4 Consideration of the evidence
4.1 The Appraisal Committee reviewed the data available on the clinical and cost effectiveness of ustekinumab, having considered evidence on the nature of plaque psoriasis and the value placed on the benefits of ustekinumab by people with the condition, those who represent them, and clinical specialists. It also took into account the effective use of NHS resources.
4.2 The Committee discussed the likely place of ustekinumab in the management of severe plaque psoriasis. It heard from the clinical specialists that there has been a substantial reduction in hospital admissions for psoriasis as a result of the increasing availability of biological therapies. However, the Committee heard from the clinical specialists that there are currently no treatments that they considered to be effective for people whose psoriasis does not respond adequately to the tumour necrosis factor (TNF) inhibitors (that is, adalimumab, infliximab and etanercept). In addition, with the withdrawal of efalizumab there are no treatment options for people in whom TNF inhibitors are contraindicated, such as people with heart failure or demyelinating disease. The Committee noted that ustekinumab has a different mechanism of action from that of the TNF inhibitors, and heard that the clinical specialists considered that its mechanism of action may be specific in the management of psoriasis. The Committee understood that ustekinumab would be considered to be of value by people with psoriasis and their clinicians.
4.3 The Committee heard from the clinical specialists and patient experts that ustekinumab may be easier to use than other biological therapies because it is administered subcutaneously just once every 12 weeks after the first 4 weeks. This could enable people to be given the drug during their routine scheduled clinic visits. The Committee was informed by the patient experts that people with psoriasis do not generally have a problem with the frequency of injections, although they prefer less frequent injections. The Committee accepted that the less frequent dosing for ustekinumab, which would allow it to be given during routine scheduled clinic visits, may also help compliance.
4.4 The Committee heard from the clinical specialists that ustekinumab is a new drug that has been given to far fewer people than the other biological therapies, and therefore its long-term safety profile is less certain. Because of this, the specialists considered that the drug may initially be prescribed more cautiously than existing treatments. The Committee also heard from the clinical specialists and patient experts that people with severe psoriasis are often well informed about drug safety and able to consider benefits and risks before starting treatment.
4.5 The Committee considered that the RCTs identified in the manufacturer's submission confirmed the clinical effectiveness of ustekinumab compared with placebo in people with moderate to severe psoriasis. The Committee also considered that ustekinumab had been demonstrated to be more clinically effective than etanercept. It noted, however, that the dosage used for etanercept in the comparative trial was different from that currently recommended in TA103. The Committee heard that the inclusion criteria used in the clinical trials were representative of people with psoriasis who are being considered for treatment with biological therapies in clinical practice.
4.6 The Committee noted that the manufacturer had conducted a mixed treatment comparison to enable a comparison of ustekinumab with all alternative biological therapies currently available for the treatment of psoriasis. The Committee noted that two analyses had been completed: one analysed data from all patients according to their randomisation, whereas the other analysed data from patients according to a weight-based dosing approach. The Committee noted that the results for both analyses suggested a higher probability of a response after treatment with ustekinumab than with etanercept or adalimumab, but a lower probability of a response compared with infliximab.
4.7 The Committee discussed comments received during consultation on the appraisal consultation document (ACD) suggesting that the efficacy of adalimumab had been underestimated in the mixed treatment comparison because of the possible exclusion of relevant outcome data and the inclusion of a study that had enrolled people with less severe psoriasis. In addition, the Committee discussed the uncertainties about how the analysis had been completed and how it compared with analyses used in previous appraisals. It also considered that randomisation may not have been maintained in the weight-based dosing analysis. The Committee recognised these issues concerning the mixed treatment comparison and took them into account in its decision-making.
4.8 The Committee considered whether the appropriate comparator for ustekinumab should be etanercept given continuously or intermittently, with the latter regimen being specified in TA103 and in the marketing authorisation for etanercept. The Committee heard from the clinical specialists that biological therapies for psoriasis, including etanercept, are usually used on a continuous basis in clinical practice, although treatment may be interrupted if a person has a sustained remission. The Committee heard that treatment withdrawal was carried out cautiously because a person's condition may deteriorate rapidly and they may subsequently not regain full control of their disease. The Committee heard from the clinical specialists that ustekinumab was likely to be used in a similar way to other biological therapies. The Committee recognised that there is variation in the administration of etanercept in clinical practice, and noted a comment received during consultation on the ACD stating that etanercept is usually given intermittently and only given continuously when required.
