Review of utility data in children and young people with psoriasis

A systematic literature review was conducted to identify utility values for plaque psoriasis in children and young people. The aim of the search was to identify any studies that reported utility values or other measures of HRQoL that could be converted into utility values specifically for the population of children and young people.

The search strategy was developed in MEDLINE (via Ovid) by an information specialist with input from the project team. The strategy included terms for psoriasis combined, using the Boolean operator AND, with terms for quality of life/utilities or named instruments. No language, geographical or date limits were applied. A search filter to limit retrieval to quality-of-life studies was used when available. The search strategy was adapted for use in the other resources searched. Full search strategies can be found in Appendix 1.

The following databases were searched on 12 July 2016: MEDLINE [including MEDLINE Epub Ahead of Print, MEDLINE In-Process & Other Non-Indexed Citations, Ovid MEDLINE(R) Daily and Ovid MEDLINE(R)], the Cost-effectiveness Analysis (CEA) Registry [see https://research.tufts-nemc.org/cear4/ (accessed 10 August 2017)] EMBASE, PubMed and the School of Health and Related Research Health Utilities Database (ScHARRHUD) [see www.scharrhud.org/ (accessed 10 August 2017)].

The Health Economics Research Centre (HERC) database of mapping studies from the University of Oxford¹⁵² was also searched to identify any suitable mapping algorithms that would allow conversion of clinical measures routinely collected in studies of psoriasis in children and young people into utility values.

The results from the searches were imported into an EndNote X7 library and deduplicated. After deduplication, 286 records in total were identified. The titles and abstracts were assessed independently by two reviewers for inclusion and any discrepancies were resolved by consensus. None of the titles identified reported utility values collected in children and young people with psoriasis. The search of the HERC mapping algorithm database identified one study on the development of a mapping algorithm to estimate EQ-5D-Y (EuroQol-5 Dimensions for Youth) utility values from PedsQL general core scales.¹⁵³ The PedsQL is a generic instrument for measuring HRQoL in children and adolescents with acute and chronic health conditions. The PedsQL measures core dimensions of health as delineated by the WHO, as well as role (school) functioning. The four multidimensional scales are physical functioning (eight items), emotional functioning (five items), social functioning (five items) and school functioning (five items).¹⁹ The EQ-5D-Y is the youth version of the EQ-5D, which has been specifically adapted in terms of language for children aged 8–11 years and adolescents aged 12–18 years. In the absence of a tariff set specifically for the EQ-5D-Y, the authors¹⁵³ applied the UK national tariff for the valuation of the EQ-5D-3L (EuroQol-5 Dimensions three-level version) to generate utility values from the EQ-5D-Y instrument.¹⁵⁴

Khan et al.¹⁵³ assessed different mapping methods for estimating EQ-5D-Y health utilities from PedsQL response scores. The study used data collected in a cross-sectional survey conducted in four secondary schools in England among children aged 11–15 years. The sample on which the mapping models were estimated included 559 children and the validation sample included 337 children. Children in the full study sample (n = 896) were on average aged 13.3 years (standard deviation 1.3 years), 54% were male and approximately 40% were of non-white ethnicity. The authors explored both direct and response mapping approaches to predict EQ-5D-Y utility values, as well as a number of functional forms, including ordinary least squares (OLS) regression, generalised linear models, two-part logit–OLS regression, censored least absolute deviation and Tobit regression. Model performance was assessed on the validation sample and models were re-estimated on the full study sample. Table 58 presents the two best-fitting models for the mapping algorithm. These correspond to the models estimated using OLS regression with (1) age and sex terms included as regressors and (2) excluding age and sex terms as regressors.

TABLE 58

Best-fitting mapping algorithms from the PedsQL to EQ-5D-Y based on the study by Khan et al.

The two models were considered to have similar prediction accuracy for mean EQ-5D-Y values. Model 1, which included age and sex as regressors, had a better fit across a wider range of EQ-5D-Y values than model 2. Model 2 reported a better fit for the EQ-5D-Y utility score range of 0.8–1.0 category. There are a number of potential limitations to the use of this algorithm to predict EQ-5D-Y utilities. The sample on which the models were estimated consisted of healthy children aged 11–15 years, which may limit the predictive accuracy in sicker populations or populations outside this age range. The authors recognise that there is a need for further validation and testing of the algorithm but, in the absence of an alternative source, it remains a useful tool for estimating EQ-5D-Y utility values in situations in which only the PedsQL has been administered.

Utility data reported in company submissions

Health-related quality-of-life assessments were carried out in the etanercept trial (20030211⁴⁹), the ustekinumab trial (CADMUS⁷²) and the adalimumab trial (M04-717⁴²) using the CDLQI and the PedsQL at selected time points. EQ-5D or EQ-5D-Y values were not collected in any of the trials. Therefore, the only way to include EQ-5D utility values in the model was by mapping from either CDLQI scores or PedsQL scores. The literature review described earlier did not identify any studies that estimated the relationship between CDLQI scores and EQ-5D values, whereas the study by Khan et al.¹⁵³ was the only study that estimated the relationship between PedsQL scores and EQ-5D values. The AG requested from the companies access to individual patient data (IPD) for PedsQL domain scores at baseline and follow-up by category of PASI response and PedsQL summary scores at the domain level by response category. The AG did not receive access to IPD; however, Janssen (ustekinumab) submitted aggregated summary data (mean and standard deviation) from the CADMUS trial for the PedsQL subscale and total scale scores by treatment arm (placebo and ustekinumab standard dose) and PASI response category at 12 weeks (< 50, 50–74, 75–89, ≥ 90) for baseline and 12, 28 and 52 weeks.

Utility estimates used in the model

The utility values associated with treatment in previous models in adults were based on the proportion of individuals in the different PASI response categories (< 50, 50–74, 75–89, ≥ 90) and the change in utility from baseline associated with the PASI response categories. Therefore, PASI response rates from the NMA were assumed to be a perfect proxy for change in utility arising from treatment.

The relationship between utility and PASI response was estimated in previous TAs in adults using either DLQI data mapped onto EQ-5D utility values or directly from EQ-5D data collected in the trials. In the population of children and young people, the only possibility of obtaining EQ-5D values was by mapping from PedsQL scores to EQ-5D-Y values, as described earlier. Without access to the IPD, which would allow full uncertainty to be reflected in the values, the mapping algorithm was applied to the summary scores at the domain level from the CADMUS trial.

Validation of the algorithm was performed by examining data reported in a study by Varni et al.,¹⁵⁵ which compared self-reported HRQoL (based on PedsQL scores) between paediatric patients with moderate to severe plaque psoriasis and a healthy population sample. The sample used to represent the psoriasis population corresponded to individuals in the main efficacy trial for etanercept (n = 208, age 4–17 years) and measurements of PedsQL scores at baseline were pooled across the two treatment arms (etanercept and placebo). The healthy population sample was taken from a US children’s health insurance programme evaluation (n = 5079) open to children and young people aged 2–16 years. Table 59 summarises the PedsQL subscale scores reported in Varni et al.¹⁵⁵ for the psoriasis and healthy populations, alongside the estimates obtained by applying the two best-fitting mapping algorithms to obtain EQ-5D-Y utility values (model 1 includes age and sex terms whereas model 2 excludes these variables).

