1 Appraisal Committee’s preliminary recommendations

1.1 Tocilizumab is not recommended for the treatment of systemic juvenile idiopathic arthritis in children and young people aged 2 to 17 years and older whose disease has responded inadequately to non-steroidal anti-inflammatory drugs (NSAIDs) and systemic corticosteroids.

1.2 The Committee is minded not to recommend tocilizumab for the treatment of systemic juvenile idiopathic arthritis in children and young people aged 2 years and older whose disease has responded inadequately to NSAIDs, systemic corticosteroids and methotrexate. The following further information on clinical and cost effectiveness should be provided by the manufacturer:

• A revised economic model that allows transitions between ACR (American College of Rheumatology)-response categories within a treatment line and:
  • restructures the health states defined by Childhood Health Assessment Questionnaire (CHAQ) scores and uses treatment response to define transition probabilities or
  • uses a patient-level simulation to define the cost and utility values of a health state depending on both the starting CHAQ score and the change in CHAQ score in relation to the ACR (American College of Rheumatology) response.
• A revised manufacturer’s base case with the assumption that treatment starts at age 5 years.
• For the comparison of tocilizumab and anakinra a revised base case using the primary outcome (ACR30 response and no fever).
• A fully incremental cost-effectiveness analysis of tocilizumab compared with anakinra and infliximab.
• Sensitivity analysis on the revised base cases that includes:
  • the uncertainty around the adjustment factor derived from the etanercept study used to take account of the other juvenile idiopathic arthritis subgroups in the infliximab study
  • a ‘stopping rule’ for tocilizumab after 2 years of treatment
  • a decreased frequency of administration of tocilizumab after 6 months to a 4-weekly regimen.
• Scenario analyses that include a fully incremental cost-effectiveness analysis of tocilizumab compared with anakinra and infliximab with:
  • infliximab as first-line treatment followed by tocilizumab
  • infliximab as first-line treatment followed by anakinra.

1.3 The Committee requests further information on:

• radiographic evidence of progression of joint damage for patients receiving tocilizumab
• long-term follow-up data from the trials being conducted in Japan
• how CHAQ responses were elicited from children aged under 5 years.

2 The technology

2.1 Tocilizumab (RoActemra, Roche Products) is a humanised monoclonal antibody that inhibits the cytokine interleukin-6 (IL-6). Reducing the activity of IL-6 can reduce inflammation in the joints, prevent long-term damage, and improve quality of life and function. ‘Tocilizumab has a marketing authorisation for the treatment of active systemic juvenile idiopathic arthritis (sJIA) in patients 2 years of age and older who have responded inadequately to previous therapy with non-steroidal anti-inflammatory drugs (NSAIDs) and systemic corticosteroids. Tocilizumab can be given as monotherapy (in case of intolerance to methotrexate or where treatment with methotrexate is inappropriate) or in combination with methotrexate’.     

2.2 Upper respiratory tract infection, with typical symptoms such as cough, blocked nose, runny nose, sore throat and headache, is one of the most common side effects of tocilizumab. Other reported side effects include rash, urticaria, diarrhoea, epigastric discomfort and arthralgia. Infusion-related reactions that can be considered serious and life-threatening (such as angioedema) have also been reported. For full details of side effects and contraindications, see the summary of product characteristics.

2.3 Tocilizumab is administered as an intravenous infusion over 1 hour and treatment is repeated at 2-week intervals. The recommended dose is 8 mg/kg in patients weighing 30 kg or more, and 12 mg/kg in patients weighing less than 30 kg. The dose should be calculated based on the patient’s body weight at each administration. A change in dose should only be based on a consistent change in the patient’s body weight over time. The cost of 80 mg in 4 ml vial is £102.40 (excluding VAT; ‘British national formulary’ [BNF] edition 61). The average cost of treatment is £7987.20 per year for a 30 kg patient and £9984 per year for a 25 kg patient, assuming no wastage. Costs may vary in different settings because of negotiated procurement discounts.

3 The manufacturer’s submission

The Appraisal Committee (appendix A) considered evidence submitted by the manufacturer of tocilizumab and a review of this submission by the Evidence Review Group (ERG; appendix B).

3.1 The manufacturer submitted evidence for the two populations defined in the decision problem: population 1 – children and young people aged 2 years and older with systemic JIA that has not responded adequately to prior NSAIDs and systemic corticosteroids; and population 2 – children and young people aged 2 years and older with systemic JIA that has not responded adequately to prior NSAIDs, systemic corticosteroids and methotrexate. For population 1 the manufacturer compared tocilizumab with methotrexate. For population 2 the manufacturer carried out indirect comparisons of tocilizumab with tumour necrosis factor-alpha (TNF-alpha) inhibitors and anakinra.

3.2 In the manufacturer’s submission, evidence of clinical effectiveness was based on one randomised controlled trial (TENDER). The manufacturer stated that by viewing the inclusion criteria of the TENDER trial, the whole TENDER population matched population 1. The manufacturer also stated that 95% of TENDER trial participants match population 2 because ‘patients are included in the study if they have symptoms of active disease’ and ‘it follows that if patients have tried in the past or are currently administered methotrexate and continue to have persistent disease then they are inadequate responders’. An inadequate response to methotrexate required patients to be on a standard dose of methotrexate for a period of 3 months and still show symptoms of active systemic JIA at baseline.

