered that the available data demonstrated that romiplostim was clinically effective in people with severe ITP who are at high risk of bleeding and need repeated and frequent courses of rescue therapies.
The Committee then discussed the adverse effects associated with romiplostim. It noted that very few patients treated with romiplostim in the trials experienced adverse effects, including bleeding. The Committee heard from the clinical specialists that romiplostim may have benefits over other active treatments because it produces a sustained platelet response during treatment, it can be continued for a longer time than other active treatments, it can be used in a wider population, and because adverse effects limit both the use and duration of other active treatments. The Committee noted from comments received from consultees and commentators in response to consultation on the first ACD that romiplostim has a different mode of action from current treatments for chronic ITP. The clinical specialists stated that romiplostim represents a 'step-change' for the treatment of ITP, in that it does not have immunosuppressive properties. The Committee also considered comments from consultees and commentators in response to consultation on the second ACD about setting a stopping rule for romiplostim, but noted that the modelling of the cost effectiveness for romiplostim did not include analyses addressing stopping. The Committee was also aware that the SPC specifies conditions under which to stop treatment with romiplostim. The Committee concluded that romiplostim has a novel mechanism of action (as a thrombopoietin receptor agonist) and a good adverse-effect profile, particularly in comparison with currently available treatments.
The Committee next considered the revised economic model submitted by the manufacturer that included the patient access scheme. The Committee agreed with the new approach taken by the manufacturer, wherein the comparator arm of the model started with active treatments instead of 'watch and rescue' and the costs of bone marrow tests and blood film assessment were included in the cost of treatment with romiplostim. The Committee noted that it would have been preferable to see romiplostim modelled after rituximab, which is where the clinical specialists currently position romiplostim in the treatment pathway for ITP, but recognised that this was not the case at the time of the consultation on the first ACD. The Committee considered the ERG's comments on the manufacturer's revised base-case analysis that included the patient access scheme, and noted that the ERG considered the approaches taken by the manufacturer in the revised analyses to being generally reasonable.
The Committee considered the ICERs in the manufacturer's revised base-case analysis that included the patient access scheme. For non-splenectomised patients the ICERs were £24,800 and £28,300 per QALY gained for the realistic and conservative scenarios respectively. For splenectomised patients, the ICERs were £4,620 and £16,500 per QALY gained for the realistic and conservative scenarios respectively.
The Committee noted that the median dose of romiplostim used in the RCTs was 3 micrograms/kg for splenectomised patients and 2 micrograms/kg for non-splenectomised patients, and heard from the clinical specialists that these doses reflect clinical practice. The Committee discussed the number of vials used in the realistic and conservative scenarios presented by the manufacturer. It heard from the clinical specialists that although they would not use doses as high as those given in the trials, there may be some instances where the entire contents of a vial of romiplostim would be used, rather than wasting some, and so the dose may occasionally exceed the maximum dose specified in the SPC. Therefore, the Committee concluded that the most plausible scenario would be somewhere between the realistic and conservative scenarios modelled by the manufacturer.
The Committee noted that in clinical practice it would be rare for clinicians to use doses of romiplostim that were aimed at obtaining a platelet count above 50×109 per litre (as was the case in the trials). Therefore, in practice, aiming for a lower target platelet count would mean less frequent use of romiplostim, and lower doses of romiplostim when it is used. The Committee noted that, in the ERG's exploratory analyses, the ICERs were sensitive to a change in the number of vials used, and concluded that romiplostim would be more cost effective if less romiplostim was used in clinical practice than was assumed in the model.
The Committee considered the estimates of the effectiveness of the comparators assumed in the model, and heard from the clinical specialists that these estimates seemed reasonable and reflected clinical practice. The Committee noted that the model did not sufficiently account for long-term adverse effects of some of the comparators. The Committee concluded that had the model incorporated these adverse effects, the ICERs of romiplostim would have been lower.
The Committee noted that the submitted model assumed that treatment with romiplostim extended the life of people with chronic ITP by 2.92 and 2.03 years for the non-splenectomised and splenectomised groups respectively. The Committee noted that the survival advantage associated with romiplostim resulted from avoiding severe life-threatening bleeds. The Committee understood from the comments of consultees and commentators that observational evidence supported the association between low platelet counts and an increased risk of dying. The Committee heard from the clinical specialists that this assumption was plausible, and that an additional mortality benefit could result from reducing the number of deaths associated with the long-term adverse effects of the comparator treatments. The Committee concluded that although there were no robust data supporting the ability of romiplostim to reduce the risk of dying compared with standard care in people with chronic ITP, this possibility was plausible.
The Committee noted the ERG's concern that if it was assumed that romiplostim was no longer effective in the patients who were lost to follow-up during the trial period, the ICERs increased markedly. The Committee heard from the manufacturer that of the 31 patients who withdrew during the open-label extension study, only 2 patients (6%) withdrew because they did not respond to romiplostim. The Committee concluded that the patients who were lost to follow-up were neither more nor less likely to have stopped responding to romiplostim than patients who were not lost to follow-up.