TABLE 59

Application of the mapping algorithm to estimate EQ-5D-Y utilities in paediatric populations

The EQ-5D utility estimates were higher in the healthy population than in the population with psoriasis, irrespective of the model used to map PedsQL scores to EQ-5D-Y values. The distinction between the models was minimal, especially in the psoriasis population: model 1 provided slightly higher utility values than model 2 (0.6% and 2.5% higher in the psoriasis and healthy populations respectively). Model 2 was subsequently used in the base-case analysis as the reference category for the variable sex was unclear in the study by Khan et al.¹⁵³

The mapping algorithm (model 2 in Table 58) was used to estimate change in EQ-5D-Y utility values from baseline based on PedsQL data from the ustekinumab trial (CADMUS) at baseline and 12 weeks’ follow-up (the time point at which response to treatment was assessed in the trial and blinding of randomised subjects in the trial was terminated; after this point crossover between treatment arms was possible). The mean change in EQ-5D-Y values between baseline and week 12 was estimated for individuals with different levels of PASI response. Table 60 reports the EQ-5D-Y utility values estimated for the base-case-analysis, with the placebo and treatment arms pooled.

TABLE 60

Baseline utility and mean change in utility by PASI response, estimated from CADMUS trial PedsQL data mapped onto the EQ-5D-Y

The baseline utility estimate is similar to that derived from the etanercept trial (20030211) (0.864 in Table 59) and is lower than the general healthy population estimate of 0.913 based on the study by Varni et al.¹⁵⁵ The mean changes in EQ-5D utility from baseline by PASI response category are much smaller than the corresponding changes in EQ-5D utility observed in previous TAs in adults. For example, the EQ-5D changes in utility by PASI response category in the York model of adults were 0.05 for PASI < 50, 0.17 for PASI 50–74, 0.19 for PASI 75–89 and 0.21 for PASI ≥ 90.

To examine whether or not the changes in EQ-5D-Y utility values were accompanied by similar changes in other measures of HRQoL in the population of children and young people, the changes in EQ-5D-Y values were compared with reported CDLQI values by PASI response (Table 61). A comparison of EQ-5D and DLQI values by PASI response in adults is also shown in Table 61 (taken from TA180¹⁰⁰ for ustekinumab, which was the only TA in adults that reported both outcomes). The mean changes in CDLQI score by PASI response in the CADMUS trial are much smaller than the mean changes in DLQI score in TA180,¹⁰⁰ which is consistent with the smaller mean changes estimated for the EQ-5D-Y in the paediatric population than for the EQ-5D in adults. These differences, however, should be interpreted with caution as CDLQI and DLQI scores are not directly comparable and the number of observations was much smaller in the population of children and young people than in the adult population in TA180.¹⁰⁰

TABLE 61

Mean change from baseline in CDLQI/DLQI scores and EQ-5D/EQ-5D-Y utilities by PASI response

The EQ-5D-Y/EQ-5D utility estimates suggest that improvements in HRQoL associated with reductions in PASI response rates are of a much smaller magnitude in children and young people than in adults; however, the evidence is highly uncertain because of the small sample size and the limited data available to validate the findings. In the absence of an alternative source to estimate EQ-5D values for the model, these values were used in the base-case analysis. It is important to highlight a number of limitations of this approach. First, the use of a mapping algorithm to estimate utilities introduces uncertainty compared with direct EQ-5D measurement. Second, the Khan et al.¹⁵³ mapping algorithm has not been validated in children aged < 11 years or in a population with psoriasis. Third, the CADMUS trial, from which the PedsQL data that were mapped to EQ-5D utilities were sourced, excluded children aged < 12 years; therefore, it remains uncertain whether or not the mapped utilities are reflective of this population. Fourth, in populations aged < 12 years, there may be issues with lack of agreement or consistency between self-reported and proxy (parent)-reported measurements.¹⁵⁶ Therefore, even if PedsQL data were available for younger children, the mapping algorithm might not consistently perform for self-reported and parent-reported measurements of the instrument. Finally, Khan et al.¹⁵³ used the EQ-5D-3L value set as a proxy for EQ-5D-Y in the absence of an alternative tariff set, but this approach is currently not recommended.¹⁵⁷ These limitations reduce the robustness of the utility estimates used in the model.

There might be other potential benefits of treatment that fall outside the QALY estimation. First, children and young people may miss schooldays to attend health-care appointments and may be absent for longer periods from school while experiencing symptoms. This can have a negative impact on their education/academic achievements and, in future, their ability to gain employment. It may also affect their social and psychological health through the reduced ability to participate in social and leisure activities and sport. Second, early treatment of children and young people with biological agents may prevent long-term multisystem morbidity (e.g. hypertension, cardiovascular disease, depression), which has a higher prevalence in adults with psoriasis than in the general population.¹⁵⁸ Finally, there may also be other aspects of HRQoL that are outside the perspective defined by NICE’s reference case,¹⁵⁰ namely the potential impact on the HRQoL of carers of children and young people with psoriasis if treatment with biologics reduces the spillover disutility of illness by improving patients’ outcomes. The impact on carers may also extend to a reduced ability to participate in normal activities, both work- and non-work related. Because of an absence of quantitative estimates of the impact on the HRQoL of children and young people with psoriasis receiving any of the interventions and the potential benefits to their carers, it was not possible to incorporate them into the economic analysis. Any attempt to add arbitrary values to the utility estimates, which are already highly uncertain, would introduce further uncertainty.

Given the uncertainty surrounding the utility estimates for children and young people, scenario analyses were conducted using utility estimates from previous TAs in adults for etanercept, adalimumab and ustekinumab. Table 62 summarises the utility estimates considered in the scenario analyses.

TABLE 62

Baseline utility and mean changes in utility by PASI response used in the base-case and scenario analyses

Drug acquisition costs

Table 64 details the dose and frequency of administration for each treatment and comparator, including ciclosporin, which forms part of BSC, and the unit costs associated with each treatment.

TABLE 64

Drug acquisition costs in children and young people

The dosages of the biological therapies were taken from the Summaries of Product Characteristics.^167–169 For methotrexate and ciclosporin, which are currently not licensed for paediatric use, the dosages were sourced from published literature^78,170,171 and confirmed with our clinical advisor to ensure that they reflected UK clinical practice in this population. Methotrexate can be administered orally or injected subcutaneously or intramuscularly. In the model it was assumed that 72% of individuals are given methotrexate in oral solution and 28% in injectable solution, which reflects the distribution of administration identified in the UK psoriasis audit of the use of systemic treatments in children and young people.¹⁴⁸ Therefore, the unit cost per mg for methotrexate is a weighted average of the unit cost per mg of the oral and injectable solutions (i.e. £0.71/mg). Unit costs were sourced from MIMS¹⁶⁰ and supplemented with data from the BNF.¹⁶¹

Figure 11 illustrates the number of doses administered in the first five cycles of the model for each treatment based on the licensed dose. Adalimumab is administered at weeks 0 (baseline) and 1 and then every 2 weeks thereafter until response assessment at the end of week 16. If individuals are responders to treatment they continue to receive adalimumab every 2 weeks until treatment withdrawal (highlighted in grey). Ustekinumab is administered at weeks 0 and 4 and then every 12 weeks thereafter, with response assessment at week 16. Etanercept and methotrexate are administered weekly, with response assessment at weeks 12 and 16 respectively.

FIGURE 11

Drug dose distribution during the first five cycles in the model. ADA, adalimumab; ETA, etanercept; MTX, methotrexate; UST, ustekinumab; •, ADA administration; ♦, ETA administration; ▪, UST administration; Δ, MTX administration. (more...)