3.3 The TENDER trial is an ongoing three-part, 5-year, phase III study. Part one consisted of a 12-week international multicentre randomised double-blind placebo-controlled parallel two-group study to evaluate the efficacy and safety of tocilizumab in children with active systemic JIA. The study enrolled 112 participants (from 17 countries, including the UK) who were unequally randomised 2:1 to tocilizumab (n = 75) or placebo (n = 37). Tocilizumab was administered every 2 weeks with a dose of 8 mg/kg for participants who weighed at least 30 kg (n = 37) and 12 mg/kg for those who weighed less than 30 kg (n = 38). Part two is a 92-week single-group open-label extension and part three is a 3-year single-group open-label continuation of the study. Ages of patients in the trial ranged from 2 to 17 years, with an average age of 10 years. Patients had to have documented persistent disease activity (at least five active joints, or at least two active joints with fever above 38°C for any 5 out of 14 days of screening) for at least 6 months, and an inadequate response to NSAIDs and corticosteroids because of toxicity or lack of efficacy. An inadequate response to previous treatment was determined by the treating physician’s clinical assessment. Before study entry, 78 out of 112 patients (70%) had been treated with methotrexate (36 entered the study on methotrexate that had been previously stopped then restarted; 42 were on their first course of methotrexate,which was ongoing). Twenty-nine patients (approximately 26%) had no background methotrexate at baseline but had received and stopped methotrexatepreviously. Five patients (approximately 4%) had never received methotrexate, and could be considered methotrexate naive. Patients taking NSAIDs, corticosteroids and methotrexate were permitted to take part but had to enter the study on a stable dose of the medicines.

3.4 The primary outcome measures were the proportion of patients achieving a JIA ACR30 response at 12 weeks and absence of fever (defined as no recorded temperature of 37.5°C or above in the preceding 7 days). A JIA ACR30 response is defined as an improvement of at least 30% from the baseline assessments in any three of six core outcome variables, with no more than one of the remaining variables deteriorating by more than 30%. The JIA core outcome variables are: physician global assessment of disease activity (100 mm visual analogue scale [VAS]); parent or patient global assessment of overall well-being (100 mm VAS); number of joints with active arthritis; number of joints with limitation of movement; erythrocyte sedimentation rate; and functional ability (using the Childhood Health Assessment Questionnaire [CHAQ], which measures eight everyday functional activities).

3.5 he secondary outcomes were: individual results for each JIA ACR core outcome variable at 12 weeks; JIA ACR 50/70/90 responses at 12 weeks (that is, an improvement of at least 50%, 70% or 90% respectively from the baseline assessments in any three of the six core outcome variables, and no more than one of the remaining variables worsening by more than 50%, 70% or 90%); corticosteroid reduction; fever; rash; pain; and laboratory outcomes (C-reactive protein [CRP]) levels, anaemia and haemoglobin levels, thrombocytosis and leucocytosis).

3.6 Efficacy endpoints were analysed using the intention-to-treat population. All patients were classified as either responders or non-responders, and those patients who withdrew or escaped were classed as non-responders. There was an ‘early escape’ option to allow children with more severe disease at baseline an opportunity to escape and receive active open-label tocilizumab. A total of 21 patients out of the 112 patients enrolled received escape therapy, with 20 of those patients being initially randomised to the placebo arm. The main reasons for escape were fever for at least 3 consecutive days or a JIA ACR30 flare (a worsening of symptoms).

3.7 The results of the TENDER trial showed that for its primary endpoint (a JIA ACR30 response and absence of fever at week 12), 85.3% of the tocilizumab patients were classed as responders compared with 24.3% of the placebo patients, a statistically significant difference (p < 0.0001). Tocilizumab patients had a greater chance of achieving JIA ACR30/50/70/90 responses at week 12 in comparison with the placebo patients. The differences in the proportions of tocilizumab patients and placebo patients at each JIA ACR response level were statistically significant (p < 0.0001). The proportion of responders showing an ACR30 response was higher in patients receiving tocilizumab 12 mg/kg (97.4%) compared with those receiving tocilizumab 8 mg/kg (83.8%). The efficacy of tocilizumab with respect to individual ACR core outcome variables was analysed as part of the secondary efficacy analyses; these results are marked by the manufacturer of tocilizumab as academic in confidence and therefore are not presented here.

3.8 The TENDER trial also included the Child Health Questionnaire (CHQ) as an instrument eliciting patient health-related quality of life. The CHQ assesses a child's physical, emotional and social well-being from the perspective of a parent or guardian. The questionnaire was completed twice during the randomised period of the study: at baseline (visit 1) and at week 12 (visit 7).

3.9 The adverse event rate in the TENDER trial data was marked as commercial in confidence by the manufacturer of tocilizumab and therefore is not presented here. Infusion-related reactions are defined as all events occurring during or within 24 hours of an infusion. In the 12-week controlled phase, 4% of patients from the tocilizumab group experienced adverse events during infusion. One event (angioedema) was considered serious and life-threatening, and the patient was discontinued from study treatment. In the 12-week controlled phase, 16% of patients in the tocilizumab group and 5.4% of patients in the placebo group experienced an adverse event within 24 hours of infusion. In the tocilizumab group, the adverse events included, but were not limited to, rash, urticaria, diarrhoea, epigastric discomfort, arthralgia and headache. One of these adverse events, urticaria, was considered serious. Clinically significant hypersensitivity reactions associated with tocilizumab and requiring treatment discontinuation were reported in 1 out of 112 patients (less than 1%) treated with tocilizumab during the controlled phase and up to and including the open-label clinical trial.