The Committee noted that the use of intravenous immunoglobulin in 'watch and rescue' care was a major driver of costs in the comparator arm. The Committee noted that the manufacturer had used data from the clinical trials to model the costs and use of resources by assuming that a certain proportion of patients would need rescue therapies during each monthly cycle in the model. The Committee noted that in the ERG's 1-way sensitivity analyses, the use of rescue therapies had the greatest impact on the ICER. When the use of rescue therapies was reduced to 80% of the base case in both the comparator and romiplostim arms, the ICER for romiplostim for the realistic scenario increased from £4,620 to £32,200 per QALY gained for the splenectomised patients, and from £24,800 to £35,200 per QALY gained for the non-splenectomised patients. The Committee heard from the clinical specialists that the use of rescue therapies was higher in the RCTs than in UK clinical practice on average. This reflected the disease severity of the patients in the RCTs and was consistent with 'severe refractory ITP'. The clinical specialists agreed with the ERG's estimates of time spent on each treatment in the model and stated that people with severe refractory ITP could potentially be on 'watch and rescue' for 18 out of 20 years, and receive rescue treatment as frequently as in the romiplostim RCTs (that is, on 'watch' for 4 months and on rescue treatment for 8 months in each year). The Committee concluded that the majority of people with chronic ITP would not receive rescue therapies as frequently as this, and that the manufacturer's revised base-case ICERs would be relevant only for the treatment of patients at the very severe end of the spectrum of ITP. However, the Committee noted that there remained some uncertainty over the manufacturer's revised base-case ICERs, because even a small reduction of the amount of rescue treatments would increase the ICERs.
The Committee considered the comments received from consultees and commentators in response to consultation on the second ACD. The consultees and commentators requested that the Committee clarify the terms 'specialist in haematology' and 'standard active treatments', and define the dosages of unlicensed treatments. In response to the first point, the Committee recommended that the wording be changed to 'haematologist'. The Committee believed that a haematologist would have a good understanding of what is meant by standard active treatments for chronic ITP, and of dosages. The Committee did not agree that recommending an arrangement of shared care with general practitioners, as proposed by 1 commentator, was appropriate at this time. Consultees and commentators also requested that a registry is set up to collect data on the use of romiplostim, in order to monitor adverse events, and to audit the implementation of this guidance on romiplostim. The Committee was aware of the UK ITP registry and supported the collection of data on treatment with romiplostim. The Committee furthermore concluded that these data would be useful for any future appraisal of romiplostim for the treatment of chronic ITP.
The Committee agreed that a starting point for the most plausible ICERs would lie between the ICERs for the manufacturer's realistic and conservative scenarios, and may be slightly higher to account for any small reduction in the amount of rescue therapy seen in clinical practice compared with the romiplostim RCTs. Therefore, the Committee concluded that the ICERs would be under £20,000 per QALY gained for the treatment of splenectomised patients, and around £30,000 per QALY gained for the treatment of non-splenectomised patients. In addition, the Committee concluded that 2 factors would reduce these ICERs: the potentially lower use of romiplostim in clinical practice when clinicians aim for target platelet counts lower than those in the romiplostim RCTs, and the fact that the model excluded the long-term adverse effects of the comparator treatments. The Committee was also aware that since the publication of the second ACD, the marketing authorisation of romiplostim had changed such that the platelet count at which the dose of romiplostim can be reduced had been lowered. The Committee appreciated that this would be likely to reduce the ICERs for romiplostim. Therefore, the Committee concluded that romiplostim is recommended as a cost-effective use of NHS resources for the treatment of adults with chronic ITP whose condition is refractory to standard active treatments and rescue therapies, or who have severe disease and a high risk of bleeding that needs frequent courses of rescue therapies, and if the manufacturer makes romiplostim available with the discount agreed as part of the patient access scheme. The Committee heard from clinical specialists that approximately 1% to 4% of the UK population with chronic ITP would be eligible for treatment with romiplostim using these criteria. Because these people have severe ITP, the Committee concluded that only a haematologist should start and supervise treatment with romiplostim.
The Committee considered whether its recommendations raised any equality issues for people with chronic ITP. The Committee was aware that certain religious groups would not consent to the use of blood products, and also that ITP might affect pre-menopausal women more than men. It also understood that romiplostim might reduce the burden of hospital admission for long hours to receive intravenous immunoglobulin, especially for people for whom it is difficult to travel to a hospital. The Committee noted that no specific representations had been made for these groups of people. The Committee concluded that its recommendations do account for the individual needs of people to receive romiplostim, and do not make it more difficult for any particular group to access treatment with romiplostim compared with any other group.