The dosages of the biological treatments are dependent on patient weight. The median weight by age and sex in the population of children and young people was extracted from the Royal College of Paediatrics and Child Health’s school-age growth charts.¹⁷² Table 65 shows the weights used in the model by age. These were based on an average of the weight of boys and girls (and when the weight estimate in the growth chart did not correspond to an integer, the next-highest integer was used).

TABLE 65

Median weight by age used in the model

The weight by age was used to estimate the correct dosage of each treatment and the corresponding cost. Table 66 summarises the dosages used in the model for each treatment by age and the corresponding costs per dose. Following clinical advice it was assumed that the vial with the lowest dose available would be used to allow administration of a single dose in the paediatric population. This inevitably results in wastage of the remainder of the vial. For example, for individuals who weigh < 60 kg the full cost of the 45-mg vial of ustekinumab is assumed as the remaining product in the vial cannot be stored. Vial splitting across individuals was considered unlikely because in most cases the majority of the vial is used for a single patient and treating patients together is less likely to occur in this population because of low patient numbers. Therefore, the cost per dose was fixed for adalimumab and ustekinumab (£352.14 and £2147.00 respectively). For etanercept, the 25-mg vial (£89.38) is used for children aged < 10 years whereas the 50-mg vial (£178.75) is used for those aged ≥ 10 years.

TABLE 66

Drug dosages and cost per dose by age in the model

Drug administration costs

Adalimumab, etanercept and ustekinumab were assumed to be self-administered. In the case of younger children it was assumed that a parent or carer would administer the subcutaneous injection. Subcutaneous injections were assumed to incur administration costs only for nurse training for self-administration (or parent/carer administration) in the induction phase. In line with previous TAs in adults, this was assumed to require 3 hours of nurse time, which was costed based on the cost per working hour of a band 5 hospital nurse with qualifications (£43 per hour). A cost of £129 was applied in the first cycle of the model for the administration of the biologics.

Monitoring costs

Table 67 summarises the resource use assumptions made in relation to monitoring and the corresponding unit costs applied in the model. In the absence of evidence specifically relating to the population of children and young people, resource use estimates associated with monitoring and routine laboratory tests for biological and non-biological systemic treatments were taken from NICE CG153,¹⁷³ which used similar assumptions to those in the original York model (TA103⁹⁷) and subsequent NICE appraisals of biological treatments.^98–102

TABLE 67

Monitoring resource use and unit costs

Individuals on biological therapy were assumed to undertake a series of tests during the initial trial period, namely a full blood count, liver function test and urea and electrolytes test. During the trial period the tests were assumed to be carried out during two routine outpatient visits that occur at treatment initiation and at the end of the trial period (treatment response assessment visit). As methotrexate was not included as a comparator in CG153¹⁷³ (only as part of BSC), it was assumed that the resource use for the monitoring of methotrexate in the trial period is the same as that for biological treatment. In the maintenance period (corresponding to the health state of ‘continued use’), individuals on systemic therapies were assumed to be monitored once every 3 months.¹⁷³ The unit costs for glomerular filtration rate and outpatient visits were taken from NHS reference costs 2014–15,¹⁵⁹ whereas the costs of the remaining monitoring items were inflated to 2014–15 prices based on estimates presented in TA103.⁹⁷

The costs of tests undertaken solely to screen individuals for eligibility for treatment were excluded from the analysis, namely chest radiography, tests for tuberculosis or biopsies of lesions atypical of psoriasis. These costs were also excluded in previous appraisals in adults. The cost of folic acid used in conjunction with methotrexate to prevent side effects was also excluded from previous appraisals as the annual cost of this drug is very low (< £1). Our clinical advisor indicated that children and young people would be tested for herpes zoster before treatment initiation; however, as this test would be performed on every patient not immune to the virus regardless of treatment, it was excluded from the analysis. The costs of liver biopsy and type III procollagen peptide (PIIINP) monitoring for the purpose of assessing liver function in individuals treated with methotrexate were also excluded from the analysis based on clinical advice; liver biopsy is seldom conducted in children and young people given its invasiveness, whereas PIIINP is a marker of growth in this population rather than of hepatic toxicity.

Best supportive care costs

Best supportive care corresponds to the management of individuals after failure of conventional systemic therapies. BSC is also considered a relevant comparator to biological treatments. If biological treatments are found not to be effective, individuals are usually offered some form of BSC rather than no treatment. BSC tends to include a mix of active non-biological systemic therapies such as methotrexate and ciclosporin and palliative care, including phototherapy, as well as outpatient visits and hospitalisations to manage disease flare-ups.

The resource use and costs associated with BSC have represented a significant area of uncertainty in the analysis of the cost-effectiveness of biological treatments for moderate to severe psoriasis in adults. In TAs prior to CG153,^97–100 the definition of BSC in terms of resource use and costs was restricted to outpatient visits and hospitalisations to manage the symptoms of psoriasis, with these largely informed by assumptions and clinical opinion. In CG153,¹⁷³ the definition of BSC was expanded to also include non-biological systemic treatments, phototherapy and attendance at tertiary day centres. As discussed previously (see Chapter 5, Resource use and costs in the York model and subsequent appraisals), this guideline used estimates of resource use from observational studies in the UK¹⁴³ and the Netherlands¹⁴⁴ but also relied heavily on clinical opinion and assumptions. In the absence of evidence for children and young people, the definition of BSC from CG153¹⁷³ was used in the model, with input from our clinical advisor on the appropriateness of the assumptions for a younger population.

Table 68 summarises the resource use assumptions for BSC by category of cost in CG153¹⁷³ and those applied in the model, alongside the associated unit costs. Unit costs were sourced from the BNF,¹⁶¹ MIMS¹⁶⁰ and NHS reference costs 2014–15.¹⁵⁹ The relative proportions of individuals on active treatment with methotrexate and ciclosporin were modified from those used in CG153¹⁷³ based on clinical opinion that children and young people are less likely to be managed with ciclosporin than adults because of the renal toxicity of the drug. Data from a UK psoriasis audit on the use of systemic treatments in children and young people were used to inform the relative proportion of individuals on methotrexate and ciclosporin.¹⁴⁸ In CG153, it was assumed that 90% of individuals receiving BSC would be on active treatment with systemic drugs. However, in the audit, 53 patients were treated with non-biological systemic treatments, of whom 25 patients were treated with methotrexate and 12 with ciclosporin. Therefore, instead of assuming that individuals are equally distributed between methotrexate and ciclosporin, a ratio (25/37 and 12/37 for those on methotrexate and ciclosporin respectively) for each treatment was applied to the overall proportion of 90% to reflect the distribution of children and young people receiving these treatments in the audit. The corresponding proportions of individuals assumed to receive methotrexate and ciclosporin as part of BSC were 61% and 29%, respectively. As in CG153,¹⁷³ treatment with ciclosporin was assumed to be discontinued after a maximum duration of 2 years (because of the increased risk of renal toxicity). Monitoring costs associated with the use of these non-biological systemic therapies were applied in the model as presented in Monitoring costs.