3.10 For the comparison of tocilizumab and methotrexate for population 1, the manufacturer used a post-hoc analysis to compare patients receiving tocilizumab with the 70% of patients in the placebo group who were receiving methotrexate. The manufacturer presented results that showed methotrexate had limited effect on the primary outcome in patients who were on placebo and those on tocilizumab. The manufacturer concluded that methotrexate as add-on therapy did not have a significant impact on the JIA ACR responses observed in the tocilizumab arms in the TENDER study. The manufacturer further presented results showing that the proportion of patients on tocilizumab achieving ACR30 was 0.907 compared with those on methotrexate which was 0.154.

3.11 No head-to-head trials were available analysing the efficacy of tocilizumab compared with TNF-alpha inhibitors or anakinra for population 2. The manufacturer included data from two studies (Ruperto et al. 2007 [NCT00036374] and the ANAJIS [anakinra in patients with systemic-onset juvenile idiopathic arthritis] study) in the indirect comparison analysis. The NCT00036374 trial compared the TNF-alpha inhibitor infliximab with placebo in patients with juvenile rheumatoid arthritis (systemic 16%, pauciarticular 23%, polyarticular 61%) described as having a suboptimal response to methotrexate. Participants were from North and South America and Europe, aged between 4 and 18 years, and were randomised to infliximab (62 patients) or placebo (60 patients). Patients received concomitant methotrexate alongside placebo or active treatment. The study was a randomised double-blind placebo-controlled trial, and the primary outcome was the proportion of patients meeting a paediatric ACR30 response based on JIA core outcome variables at week 14.

3.12 The ANAJIS trial recruited children with systemic JIA and compared anakinra with placebo. This was a multicentre study with 24 participants (12 in each arm) aged 2 to 20 years, from North America and Europe. The study included patients whose systemic JIA had not responded to methotrexate and any of the disease-modifying anti-rheumatic drugs (DMARDs), and the protocol did not permit the administration of any DMARDs for the duration of the trial. The outcomes of the randomised controlled phase were reported after 1 month. The primary outcome was the paediatric ACR score, absence of fever and normalisation of CRP levels and erythrocyte sedimentation rate after 1 month.

3.13 The manufacturer conducted some analysis for the indirect comparison of tocilizumab with anakinra. Data from the ANAJIS and the TENDER trials were used. The manufacturer used all patients in the TENDER trial, including those who were methotrexate naive, for the analysis. The relative risk for achieving an outcome of a JIA ACR30 response for patients on tocilizumab compared with anakinra was 2.37 (95% CI 1.10 to 5.10), which was statistically significant. There was no significant difference in achieving a JIA ACR30 response and absence of fever between the anakinra and tocilizumab populations. The manufacturer also conducted an indirect comparison of tocilizumab and infliximab using the results from NCT00036374 trial and the TENDER trial. The outcomes JIA ACR30, 50 and 70 responses were measured. Patients on tocilizumab showed a statistically significantly greater chance of achieving these outcomes than those on infliximab. The relative risks were 2.87 (95% CI 1.49 to 5.55), 5.35 (95% CI 1.91 to 14.97) and 4.61 (95% CI 1.16 to 18.38) for JIA ACR30, 50 and 70 responses respectively.

3.14 The manufacturer used an adjustment factor derived from a study of etanercept by Prince et al. 2009. This was an observational study of 146 patients, of whom 27% had systemic JIA. The adjustment factor is the difference in the proportion of responders between the total population with JIA and the subpopulation with systemic JIA. This factor was used to correct for ACR response rates in the indirect comparison results which the manufacturer had derived from the NCT00036374 (infliximab) study (in which 16% of JIA patients had systemic JIA) and the TENDER trial. These resulting ACR response rates were assumed to represent the responses achieved with all of the TNF-alpha inhibitors.

3.15 The manufacturer submitted a Markov model to evaluate the cost effectiveness of tocilizumab as part of a sequence of treatments. In the tocilizumab versus methotrexate model, patients progressed to anakinra, etanercept and then adalimumab; in the tocilizumab versus anakinra model, patients progressed to etanercept, adalimumab and then abatacept.