4.9 The Committee was aware that the clinical specialists had indicated that ustekinumab may be used after a person's psoriasis had shown an inadequate response to other biological therapies. It was also aware that guidelines in preparation from the British Association of Dermatology might include advice on the sequential use of such therapies. The Committee took note of comments received on the ACD suggesting the use of ustekinumab after the failure of TNF inhibitors. However, no evidence for the use of ustekinumab after an inadequate response to other biological therapies was placed before the Committee. It noted that 40–50% of people in the PHOENIX trials had received previous treatment with biological therapies, but that a person's psoriasis had not necessarily shown an inadequate response to these therapies before the trial use of ustekinumab. Furthermore, data for this subgroup had not been presented separately. Therefore the Committee felt that it could not make any specific recommendations on the use of ustekinumab after a person's psoriasis had failed to respond to other biological therapies. However, it considered that data on the effectiveness of biological therapies, including ustekinumab, for the sequential treatment of severe plaque psoriasis would be an important part of future assessments.
4.10 The Committee discussed the results of the economic analysis conducted by the manufacturer. It considered the overall approach to modelling adopted by the manufacturer to be appropriate, but noted comments received during consultation on the ACD relating to the potential limitations of the probabilistic sensitivity analyses. The Committee noted the ERG's concerns that no formal subgroup analysis that justified weight-based dosing had been done. It also discussed comments received during consultation on the ACD that other biological therapies might also demonstrate a weight–dose relationship. However, the Committee noted that weight-based dosing is included in the marketing authorisation for ustekinumab and that evidence had been presented for a dose–response relationship with this drug.
4.11 The Committee discussed the assumption in the economic model that 20% of people receiving ustekinumab would weigh more than 100 kg. It recognised that this might be an underestimate, because around 30% of the people included in the PHOENIX trials weighed more than 100 kg. However, the Committee considered that comments received during consultation on the ACD had shown that changing this assumption had minimal impact on estimates of cost effectiveness.
4.12 The Committee noted the assumption in the model that a hospital inpatient period of 21 days would be required for people whose psoriasis had not responded adequately to treatment. The Committee noted that this assumption had been used in the appraisals of other biological therapies for psoriasis. The Committee heard from the clinical specialists and patient experts that 21 days of inpatient treatment in a year was plausible for a person with severe psoriasis that had not responded adequately to treatment. The Committee also heard from the clinical specialists that the cost of £288 per day for an inpatient stay, as assumed in the model, may be too low. Costs as high as £700 per day may be incurred, but these are usually associated with shorter, more intensive inpatient admissions. Additionally, the Committee heard that the cost of supportive care may be higher than calculated in the model because people may receive methotrexate or ciclosporin even if their disease is not adequately controlled by these treatments. The Committee recognised that the costs were similar to those used in previous appraisals, but was concerned about their accuracy.
4.13 The Committee noted that the economic model assumed that the efficacy of intermittent etanercept was lower than that of continuous etanercept. The Committee was informed that this was based on an RCT showing that, for the outcome measured (PGA score), intermittent etanercept was less effective than continuous etanercept. This difference in effectiveness had then been applied to the PASI response data for continuous etanercept in the mixed treatment comparison in order to determine the efficacy of intermittent etanercept. The Committee considered that an assumption of reduced efficacy of intermittent etanercept may be reasonable, but that the way this had been calculated in the model increased the uncertainty in the results.
4.14 The Committee discussed comments received during consultation on the ACD about the cost of etanercept 25 mg given intermittently. It recognised that the appraisal of etanercept and efalizumab (TA103) had assumed that the cost of intermittent etanercept 25 mg was 74% that of continuous etanercept. However, in the current appraisal of ustekinumab an estimate of 88% had been used, which reflected that used by another manufacturer in the appraisal of adalimumab. The Committee noted comments received during consultation on the ACD that if the cost of intermittent etanercept 25 mg was 74% of that of continuous etanercept, the ICER for ustekinumab in comparison with intermittent etanercept 25 mg was £68,300 per QALY gained. However, the Committee was mindful of comments from clinical specialists that for people with severe psoriasis, treatment may be given continuously or may have short re-treatment intervals. The Committee recognised that in a scenario where etanercept was given continuously, the manufacturer's analysis suggested that ustekinumab was less costly and more effective.