TABLE 68

Resource use assumptions and unit costs for BSC in CG153 and the current analysis

In line with CG153,¹⁷³ 16% of the population were assumed to undergo 24 sessions of phototherapy per year (NBUVB) and five outpatient visits per annum were assumed for the 10% of individuals not managed with systemic therapies. All individuals were assumed to incur the costs of five visits per annum to a specialist dermatology day centre, in line with CG153.¹⁷³

The resource use associated with hospitalisations for individuals on BSC was identified as an area of high uncertainty and a key driver of cost-effectiveness in the previous TAs in adults. The number of bed-days assumed in CG153¹⁷³ (26.6 days per year) was based on the average LOS for psoriasis patients with a baseline PASI of 10–20 points taken from a UK observational study¹⁴⁵ combined with the average number of hospitalisations for individuals at high need (one hospitalisation per year) and very high need (2.55 hospitalisations per year) from a Dutch observational study.¹⁴⁴ The total of 26.6 days of hospitalisation per annum was considered by the NICE Appraisal Committees for TA350¹⁰¹ (secukinumab) and TA368¹⁰² (apremilast) in adults to be too high. Our clinical advisor suggested that hospitalisations in children and young people are very rare. This is largely because children and young people have not yet developed the comorbidities that often lead to hospitalisations in adults with psoriasis. Therefore, in the base-case analysis it was assumed that children and young people do not incur any inpatient stays. In separate scenario analyses, estimates of 26.6 days of hospitalisation per annum¹⁷³ and 6.49 days of hospitalisation per annum¹⁴⁴ were considered.

Adverse event costs

As discussed in Chapter 5 (see Resource use and costs in the York model and subsequent appraisals), only one previous TA in adults¹⁰¹ considered the costs of hospitalisations resulting from AEs in the cost-effectiveness analysis. The AEs that were assumed to lead to relevant resource use consumption (i.e. those leading to hospitalisations) in this evaluation were (1) NMSC, (2) malignancies other than NMSC and (3) severe infections. The rates of AEs as reported in the literature (for adalimumab, etanercept, ustekinumab and infliximab) and from trial data (for secukinumab) were applied to each treatment arm as per the rates of these events occurring.

The safety data from the clinical trials of biological drugs for the treatment of severe-to-moderate psoriasis (see Chapter 3, Safety of adalimumab, Safety of etanercept and Safety of ustekinumab) suggested that there was little difference in the short- and long-term rates of AEs between trial arms, with the potential exception of etanercept, for which a higher rate of infections (not statistically significant) was observed than for placebo. However, the trial data included a small number of observations for each treatment and a limited follow-up period (from 52 weeks for adalimumab^46,47 to 312 weeks for etanercept).^49,52,80 Observational studies in children and young people with psoriasis^76,77 (see Chapter 3, Additional observational evidence) did not report any increase in infections or SAEs associated with the use of biological therapies.

Given the paucity of robust evidence on the incidence of AEs in children and young people with moderate-to-severe psoriasis, the costs of these were not included in the base-case analysis. However, scenario analyses were conducted to explore the impact on the cost-effectiveness results of including the costs associated with hospitalisations resulting from serious infections and malignancies (both NMSC and other). The rates of AEs were sourced from TA350¹⁰¹ and supplemented with data from Dixon et al.¹⁷⁴ for methotrexate, whereas the unit costs were taken from NHS reference costs 2014–15.¹⁵⁹ Table 69 summarises the adverse event rates applied in the model, alongside the corresponding unit costs.

TABLE 69

Adverse event rates applied in the model

The costs of AEs associated with biological therapies and methotrexate were applied in the model to individuals while on treatment. Individuals treated with BSC were assumed not to develop AEs.

Base-case analysis

The expected costs and QALYs of the interventions and comparators were determined for each population and the relative cost-effectiveness was established using standard decision rules and reported using ICERs as appropriate. The ICER examines the additional cost that one treatment option incurs over another and compares this with the additional health benefits to give the additional cost of the treatment for each additional QALY gained. When more than two treatment options are being compared, the ICERs are calculated using the following process:

1. The treatment options are ranked in terms of mean QALYs (from the least effective to the most effective).
2. If a treatment option is more costly and less effective than any other option, then this treatment is said to be dominated and is excluded from the calculation of the ICERs.
3. The ICERs are calculated for each successive alternative, from the least effective to the most effective. If the ICER for a given treatment option is higher than that of any more effective option, then this treatment option is ruled out on the basis of extended dominance.
4. Finally, the ICERs are recalculated, excluding any treatment options that are ruled out by principles of dominance or extended dominance.

The resulting ICERs provide the basis for establishing which treatment appears optimal based on cost-effectiveness considerations. Guidance from NICE¹⁵⁰ suggests that an incremental cost per additional QALY of around £20,000–30,000 is considered to represent an appropriate threshold for the health opportunity costs to the NHS.

The ICER comparing all interventions and comparators relates to a situation in which the decision-maker can choose only one of the treatment options. However, in psoriasis, as indicated previously, if an individual patient does not respond to or tolerate one of the biological therapies, an alternative one is usually tried. This means that treatments are usually trialled on an individual basis until an effective option is found. The ICERs comparing each intervention with BSC (after systemic therapy) or methotrexate (before systemic therapy) are also presented, to indicate the optimum ordering of treatments in terms of their cost-effectiveness. The most cost-effective order in which to give the therapies based on total expected costs and QALYs associated with each treatment option is dependent on the cost-effectiveness threshold.

Probabilistic sensitivity analysis was used to represent uncertainty in the cost-effectiveness results. The effectiveness data were entered as simulated posterior distributions from the Markov chain Monte Carlo analysis to reflect uncertainty in the mean estimates. Monte Carlo simulation was used to propagate the uncertainty in the input parameters over 10,000 draws, from which mean costs and QALYs were then obtained by averaging over the 10,000 simulations. The probability that a treatment is first in the sequence was also estimated.

Differences in the marketing authorisations of the interventions by age and the positioning of adalimumab before non-biological systemic therapy means that the comparative cost-effectiveness of the interventions needs to be evaluated by age and before or after use of systemics. The relevant comparator also depends on the position of the particular intervention in the pathway. Before systemic therapy, methotrexate is the relevant comparator (as the current standard of care), whereas after systemic therapy BSC represents the most relevant comparator. Three base-case populations are presented:

1. children and young people aged 4–17 years with adalimumab compared with methotrexate, that is, as a second-line therapy in individuals who are inadequately controlled by, or who are intolerant to, topical therapy and phototherapies
2. children and young people aged 6–11 years with adalimumab and etanercept compared with BSC and with each other, that is, as third-line therapy in individuals who are inadequately controlled by, or who are intolerant to, systemic therapies or phototherapies
3. children and young people aged 12–17 years with adalimumab, etanercept and ustekinumab compared with BSC and with each other, that is, as third-line therapy in individuals who are inadequately controlled by, or who are intolerant to, systemic therapies or phototherapies.

Table 70 summarises the input parameters used in the base-case analysis.

Scenario analysis

A number of alternative scenarios were considered in which the assumptions used as part of the base-case analysis were varied. These analyses were undertaken to assess the robustness of the base-case results to variation in the assumptions and sources of the data used to populate the model. Table 71 summarises the alternative scenarios considered. For each element, the position in the base-case analysis is outlined, alongside the alternative assumptions applied. The cost-effectiveness of the interventions was considered under each of the scenarios for each of the licensed populations.