3.16 The Markov chain has 22 states. The model clusters the states into five groups: four groups represent different lines of treatment and the fifth group contains death and uncontrolled disease. Each line of treatment consists of five health states: ACR responses at the 30, 50, 70 and 90 levels and ‘no ACR response’. A patient can move from a particular ACR response in a particular line only to ‘no ACR response’ in the next line or to death. From ‘no ACR response’ the patient can move only to one ACR response level within this line of treatment or to ‘no ACR response’ in the next line. The main assumption of the model is that there were no transitions between ACR response categories (that is, the patient cannot move within a given line to a better or worse health state [say, from ACR50 to ACR70]). The analysis assumes that patients stay in the same health state unless they change treatment line. After 12 weeks of treatment, the cohort is put on the next treatment in the sequence. Only after being through all four lines does a patient move to the health state ‘uncontrolled disease’. The probability of a response or non-response within a line of treatment depends on the treatment. The order in which the treatments are applied does not change these transitions. The probability of death is treatment-independent and health-state-independent. The probability of withdrawal is health-state-independent, but is higher for methotrexate than for other treatment options (all other treatment options have the same probability as each other). All transitions stay constant over time; that is, they are independent of age or disease duration. In each cycle, the proportion of patients in a given state is calculated. The distribution across states is used to calculate cycle-specific quality-adjusted life years (QALYs) and treatment costs, which are discounted and summed over the length of treatment. The manufacturer’s model has a time horizon of 16 years. This means that a patient in the model starting treatment aged 2 years turns 18 and can be considered an adult at the end of the simulation. The model allows shorter and longer time durations for sensitivity analysis (up to 30 years). The discount rates applied were 3.5% for utilities and costs, and costs are considered from an NHS and personal social services perspective. A half-cycle correction was applied.

3.17 The initial CHAQ score at baseline for the cohort of patients used in the economic model is equal to that observed in the TENDER trial. The change in the patient CHAQ score is determined by the level of ACR response after 12 weeks. Improvement in each health state as measured by relative ACR change leads to an absolute change in the initial CHAQ score. For a given CHAQ score, a utility is assigned to calculate QALYs. The health-state costs vary with the health state and the treatment costs.

3.18 The data inputs for the manufacturer’s model included utility values. To derive utility values, the manufacturer had to map the CHAQ scores to utilities, using a mapping formula derived in adults with rheumatoid arthritis that mapped Health Assessment Questionnaire [HAQ] results onto EQ-5D utilities. The manufacturer recognised that the assumptions that CHAQ is equal to HAQ and that adult EQ-5D is equal to the health-related quality of life of a child are not evidence based, and acknowledged that this mapping method was only preferred for the analysis to derive QALYs for the economic model because of the lack of other available data. The manufacturer stated that health states reflect the condition of patients dependent on ACR response after a 12-week period on treatment. The utility of the health state is characterised by the resulting CHAQ triggered by the ACR response. The utilities for each health state are based on the average CHAQ score per ACR response state, combined with the mapping formula. The CHAQ scores per ACR response were 1.7442 for no response or uncontrolled disease (with as assumed quality of life value of 0.4152) and 1.2699, 1.1351, 0.8601 and 0.6692 for ACR 30, 50, 70 and 90 responses respectively (with corresponding quality of life values of 0.5674, 0.6050, 0.6736 and 0.7150 respectively).  

3.19 Treatment costs in the model are a composite of the cost of the medication and the cost of administering it. For some drugs, the required dosage depends on the body weight of the patient. The manufacturer based the unit costs on UK reference costs, literature and expert opinion. The health-state costs depend only on the ACR response level and are independent from any other health outcomes. The manufacturer stated that, in all comparisons, the identified adverse events are of minor severity and short duration, and their management would have a minuscule cost impact. Therefore, it can be assumed that they do not have a considerable bearing on the incremental cost of the two model arms.

3.20 The manufacturer presented its base case showing that treatment with tocilizumab compared with methotrexate resulted in an incremental cost-effectiveness ratio (ICER) of £20,806 per QALY gained. The base case for tocilizumab compared with anakinra resulted in an ICER of £23,219 per QALY gained. When tocilizumab was compared with etanercept instead of anakinra as part of the scenario analysis, the ICER was £21,379 per QALY gained. The manufacturer presented results based on the mean costs and QALYs. In comparison with methotrexate, the probability of tocilizumab being cost effective was 0.39 at £20,000 per QALY and 0.72 at £30,000 per QALY. When compared with anakinra, the probability of tocilizumab being cost effective was 0.38 at £20,000 per QALY and 0.63 at £30,000 per QALY. The manufacturer undertook deterministic sensitivity analysis and scenario analysis to examine how varying various parameters and assumptions affected the robustness of the model. The manufacturer reported that doubling infusion administration costs had a substantial effect on the ICER – increasing it to £57,350 per QALY gained in the tocilizumab versus anakinra model. Another parameter that seemed to be a key driver in the cost-effectiveness analysis was the cost of inpatient stay. Reducing the cost of inpatient stay by 50% increased the ICER to £37,491 per QALY gained in the methotrexate strategy and £39,765 per QALY gained in the anakinra strategy. Varying the starting age in the model progressively increased the ICER. When the starting age was 10 years in the tocilizumab versus methotrexate model, the ICER was £46,844 per QALY; in the comparison of tocilizumab with anakinra, the ICER increased to £46,369 per QALY.

3.21 The ERG noted that the TENDER trial compares tocilizumab plus standard care with placebo plus standard care. The ERG observed that the comparator in this study did not match that specified in the scope and decision problem. For population 1 the comparator in the scope is methotrexate. The manufacturer had used a post-hoc analysis to compare patients receiving tocilizumab with those patients in the placebo group also receiving methotrexate. The ERG noted that this was not methodologically acceptable because the trial participants were not originally randomised into those populations. In the TENDER trial, 5% of patients were methotrexate naive. The ERG considered that this population would represent population 1 in the decision problem, but the analyses were inadequate. The ERG thus considered that there was no evidence for any comparison of tocilizumab with methotrexate.