4.15 The Committee noted that the economic model included a 20% annual dropout rate for people whose psoriasis responded to treatment and that this rate was assumed to be the same for all biological therapies. The Committee heard from the clinical specialists that people on biological therapies do stop treatment because of a reduction in response or adverse events, and that they considered this estimate to be reasonable.
4.16 The Committee was aware that EQ-5D data had not been obtained in the clinical trials, and noted that the manufacturer had mapped DLQI scores to EQ-5D scores to obtain estimates of utility. The Committee noted that this approach had been used in TA103. The Committee recognised that the manufacturer had also provided a secondary analysis using SF-36 values from the PHOENIX-1 trial transformed into SF-6D scores. The Committee accepted the manufacturer's use of mapping to determine utility estimates.
4.17 The Committee noted that the cost-effectiveness analysis included the patient access scheme. It noted that without the patient access scheme the ICERs for ustekinumab would be £41,000 per QALY gained compared with supportive care, £102,000 per QALY gained compared with intermittent etanercept 25 mg, and £300,000 per QALY gained compared with adalimumab. The Committee therefore concluded that ustekinumab could not be considered a cost-effective use of NHS resources without the patient access scheme. The Committee was reassured that the patient access scheme would remain in place until either a review of the guidance by NICE or the introduction of any new formulations that would render the scheme obsolete, and that it would not be withdrawn without the agreement of NICE and the Department of Health. The Committee concluded that it was reasonable to consider the estimates of cost effectiveness that included the patient access scheme.
4.18 The Committee noted that in the manufacturer's base-case analysis, which included the patient access scheme, ustekinumab had an ICER of £29,600 per QALY gained compared with supportive care, and an ICER of £27,100 per QALY gained compared with etanercept 25 mg given intermittently. The Committee was mindful that this analysis assumed that the cost of intermittent etanercept was 88% of the cost of continuous etanercept. The Committee also noted that the manufacturer's analysis suggested that ustekinumab was less costly and more effective than adalimumab. However, it was aware that revised estimates for the efficacy of adalimumab had been provided during consultation on the ACD, and the resulting ICERs suggested that ustekinumab was not a cost-effective alternative to adalimumab. The Committee considered that the differences in incremental costs and QALYs between all treatments were small, and that this was particularly the case when considering ustekinumab and adalimumab. This meant that these ICERS were very sensitive to small changes in either costs or QALYs and therefore did not represent stable estimates of cost effectiveness. Therefore the Committee concluded that no robust differences in cost effectiveness between adalimumab and ustekinumab had been shown.
4.19 The Committee considered how the population with severe psoriasis should be defined. It heard from the clinical specialists that a combination of DLQI and PASI scores is used routinely in clinical practice, and agreed that it would be appropriate to define severe disease as a PASI score of 10 or more and a DLQI score of more than 10, in line with TA103. Furthermore, the clinical specialists indicated that the treatment continuation rules defined in section 1.2 of TA103 remain relevant to clinical practice. However, the Committee noted that the response should be measured at 16 weeks for ustekinumab, rather than at 12 weeks as defined for etanercept in TA103, and that this measurement should be carried out before the third (16-week) dose is given.
4.20 The Committee was mindful of the uncertainties in the resource and cost data and the potential methodological limitations of the mixed treatment comparison. The Committee considered that it would be of value to review all of the biological therapies for psoriasis in a multiple technology appraisal. It also noted that data collection, as described in its recommendations for further research (see section 6), would help decisions to be made in future appraisals. It concluded that the estimates of the cost effectiveness of ustekinumab compared with supportive care were acceptable. It also concluded that, in comparisons of ustekinumab with other biological therapies, the ICERs depended on small differences in costs and benefits that were subject to uncertainty. On balance, the Committee was persuaded that ustekinumab should be recommended as a treatment option for people with severe plaque psoriasis when standard systemic therapies have not produced an adequate response, or if a person is intolerant of or has a contraindication to these therapies.
4.21 The Committee was aware that there might be some situations when the DLQI may not be a clinically appropriate tool to inform a clinician's conclusion about the severity of psoriasis; for example, if a person has physical, sensory or learning disabilities, or communication difficulties that could affect their responses to the questionnaire. The Committee heard from the clinical specialists that the DLQI is now available in more than 50 languages and that this has improved assessment for those people whose first language is not English. The Committee concluded that healthcare professionals should take any physical, sensory or learning disabilities and communication difficulties into account when using the DLQI and make any adjustments they consider appropriate.