TABLE 71

Details of the key elements of the base-case analysis and the variations used in scenario analyses

Intervention and comparators

Scenario 1: off-label use of biologics outside age constraints and position in the pathway

As discussed in Decision problem and patient population, the biological interventions differ in their marketing authorisation by age and positioning of treatment in the pathway. Adalimumab is licensed for the youngest age group from ≥ 4 years and is the only biological treatment positioned as a second-line therapy in individuals who are inadequately controlled by, or who are intolerant to, topical therapy and phototherapies, that is, as an alternative to systemic therapy. This makes the comparison of adalimumab with etanercept and ustekinumab more problematic as the latter interventions are licensed as third-line therapies in individuals who are inadequately controlled by, or who are intolerant to, systemic therapies or phototherapies and who are aged ≥ 6 years in the case of etanercept and ≥ 12 years in the case of ustekinumab. In this scenario, the off-label use of the biologics outside their age constraints and positioning in the management pathway is considered.

In the absence of clinical effectiveness evidence in a systemic therapy-naive population, the same efficacy estimates as in the base-case analysis were used in this scenario. Therefore, the only difference between this scenario and the base-case assumptions is the comparator, which is methotrexate in the analysis that considers biologics as an alternative to systemic therapy, and the time horizon of the model, which extends to 14 years because the starting age in the model is now 4 years.

Table 74 presents the cost-effectiveness results for the use of the interventions as an alternative to systemic therapy for all ages (4–17 years). Ustekinumab is the most effective treatment, followed by adalimumab, etanercept and methotrexate, as the efficacy of the treatments follow in this order. In terms of costs, ustekinumab is the most costly treatment, followed by adalimumab, etanercept and methotrexate. The reason for this ordering is the same as in the base-case results, with ustekinumab costing £715.67 per 4-week cycle compared with a cost per cycle of £704.28 for adalimumab, £357.50 and £715 for etanercept for those aged < 10 years and ≥ 10 years, respectively, and approximately £60 for methotrexate, with the reduction in costs associated with improved efficacy (i.e. less time spent on BSC) not sufficient to offset the additional treatment costs. Based on a fully incremental analysis, the incremental costs of etanercept and adalimumab relative to methotrexate are greater for each additional gain in QALY than the incremental costs of ustekinumab relative to methotrexate for each QALY gain. Therefore, etanercept and adalimumab are extendedly dominated by ustekinumab. The ICER of ustekinumab compared with methotrexate is very high at £293,117 per QALY gained. As a result, the optimal treatment option in a systemic therapy-naive population is methotrexate.

TABLE 74

Scenario 1 results for interventions as an alternative to systemic therapy: off-label use of biologics outside age constraints and position in the pathway

Table 75 presents the cost-effectiveness results for treatment with the interventions after failed systemic therapy for all ages (4–17 years). The only difference between this scenario and the base-case analysis is the starting age of 4 years used in the model. The total absolute costs and QALYs are greater than in the base case because of the longer model time horizon of 14 years. The ordering of the treatments in terms of costs and QALYs follows that in the base case, with ustekinumab the most effective but most costly treatment, followed by adalimumab, etanercept and BSC. Based on a fully incremental analysis, adalimumab is extendedly dominated by ustekinumab. Compared with the base-case population aged 12–17 years, etanercept is no longer extendedly dominated because more individuals receive a lower dose of etanercept, at the cost of a 25-mg vial rather than a 50-mg vial. The ICERs are lower than in the base-case analysis but the optimal treatment option remains BSC. BSC is the optimal option until the threshold reaches £60,000 per QALY gained, when etanercept would then enter as the first treatment in the sequence.

TABLE 75

Scenario 1 results for treatment with the interventions after failed systemic therapy: off-label use of biologics outside age constraints

Model time horizon

Scenario 2: time horizon of the model

The time horizon of the model was chosen to reflect the fact that once individuals reach 18 years of age separate NICE recommendations for the use of the interventions in adults apply. To incorporate these recommendations, evidence on the efficacy of the treatments in biologic-experienced patients (i.e. effectiveness estimates conditional on previous biological therapy) would be required. This would involve modelling the sequential use of therapies, with every possible potential treatment sequence considered based on current recommendations in adults. As well as being outside the scope of this appraisal, this would represent a significant challenge for two reasons: first, there is very limited evidence on the efficacy of biologics when used in sequence, that is, in biologic-experienced patients, and, second, current NICE recommendations for the use of biologic therapies in moderate to severe psoriasis in adults have been informed by a series of STAs^98–102 rather than a MTA that establishes the optimal sequence of treatments in adults.

Furthermore, the differences in the marketing authorisations of the interventions by age inevitably mean that the time horizon of the model will differ according to age group. In this scenario, the impact of the time horizon was assessed by considering a common time horizon of 14 years for all age groups, but with the same starting age for each group as used in the base-case analysis. The time horizon of 14 years is sufficient to capture differences in costs and effects between the interventions under comparison because all individuals on each treatment in the model have moved to BSC by 14 years. This time horizon is also greater than the 10 years used in previous TAs in adults.

The base case already considers a time horizon of 14 years for adalimumab as an alternative to systemic therapy because the starting age is 4 years. Therefore, Table 76 presents the cost-effectiveness results for treatment with the interventions after failed systemic therapy for a common time horizon of 14 years. By extending the time horizon, the total costs and QALYs for the interventions are greater than in the base case, but the relative cost-effectiveness of the interventions remains the same, that is, the ICERs for each intervention relative to the next best treatment option or BSC are similar to those observed in the base case. Therefore, the model time horizon used in the base-case analysis is sufficient to capture the differences between the interventions in terms of costs and QALYs.

TABLE 76

Scenario 2 results for treatment with the interventions after failed systemic therapy: common time horizon of 14 years

Treatment effectiveness estimates

Scenario 3a: direct trial evidence for treatment effects in children and young people

As discussed in Chapter 4, a NMA was used in the base-case analysis to connect the evidence from the adalimumab trial (M04-717) in children and young people to the evidence from the etanercept (20030211) and ustekinumab (CADMUS) trials by drawing strength from the wider network of evidence in adults. In this scenario, the relative cost-effectiveness of adalimumab compared with methotrexate and of etanercept and ustekinumab compared with BSC is considered using the direct efficacy estimates derived from their corresponding trials. The limitation of this approach is that it does not allow the relative cost-effectiveness of all three biologics to be assessed in the same analysis. However, it may give an indication of how much influence the wider network of evidence has on the individual pairwise comparisons.

Table 77 presents the cost-effectiveness results for use of adalimumab as an alternative to systemic therapy using the efficacy estimates from the M04-717 trial alone. The incremental cost (£20,256) and QALYs (0.037) for adalimumab compared with methotrexate are lower than the base-case incremental cost (£27,084) and QALYs (0.088). The PASI 75 response rate is 58% for adalimumab and 32% for methotrexate in the M04-717 trial compared with 79% and 49%, respectively, in the NMA. The NMA estimates higher absolute values for PASI 75 response but the incremental difference between adalimumab and methotrexate is of a similar magnitude in the NMA (30% difference in PASI 75 response) and the M04-717 trial (26% difference in PASI 75 response). This smaller difference in relative effectiveness between adalimumab and methotrexate in the M04-717 trial means that the incremental cost for each additional gain in QALYs is greater for adalimumab compared with methotrexate. The resulting ICER increases from £308,329 per additional QALY in the base-case analysis to £549,899 per additional QALY using the direct trial evidence.