3.22 For population 2 (children with systemic JIA whose condition has an inadequate response to NSAIDs, corticosteroids and methotrexate) the manufacturer’s submission provided data for an indirect comparison of tocilizumab with anakinra, using data from the TENDER trial and a trial of anakinra versus placebo. The ERG believed that the 5% of participants in the TENDER trial who were methotrexate naive should be excluded from these analyses. The manufacturer’s submission provided data only for all participants in the TENDER trial. However, in response to the request for clarification, some data were provided in which methotrexate-naive patients were excluded. These data were not reported for the TENDER trial, but only for the indirect comparison with anakinra. Where possible, the ERG used data for this population.

3.23 For the comparators, the ERG noted that the manufacturer had decided to broaden the inclusion criteria to include all subtypes of juvenile arthritis, not just systemic JIA. The manufacturer had taken this approach because of the dearth of clinical evidence for systemic JIA. The ERG was concerned that this approach had been taken despite the manufacturer’s clinical specialists stressing the differences between systemic JIA and other subtypes and advising against comparing the evidence from different JIA populations.

3.24 The ERG noted that the current economic model does not adhere to conventions in Markov modelling. In a Markov cohort model the health states defined should comprise the full range of conditions that are relevant to a patient population, and the states should be mutually exclusive. In the manufacturer’s submission, the health states were defined to reflect a change in a patient’s condition (change in CHAQ score based on ACR response) instead of the absolute condition of the patient. The ERG further noted that the change in a patient’s condition should be included in a Markov model as a health-state transition rather than a health state. The consequence of using a change in a patient’s condition as a health state is that the Markov states are heterogeneous rather than mutually exclusive, depending on the disease variation of the cohort at the start of the model.

3.25 The ERG also noted the assumption in the model that patients move to a certain ACR response and stay in that state until they either withdraw (move to the next treatment line) or die. The ERG thought that, given the nature of the disease, this assumption was unlikely to be correct.

3.26 The ERG noted the lack of health-related quality of life data both in the TENDER trial and in the literature, and recognised that very large assumptions (such as assuming that the CHAQ score of a child is equal to the HAQ score of an adult and that adult EQ-5D is equivalent to the health-related quality of life of a child) were needed to assign a utility to each health state in the model. Because of the lack of data in the trial and the literature, the ERG considered the approach used by the manufacturer to be reasonable and acceptable.

3.27 The ERG noted that the manufacturer assumed that the CHAQ scores for ACR response found in the TENDER trial are also valid for the anakinra and tocilizumab comparison. The ERG found it difficult to judge whether this is a reasonable assumption. The baseline CHAQ score in the anakinra group is slightly lower than that observed in the TENDER trial, and the baseline score in the infliximab study is much lower. The latter reflects the fact that the infliximab study included patients with a number of JIA subtypes, of which 16% had systemic JIA.

3.28 The ERG considered the sensitivity analyses performed on the utilities to be limited. The CHAQ scores, which are directly mapped into utilities, are varied only slightly. Only the starting CHAQ score (base case: 1.7 ± 0.8) is varied slightly (to 1.63, 1.73 and 2) to reflect the mean starting CHAQ score when accounting for different subgroups. In response to the request for clarification, the manufacturer stated that the way in which absolute CHAQ scores are modelled leads to the assumption that all patients have the same initial CHAQ score and means that all relative improvement leads to the same absolute (improved) CHAQ score. A clarifying example can be given by the initial CHAQ distribution, which has a mean of 1.7 ± 0.8. Assuming normality, this translates into 16% of all patients having an initial CHAQ score lower than 0.9 and 16% of all patients having an initial CHAQ score higher than 2.5. As a relative increase is modelled, the heterogeneity in the treatment health states is of a similar magnitude. The ERG highlighted that all these heterogeneities needed to be addressed.

3.29 The ERG also noted that the incremental change in CHAQ score and utilities between health states when having a treatment response, which is also affected by uncertainty, is part of neither the deterministic sensitivity analysis nor the probabilistic sensitivity analysis.

3.30 The ERG questioned the cost estimates for health states were defined by expert opinion because they present a cost for non-responders (£3300) that is more than six times higher than the cost for an ACR30 response (£500), whereas a ACR90 response is associated with only a 30% decrease in cost (to £350) compared with an ACR30 response.

3.31 The ERG stated that no reliable data have been presented to inform population 1 in the decision problem and so did not go on to explore this comparison further.

3.32 The ERG noted that some evidence suggests that the peak age of onset of systemic JIA is between 18 months and 2 years. In a UK cohort, the peak age was 2 years, with a mean of 6 years. In another UK prospective study, the Childhood Arthritis Prospective Study, the median age of onset is reported as 6.4 years (interquartile range 4.2 to 9.8 years). The ERG also noted that the average age in the TENDER trial was close to 10 years. The manufacturer’s scenario analysis shows that patient age has a substantial effect on the cost effectiveness of the intervention; for example, a starting age of 5 years leads to an ICER of approximately £36,000 per QALY gained and a starting age of 10 years leads to an ICER of approximately £47,000 per QALY gained when comparing tocilizumab with anakinra. The increase in the ICER with age is a result of higher doses of tocilizumab being used as the child gets older (because the dosing of tocilizumab is based on the weight of the patient), so the cost of tocilizumab increases, whereas the dose for comparator treatments does not increase to the same extent. Therefore, when comparing tocilizumab with the other options at a later starting age, the years with a relatively low cost of tocilizumab are no longer included in the analyses, hence the increase in the ICER.

3.33 The ERG conducted some exploratory analyses using the following assumptions and modifications:

• The starting age is 7 years (based on the observed average of 6 years, plus 1 year for diagnosis and [failed] treatments with NSAIDs, corticosteroids and methotrexate), with a time horizon of 11 years.
• The cycle length is adjusted to 12 weeks instead of the current 3 months.
• The withdrawal rate is based on the exponential distribution.
• The ACR response probabilities for tocilizumab are adjusted to reflect the methotrexate non-responder population (95% of the whole population).
• The relative risk for anakinra is adjusted to reflect the non-methotrexate-naive population in the indirect comparison.
• Parameters for the distribution of treatment response for anakinra and other TNF-alpha inhibitors for the probabilistic sensitivity analysis are adjusted to include additional uncertainty around the relative risks and around the adjustment factor.

3.34 The ERG’s resulting base-case ICER for tocilizumab compared with anakinra was £42,500 per QALY gained. The ERG did an exploratory analysis based on their base case. The ERG varied the withdrawal probabilities so that high responders would have a lower probability of withdrawing than low responders. This was implemented by assuming withdrawal of 5% for ACR30 response, 3.5% for ACR50response, 2.7% for ACR70 response and 1.5% for ACR90 response. There is no evidence base for the specific values used; the main aim was to use realistic values and ensure that the base-case withdrawal risk of 3.13% would be between the ACR50 and ACR70 responses. The resulting ICER was £40,916 per QALY gained, slightly lower than the base-case ICER.

3.35 In the second scenario, the ERG explored the effect of the manufacturer’s assumption that after the initial response patients stay in their current health state, withdraw and move to next line, or die. The ERG assumed instead that patients would move between all health states with a probability of 10% per transition; that is, patients in the ACR30 state had (per cycle) a 10% chance of moving to ACR50, a 10% chance of moving to ACR70 and a 10% chance of moving to ACR90. The resulting ICER was £53,051 per QALY gained, 24% higher than the base-case ICER. This indicates that the assumption that patients who do not move to the next treatment line stay in the same health state indefinitely is optimistic.

3.36 As an alternative to the ERG’s starting age of 7 years, which was derived from the literature, the ERG also explored the starting age of 9.7 years that is observed in the TENDER trial data across all patients; this increases the ICER to £46,611 per QALY gained.

3.37 Finally, the ERG explored the costs and effects from various alternative sequences for treatment. The decision problem states that tocilizumab should be compared with anakinra and TNF-alpha inhibitors. The main focus of the manufacturer’s submission is on anakinra as comparator, and etanercept as comparator is explored in a scenario analysis. However, only a pair-wise comparison is done, instead of the full incremental analysis of possible sequences using the three treatment options. Additionally, the ERG considers that anakinra as a second-line treatment after tocilizumab is also a viable option. The results of the ERG’s full incremental analysis show that the strategy of etanercept followed by anakinra is dominated by (that is, more costly and less effective than) anakinra followed by etanercept. The strategy of tocilizumab followed by etanercept is extendedly dominated by tocilizumab followed by anakinra. Thus, the ERG stated that the ICER of interest becomes that of tocilizumab followed by anakinra compared with anakinra followed by etanercept. This is £39,026 per QALY gained, slightly lower than the ERG’s base-case ICER.

3.38 Full details of all the evidence are in the manufacturer’s submission and the ERG report, which are available from www.nice.org.uk/guidance/TAXXX

4 Consideration of the evidence

4.1 The Appraisal Committee reviewed the data available on the clinical and cost effectiveness of tocilizumab, having considered evidence on the nature of systemic JIA and the value placed on the benefits of tocilizumab 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 clinical pathway of care for systemic JIA. The Committee heard that there are currently no treatments specifically licensed for systemic JIA, although it was noted that etanercept is licensed for polyarticular-course JIA, which would include some patients with systemic JIA. The Committee heard from the clinical specialists that in routine clinical practice in the UK, patients with systemic JIA would be treated first with NSAIDs and systemic corticosteroids. If disease activity persisted, or if it was severe initially, then methotrexate would be used. If the child was intolerant of methotrexate or their condition did not adequately respond to an adequate trial of methotrexate, anakinra and infliximab would be the next treatment options to be used. The Committee heard that infliximab and anakinra were the biologic therapies used most commonly in clinical practice for treating systemic JIA. It also heard that if there was an inadequate response to these biologics, other treatment options included: tocilizumab; steroid joint injections; high-dose intravenous immunoglobulin; oral ciclosporin; oral thalidomide; autologous stem cell rescue after marrow ablation; and cyclophosphamide.