TABLE 77

Scenario 3a results for adalimumab as an alternative to systemic therapy: direct trial evidence for treatment effects in children and young people

Table 78 presents the cost-effectiveness results for treatment with the interventions after failed systemic therapy using the efficacy estimates from the etanercept trial (20030211) and the ustekinumab trial (CADMUS). The total costs and QALYs for etanercept and ustekinumab compared with BSC are very similar to those in the base case. This is because the PASI 75 response rates estimated from the NMA for etanercept (54%), ustekinumab (82%) and placebo (11.5%) are very similar to the corresponding response rates from the individual trials (CADMUS: 81% for ustekinumab vs. 11% for placebo; 20030211: 57% for etanercept vs. 11.4% for placebo). As a result, the pairwise ICERs for etanercept and ustekinumab compared with BSC are similar to those in the base-case analysis: the ICER for etanercept compared with BSC increases from £71,903 per QALY in the base-case analysis to £75,350 per QALY using the direct trial evidence and the ICER for ustekinumab compared with BSC increases marginally from £116,568 per QALY in the base-case analysis to £116,982 per QALY using the direct trial evidence.

TABLE 78

Scenario 3a results for treatment with the interventions after failed systemic therapy: direct trial evidence for treatment effects in children and young people

Scenario 3b: indirect treatment comparison estimates in children and young people

In this scenario, the relative cost-effectiveness of etanercept and ustekinumab compared with BSC is considered using the indirect treatment comparison estimates from the 20030211 and CADMUS trials, with placebo used as a common comparator. The limitation of this approach is that it does not allow the relative cost-effectiveness of etanercept and ustekinumab compared with adalimumab to be determined because of the absence of a placebo arm in the M04-717 trial.

Table 79 presents the cost-effectiveness results for treatment with the interventions after failed systemic therapy using efficacy estimates from an indirect treatment comparison of etanercept from the 20030211 trial and ustekinumab from the CADMUS trial. The total costs and QALYs for etanercept, ustekinumab and BSC are similar to those in the base-case analysis. This is expected as the efficacy estimates from the individual trials for these interventions are similar to those estimated in the NMA. Etanercept is extendedly dominated by ustekinumab as the incremental costs of etanercept relative to BSC are greater for each additional gain in QALYs than the incremental costs of ustekinumab relative to BSC for each QALY gain. This occurs because ustekinumab has a better efficacy (78% PASI 75 response) than etanercept (57% PASI 75 response), which results in improved health outcomes for ustekinumab. Interestingly, the total cost for ustekinumab is greater than that for etanercept despite the fact that the drug acquisition costs are similar between the two treatments in children and young people aged 12–17 years. This arises because, although the improved efficacy of ustekinumab reduces the time spent on BSC, it also means that a greater proportion of time is spent on a cost-ineffective treatment option. The ICER of ustekinumab compared with BSC is £119,092 per QALY gained. As a result, the optimal treatment option is BSC unless the cost-effectiveness threshold reaches £120,000 per additional QALY.

TABLE 79

Scenario 3b results for treatment with the interventions after failed systemic therapy: indirect treatment comparison estimates in children and young people

Scenario 3c: treatment effects from the network meta-analysis using minimum evidence from the adult population

In Chapter 4 the disconnected network of evidence in children and young people was connected in the first instance by bringing together the minimum amount of evidence required from the adult population to link the adalimumab trial with the other paediatric trials. The CHAMPION study in adults,¹⁰⁶ which was a three-arm trial comparing adalimumab, methotrexate and placebo, represented the best way of connecting adalimumab to etanercept and ustekinumab using the least amount of evidence borrowed from the adult population. In this scenario, the relative cost-effectiveness of the interventions was considered in the base-case populations using the treatment effects estimated from the minimum network of evidence.

Table 80 presents the cost-effectiveness results for adalimumab as an alternative to systemic therapy using the NMA with minimum links to the adult evidence. The incremental cost of £18,422 for adalimumab compared with methotrexate is lower than the incremental cost of £27,084 in the base-case analysis. This is because of a larger difference in PASI 75 response rates between adalimumab and methotrexate in the minimum NMA (approximately 40% difference) than in the full network of evidence (approximately 30% difference). Although there is a higher efficacy difference between adalimumab and methotrexate in this scenario, the health outcomes also depend on the utility associated with BSC, which is based on the proportion of individuals in the different PASI response categories in the placebo arm in the NMA. The PASI response rates for placebo are greater in the minimum NMA than in the full network. Therefore, the gain in utility associated with better efficacy on adalimumab is offset by a higher gain in utility associated with BSC. As a result, the incremental QALYs for adalimumab compared with methotrexate are very similar to those in the base-case analysis. The corresponding ICER for adalimumab compared with methotrexate is reduced from £308,329 per additional QALY in the base case to £211,259 per additional QALY.

TABLE 80

Scenario 3c results for adalimumab as an alternative to systemic therapy: treatment effects from the NMA using minimum evidence from the adult population

Table 81 presents the cost-effectiveness results for treatment with the interventions after failed systemic therapy using the NMA with minimum links to the adult evidence. The incremental costs and QALYs for etanercept and ustekinumab compared with BSC are similar to those in the base-case analysis, but the incremental costs and QALYs for adalimumab are reduced in both age groups. This is because the differences in PASI 75 response rate between the interventions and BSC in the minimum NMA and the full NMA are similar for etanercept (44% vs. 43%) and ustekinumab (66% vs. 71%) but are much smaller for adalimumab (44% vs. 68%). As a result, adalimumab is less cost-effective in children and young people aged 6–11 years (ICER vs. BSC increases from £115,825 per additional QALY in the base case to £137,329 per additional QALY) and is extendedly dominated by ustekinumab in the 12–17 years age group. The ICER for etanercept is reduced by £3400 in children aged 6–11 years, but etanercept is also extendedly dominated by ustekinumab in children aged 12–17 years. The ICER for ustekinumab compared with BSC increases slightly from the base-case value of £116,568 per QALY gained to £118,515 per QALY gained using the minimum NMA.

TABLE 81

Scenario 3c results for treatment with the interventions after failed systemic therapy: treatment effects from the NMA using minimum evidence from the adult population

Scenario 3d: Psoriasis Area and Severity Index response assessment

In this scenario, PASI 50 is considered as the primary efficacy end point for response assessment at the end of the trial period instead of PASI 75, as used in the base-case analysis.

Table 82 presents the cost-effectiveness results for adalimumab as an alternative to systemic therapy using PASI 50 as the primary efficacy end point. The incremental costs and QALYs for adalimumab compared with methotrexate increase compared with the base case because there is a smaller difference in PASI 50 response rates between the interventions (91.5% adalimumab vs. 71% methotrexate) than for PASI 75 response rates (79% adalimumab vs. 49% methotrexate). As a result, the ICER increases from £308,329 per QALY gained to £353,148 per QALY gained.

TABLE 82

Scenario 3d results for adalimumab as an alternative to systemic therapy: PASI 50 response assessment

Table 83 presents the cost-effectiveness results for treatment with the interventions after failed systemic therapy using PASI 50 as the primary efficacy end point. The incremental cost per additional QALY gained is greater for all interventions than in the base-case analysis. This is because the total costs have increased (a greater proportion of individuals continue treatment as responders) but the total QALYs have decreased across the interventions. The difference in PASI 50 response rate between the interventions and BSC is similar to the difference observed in PASI 75 response rates. The decrease in QALYs results from the proportionally smaller utility gain associated with the PASI 50–75 response category than with the PASI 75–90 and PASI ≥ 90 response categories. BSC remains the optimal treatment option and the probability that any of the biologics are cost-effective at a threshold of £30,000 per additional QALY is zero.

TABLE 83

Scenario 3d results for treatment with the interventions after failed systemic therapy: PASI 50 response assessment

Health-related quality-of-life utility values

Scenario 4a: EuroQol-5 Dimensions utility estimates from adults

The HRQoL utility values in children and young people are subject to considerable uncertainty. EQ-5D-Y values mapped from PedsQL data from the CADMUS trial (ustekinumab) at baseline and 12 weeks’ follow-up were used to estimate utility gains from baseline associated with different PASI response categories (< 50, 50–74, 75–89, ≥ 90). The utility values associated with treatment were then based on the proportion of individuals in the different PASI response categories from the NMA and the associated utility gain for each PASI category. As discussed in Utility estimates used in the model, the estimated EQ-5D-Y utility gains mapped from the PedsQL data were of a much smaller magnitude than the EQ-5D values used in previous TAs in adults.^97–102 It was also noted that the gains in CDLQI score by PASI response category were of a smaller magnitude than the DLQI values reported in adults. It is not clear whether these smaller utility increments observed in children and young people are a reflection of a lower impact of severe psoriasis on quality of life in a paediatric population or a result of the small sample sizes and the limited data in this population.

In this scenario, EQ-5D utility values from the adult population were used to inform the gains in utility associated with PASI response in children and young people. Utility values from TA103⁹⁷ (etanercept) were used; however, the implications of using alternative adult utility values from TA146⁹⁹ (adalimumab) and TA180¹⁰⁰ (ustekinumab) were also considered (see Table 62 for a comparison of utility values in children and young people and adults).

Table 84 presents the cost-effectiveness results for adalimumab as an alternative to systemic therapy using utility estimates from an adult population. The total QALYs for the interventions are lower than those in the base-case analysis, but this is because of the use of a lower baseline utility value in this scenario to prevent the utility values rising above 1.0. Note that changing the baseline utility value used in the model does not significantly affect the cost-effectiveness results because the model is driven by the incremental changes in utility from baseline. The incremental QALYs of 0.150 for adalimumab compared with methotrexate are significantly higher than the incremental QALYs of 0.088 in the base case. As a result, the ICER for adalimumab compared with methotrexate reduces from £308,329 to £180,773 per additional QALY. The implications of using adult utility values from TA180¹⁰⁰ and TA146⁹⁹ are even more pronounced, with an incremental gain in QALYs of 0.204 and 0.260, respectively, for adalimumab compared with methotrexate, resulting in corresponding ICERs of £132,616 and £104,010 per additional QALY (see Appendix 8 for results based on utility estimates from TA180¹⁰⁰).

TABLE 84

Scenario 4a results for adalimumab as an alternative to systemic therapy: EQ-5D utility estimates from adults

Table 85 presents the cost-effectiveness results for treatment with the interventions after failed systemic therapy using utility estimates from an adult population. The incremental QALYs for the interventions compared with BSC are substantially greater than those in the base case. Ustekinumab is the most effective intervention, followed by adalimumab, etanercept and BSC, and the incremental gain in QALYs from moving from one intervention to the next is greater than in the base case. As a result, all of the ICERs are substantially lower than in the base case, falling by 55–61%. Etanercept shows the largest reduction in the ICER and, at a threshold of £30,000 per additional QALY, etanercept becomes the optimal treatment in the 6–11 years age group. In the 12–17 years age group, etanercept is extendedly dominated by adalimumab because of the higher drug acquisition costs associated with this age group requiring more than a 25-mg dose. In those aged 12–17 years, the optimal treatment option remains BSC up until a threshold of £51,000 per QALY gained, when adalimumab would then enter as the first treatment in the sequence. At a threshold of £60,000 per QALY, adalimumab represents the only cost-effective treatment option based on a fully incremental analysis, whereas all of the biologics would be considered cost-effective based on a pairwise comparison with BSC.

TABLE 85

Scenario 4a results for treatment with the interventions after failed systemic therapy: EQ-5D utility estimates from adults

The implications of using adult utility values from TA180¹⁰⁰ and TA146⁹⁹ are even more pronounced than when using utility gains from TA103⁹⁷ because of the greater utility gains in the PASI 75–89 and ≥ 90 categories. The ICERs for children and young people aged 6–11 years are £22,578 (TA146⁹⁹) and £21,546 (TA180¹⁰⁰) for etanercept compared with BSC and £37,125 (TA146⁹⁹) and £39,682 (TA180¹⁰⁰) for adalimumab compared with BSC. The lowest ICERs for children and young people aged 12–17 years are £33,517 for adalimumab compared with BSC, £35,612 for ustekinumab compared with BSC and £39,247 for etanercept compared with BSC.

Scenario 4b: utility estimates for best supportive care

The base-case analysis assumes that the utility associated with BSC is based on the proportion of individuals in the different PASI response categories in the placebo arm of the NMA. In this scenario, the utility for BSC was set equal to the baseline value, that is, assuming that there is no utility gain associated with BSC.

Table 86 presents the cost-effectiveness results for adalimumab as an alternative to systemic therapy assuming that there is no utility benefit associated with BSC. For the comparison of adalimumab and methotrexate, the assumption of no utility benefit on BSC affects only the utility of non-responders. The total QALYs for both interventions are reduced and the incremental QALYs for adalimumab compared with methotrexate increase from 0.088 in the base case to 0.102 because of the higher efficacy of adalimumab, which reduces the time spent in BSC.

TABLE 86

Scenario 4b results for adalimumab as an alternative to systemic therapy: utility in BSC equal to the baseline value

Table 87 presents the cost-effectiveness results for treatment with the interventions after failed systemic therapy assuming that there is no utility benefit associated with BSC. This assumption reduces the total QALYs for the comparator of BSC and the utility of non-responders. As a result, the incremental QALYs for the interventions compared with BSC increase and the incremental gain in QALYs for the interventions relative to the next best option (e.g. ustekinumab is the most effective treatment, followed by adalimumab and etanercept) also increases as less time is spent on BSC. Consequently, the ICERs for the interventions are reduced compared with the base-case values.

TABLE 87

Scenario 4b results for treatment with the interventions after failed systemic therapy: utility in BSC equal to the baseline value

Costs associated with best supportive care

Scenario 5a: number of hospitalisations per annum for best supportive care

The resource use associated with BSC, in particular the number of hospitalisations per annum, was identified as an area of high uncertainty and a key driver of cost-effectiveness in previous TAs in adults.^97–102 Two main sources have been referred to in previous appraisals: (1) NICE CG153,¹⁷³ in which an average of 26.6 inpatient days per year was estimated for individuals whose psoriasis has not responded to treatment, and (2) Fonia et al.,¹⁴⁴ who estimated an average of 6.49 days of hospitalisation per annum. During previous NICE appraisals, the clinical experts considered that both sources are likely to overestimate the actual number of hospital days and resource use associated with BSC. This is in part because of the populations considered in CG153¹⁷³ and the study by Fonia et al.,¹⁴⁴ with CG153¹⁷³ considering a high-need population with very severe psoriasis and the study by Fonia et al.¹⁴⁴ describing care in a tertiary care centre known for treating the most severely affected individuals. The clinical experts in recent appraisals also noted that the number of individuals hospitalised for severe psoriasis has fallen over time and is continuing to fall. They also indicated that BSC is mostly given to individuals during their outpatient visits. As a result, the resource use associated with BSC is an area of considerable uncertainty and both sources of data have a number of shortcomings, even in the adult population.