4.3 The Committee considered the nature of the condition, and noted evidence submitted and presented by the patient experts and clinical specialists on the clinical symptoms associated with systemic JIA. The Committee heard that children with systemic JIA will experience severe pain and fatigue, and considerable disability. This has a substantial impact on the child’s family life, school life, and physical and emotional well-being. The condition also has an effect on the wider family, with siblings finding it distressing to see the child living with the condition, and parents and carers are often required to stay at home and care for the child if they are unable to attend school. The Committee heard that it was extremely important to get the conditon under control as quickly as possible to prevent further long-term damage to the child. The Committee heard that these children will be prescribed systemic corticosteroids over long periods of time and therefore have an increased mortality risk and experience adverse effects that can lead to chronic conditions, including infections, diabetes mellitus, cardiovascular complications and osteoporosis in later life, and risk of long-term joint damage and need for joint replacement. In addition, they may have visible side effects such as growth restriction and Cushing’s syndrome. The Committee heard from the clinical specialists and patient experts how tocilizumab had made a dramatic difference to the majority of children who had been treated with it. The Committee heard from the patient expert how tocilizumab had made such a considerable difference to a child’s symptoms that the child now had significantly less pain and better energy levels, could take part in everyday activities and sports, and could concentrate sufficiently to participate in school. The Committee also heard that the majority of children taking tocilizumab had been able to reduce or completely stop using steroids and therefore the visible side effects of the corticosteroid treatment were no longer present.

4.4 The Committee heard that there is variation in the use of tocilizumab in the UK, but that in general tocilizumab is currently used for patients whose condition does not respond to methotrexate, and following either infliximab or anakinra. The Committee heard that there was some concern about the possible long-term effects of tocilizumab in children whose treatment extends into adolescence or adulthood. But this had also been the case with previous treatments and the effects would only be known over time, following the collection of data in a registry. The Committee also heard that long-term studies would be required to confirm the potential sustained effects of tocilizumab. However, the Committee heard that the rate of response to tocilizumab allows many patients’ disease to be controlled, allowing them to stop taking the drug until they have a further relapse. For those patients still on tocilizumab after 18 months, depending on response, there is the possibility of reducing the frequency of administration from once every fortnight to once every 4 weeks. The Committee heard that the peak age of onset of systemic JIA is around 18 months to 2 years, and the mean age of systemic JIA patients who would most benefit from tocilizumab would be preschool age children between 2 and 4 years, because the younger the child at the onset of disease the more severe it is.

Clinical effectiveness

4.5 The Committee considered the evidence for the effectiveness of tocilizumab and noted that the manufacturer derived data from the TENDER trial, a 12-week randomised controlled trial that compared the efficacy of tocilizumab with placebo. The Committee heard from the clinical specialists that the population was largely generalisable to the UK, but that the mean age of approximately 10 years in the TENDER trial was older than the population they would treat with tocilizumab in routine clinical practice.

4.6 The Committee considered the evidence presented for the two populations defined in the scope and the different views of the population definitions from the manufacturer and the ERG. For the population of patients whose systemic JIA had failed to respond to NSAIDs and systemic corticosteroids, the Committee noted that only 5% of the TENDER trial population were methotrexate naive. The Committee also noted that the manufacturer had used a post-hoc analysis to compare patients receiving tocilizumab with those patients in the placebo group receiving methotrexate and that this was not methodologically acceptable. The Committee therefore concluded that there was no evidence to allow them to further consider the clinical or cost effectiveness of tocilizumab compared with methotrexate.

4.7 The Committee then considered the evidence for patients whose systemic JIA had not responded to NSAIDs, systemic corticosteroids and methotrexate. The Committee agreed that the 95% of patients in the TENDER trial who either had been given methotrexate previously or were currently being administered methotrexate could be considered to have not adequately responded to methotrexate. The Committee therefore concluded that only data for these patients should be considered in any comparison of tocilizumab with TNF-alpha inhibitors or anakinra.

4.8 The Committee considered the evidence on the clinical effectiveness of tocilizumab in the population who had experienced treatment failure with NSAIDs, systemic corticosteroids and methotrexate. It noted that there were statistically significant improvements in the primary efficacy endpoint (ACR30 response and no fever) and all secondary endpoints at 12 weeks with tocilizumab compared with placebo. The Committee noted that data on joint damage assessed by radiographic progression were not available but heard from the manufacturer that these would be available ‘soon’. The Committee also heard from the manufacturer that there were some long-term data for tocilizumab from trials being conducted in Japan. The Committee concluded from the 12-week results of the TENDER trial that tocilizumab was efficacious for the treatment of patients whose systemic JIA had not responded to NSAIDs, corticosteroids and methotrexate. The Committee agreed that the manufacturer should submit further information on radiographic evidence of progression of joint damage for patients receiving tocilizumab and any long-term follow-up data from the trials being conducted in Japan.