In the base-case analysis in children and young people, it was assumed that there are no hospitalisations for psoriasis in this population. This was informed by clinical opinion, with our clinical advisor suggesting that hospitalisations in children and young people are very rare, partly because this population has not yet developed the comorbidities that often complicate more severe cases of psoriasis in adults. In this scenario, the implications of assuming no inpatient stays for children and young people were explored by using an estimate of 6.49 hospitalisations per annum based on the study by Fonia et al.¹⁴⁴ and 26.6 hospitalisations per annum based on CG153¹⁷³ in adults.

Table 88 presents the cost-effectiveness results for adalimumab as an alternative to systemic therapy, assuming hospitalisations for BSC. For the comparison of adalimumab and methotrexate, the total costs for both interventions are increased, but the incremental cost of adalimumab compared with methotrexate decreases because of the higher efficacy associated with adalimumab, which reduces the time spent in BSC. The resulting ICER decreases from £308,329 per additional QALY in the base case to £281,029 per additional QALY for 6.49 inpatient days per annum and £202,571 per additional QALY for 26.6 inpatient days per annum.

TABLE 88

Scenario 5a results for adalimumab as an alternative to systemic therapy: number of hospitalisations per annum for BSC

Table 89 presents the cost-effectiveness results for treatment with the interventions after failed systemic therapy, assuming hospitalisations for BSC. Under this assumption, the costs of BSC increase by £147.67 and £605.25 per 4-week cycle for 6.49 and 26.6 inpatient days per annum respectively. As a result, the total costs associated with the comparator of BSC increase and the costs for non-responders increase. For children and young people aged 6–11 years, the reduction in the incremental cost of etanercept compared with BSC is sufficient to make etanercept the optimal treatment option at a threshold of £30,000 per QALY for a stay of 6.49 inpatient days per annum. When the hospitalisation LOS per annum is increased to 26.6 days in this age group, etanercept becomes the least costly treatment option and BSC becomes dominated by etanercept (i.e. BSC costs more than etanercept but produces fewer QALYs). Adalimumab enters as a cost-effective option only if the threshold increases to £70,000 per QALY gained.

TABLE 89

Scenario 5a results for treatment with the interventions after failed systemic therapy: number of hospitalisations per annum for BSC

For children and young people aged 12–17 years, etanercept is dominated by adalimumab. The ICER for adalimumab compared with BSC is £74,501 per QALY when 6.49 inpatient days per annum are assumed. When 26.6 inpatient days per annum are assumed, adalimumab becomes the least costly treatment option and the most cost-effective option at a threshold of £30,000 per QALY. The ICER for ustekinumab compared with adalimumab is £118,665 per QALY for a stay of 26.6 inpatient days per annum.

Costs associated with adverse events

Scenario 6: costs of severe infections and malignancies

In the absence of robust evidence on the incidence of AEs associated with treatment in children and young people, the base-case analysis assumed that there were no AEs associated with treatment. In this scenario, the costs associated with SAEs, including NMSC, malignancies other than NMSC and severe infections, are included. These events are expected to be very rare.

Table 90 presents the cost-effectiveness results for adalimumab as an alternative to systemic therapy with the costs of AEs included. The incremental cost of adalimumab increases by only £400. The resulting impact on the ICER is minor, with an increase from £308,329 per QALY gained in the base case to £311,067 per QALY gained.

TABLE 90

Scenario 6 results for adalimumab as an alternative to systemic therapy: costs of severe infections and malignancies included

Table 91 presents the cost-effectiveness results for treatment with the interventions after failed systemic therapy with the costs of AEs included. As expected, the incremental costs for the interventions relative to BSC increase, but the resulting impact on the ICERs for all interventions is very minor.

TABLE 91

Scenario 6 results for treatment with the interventions after failed systemic therapy: costs of severe infections and malignancies included

Treatment withdrawal rates

Scenario 7: withdrawal rates from treatment

In the base-case analysis discontinuation from treatment is modelled as an all-cause withdrawal probability of 20% per annum, which is applied to all interventions. This withdrawal rate has been used in all previous TAs in adults^97–102 and is consistent with long-term survival rates of biologics from the BADBIR audit.⁹⁴ In the absence of alternative data in children and young people, this scenario considered two separate withdrawal rates of 10% and 30% per annum.

Table 92 presents the cost-effectiveness results for adalimumab as an alternative to systemic therapy for treatment withdrawal rates of 10% and 30% per annum. The lower withdrawal rate implies that individuals spend longer on treatment before moving to BSC, whereas the higher withdrawal rate means that individuals spend less time on treatment and more time on BSC. The total costs for adalimumab increase for the 10% rate and decrease for the 30% rate, whereas the total costs for methotrexate decrease for the 10% rate and increase for the 30% rate. This opposite effect between the treatments arises because the drug acquisition cost of adalimumab (£704.28 per 4-week cycle) is proportionally greater than the cost of BSC (approximately £284 per 4-week cycle) compared with the drug acquisition cost of methotrexate (approximately £60 per 4-week cycle) relative to the cost of BSC. As a result, the withdrawal rate has less impact on the total cost of methotrexate than on the total cost of adalimumab. The incremental costs of adalimumab compared with methotrexate are £40,781 and £19,692 for the 10% and 30% annual withdrawal rates, respectively, compared with the base-case incremental cost of £27,084. In terms of health outcomes, the more time spent on treatment the higher the utility gains; therefore, the QALYs increase for the lower withdrawal rate and decrease for the higher withdrawal rate. The resulting ICERs for adalimumab compared with methotrexate are £298,846 and £318,188 per additional QALY for the 10% and 30% annual withdrawal rates, respectively, compared with the base-case value of £308,329 per additional QALY.

TABLE 92

Scenario 7 results for adalimumab as an alternative to systemic therapy: treatment withdrawal rates of 10% and 30% per annum

Table 93 presents the cost-effectiveness results for treatment with the interventions after failed systemic therapy for treatment withdrawal rates of 10% and 30% per annum. The total costs for all of the interventions increase for the 10% rate and decrease for the 30% rate because of the accumulation of higher drug acquisition costs while on treatment for longer. This increase in costs is counterbalanced by an increase in utility gains while on treatment. The resulting impact on the ICERs is minimal. BSC remains the optimal treatment option at a threshold of £30,000 per QALY gained.

TABLE 93

Scenario 7 results for treatment with the interventions after failed systemic therapy: treatment withdrawal rates of 10% and 30% per annum

Biosimilars

Scenario 8: reduction in the cost of etanercept

The biosimilar of etanercept, Benepali (50 mg), does not have marketing authorisation for use in children and young people. In this scenario, the drug acquisition cost of etanercept was reduced by approximately 10% to match the cost of Benepali in adults.

Table 94 presents the cost-effectiveness results for treatment with the interventions after failed systemic therapy for a 10% reduction in the acquisition cost of etanercept. For children and young people aged 6–11 years, the incremental cost of etanercept relative to BSC is reduced by £580, which reduces the ICER from £71,903 to £66,240 per additional QALY. For children and young people aged 12–17 years, the incremental cost of etanercept relative to BSC is reduced by £1480, which is a greater reduction than that observed in the younger age group because it is assumed that the 10% reduction in the drug acquisition cost of etanercept applies only to children aged ≥ 10 years who require 50 mg of etanercept. The cost reduction has a very minor impact on the cost-effectiveness results.

TABLE 94

Scenario 8 results for treatment with the interventions after failed systemic therapy: reduction in the cost of etanercept to match the unit cost of Benepali