4.9 The Committee next considered the manufacturer’s indirect comparison of tocilizumab with TNF-alpha inhibitors and anakinra. The Committee noted that the manufacturer had used data from the TENDER trial and the NCT00036374 trial to perform an indirect comparison between tocilizumab and infliximab, and had used infliximab to represent the class effect of the TNF-alpha inhibitors. The Committee was aware that the NCT00036374 trial was not specifically for patients with systemic JIA and had included patients with other subtypes of JIA. However, the Committee noted that patients on tocilizumab were significantly more likely to reach an ACR30 response than patients on infliximab. The Committee also noted the manufacturer had presented evidence from the ANAJIS trial that compared anakinra with placebo. The primary outcome measured in the ANAJIS trial was a modified ACR30 response without fever measured after 4 weeks. The Committee noted that the TENDER and ANAJIS trials had been used for an indirect comparison analysis of tocilizumab with anakinra. The Committee also noted that the manufacturer had used the whole population from the TENDER trial to represent tocilizumab, whereas the ERG used only the 95% of patients whose condition had not responded to methotrexate. The Committee noted that patients on tocilizumab were significantly more likely to reach an ACR30 response than patients on anakinra. The Committee also noted that there was no significant difference between tocilizumab and anakinra in terms of ACR30 response plus absence of fever. The Committee concluded that tocilizumab was clinically effective compared with anakinra and infliximab.

Cost effectiveness

4.10 The Committee considered the economic model submitted by the manufacturer for the cost-effectiveness analysis. The Committee accepted the ERG’s comments that the manufacturer’s economic model did not adhere to the conventions of a Markov model. The Committee noted that the model structure meant that the health states were not homogeneous, and did not allow transition from one ACR response to another in the same line of treatment. The Committee considered that after each treatment cycle, patients could move from one health state to another, which was represented by a change in CHAQ score. An improvement in the health state did not necessarily translate to a specific improvement in ACR response. The Committee disagreed with the manufacturer’s assumption of an absolute change in CHAQ score but a relative change in ACR response. The Committee heard from the clinical specialists that clinicians used the ACR response to quantify the improvement in health outcome of patients and that there would be continuous improvement in the ACR response of patients with systemic JIA who were on tocilizumab beyond 12 weeks. The Committee concluded that the manufacturer’s economic model did not accurately represent the natural history of systemic JIA and its response to treatment.

4.11 The Committee noted that the manufacturer had identified the limitations of the CHQ (which had been used in the TENDER trial to elicit patients’ health-related quality of life) and therefore had instead used data from the CHAQ. The manufacturer had made the assumption that the CHAQ score of a child was equal to the HAQ score of an adult and the adult EQ-5D is equal to the health-related quality of life of a child. The Committee expressed concern about the methods and assumptions that had been used by the manufacturer but, given that no other health-related quality of life measure was used in the trial, the Committee noted that the resulting utility values may represent plausible surrogate values. The Committee noted that the CHAQ had not been validated for children aged under 5 years and therefore it did not know how the  CHAQ responses had been elicited from the 21 children in the trial aged under 5 years. The Committee also noted that the model did not take into account parent and carer utilities, or the reduction in adverse effects from steroid use as a result of steroid sparing.

4.12 The Committee considered the costs for tocilizumab used in economic model. The Committee noted that the costs of treatment were a composite of cost of medication and cost of administering the medication. The Committee understood that, in practice, clinicians decreased the frequency of administration of tocilizumab, and in some instances stopping treatment altogether after 18 months. The Committee heard from the clinical specialists that the costs for the health states in the model were a reasonable reflection of clinical practice in the UK. The Committee also noted that potential cost savings could result from reductions in orthopaedic surgery for future joint damage and in bone marrow transplant and stem cell procedures. These factors had not been taken into account in the model.

4.13 The Committee noted that the manufacturer’s base-case ICER for the comparison of tocilizumab with anakinra was £23,200 per QALY gained. However, given the Committee’s concerns around the model structure, it could not accept this as a reliable ICER. The Committee noted that the primary endpoint of the TENDER trial (ACR30 response and no fever) had not been the outcome used in the base-case model, but was included in the sensitivity analysis. Given that both the TENDER and ANAJIS trials included the outcome ACR30 response and no fever, the Committee concluded that this should have been considered in the base case.

4.14 The Committee considered the sequencing of the comparators in the model. The Committee noted that even though the manufacturer had used infliximab in the indirect comparison analyses, it had not been used in the sequencing. The Committee heard from the clinical specialists that infliximab is often used because of the option to adjust the dose and the potential for an improvement in the condition of patients with active systemic JIA who receive higher doses. The Committee also heard that the administration of anakinra involves daily injections, compared with an infusion of tocilizumab every 2 weeks. In addition, the Committee heard that, in many instances, although anakinra helps in the short term, its effects are reduced over time, whereas tocilizumab appears to have a more sustained activity, and so anakinra should be considered after tocilizumab in treatment sequencing. The Committee noted that the manufacturer did not present a full incremental analysis of the treatment sequences, but only an incremental analysis for the comparison of tocilizumab with anakinra followed by etanercept, adalimumab and abatacept for the population that had an inadequate response to NSAIDs, systemic corticosteroids and methotrexate. The Committee therefore concluded that they would like to see a fully incremental analysis of tocilizumab compared with anakinra and infliximab in the base case, and also scenario analyses that include infliximab as a first-line treatment and have tocilizumab and anakinra at different places in the sequencing. The Committee noted that an adjustment factor from the Prince et al. observational study on etanercept had been used to adjust for the differences in responses of other subtypes of JIA to TNF-alpha inhibitors, even though the NCT00036374 trial using infliximab had been used in the indirect comparison analysis. It therefore concluded that the uncertainty around this adjustment factor should be explored in any sensitivity analysis.

4.15 The Committee considered the starting age of 2 years in the manufacturer’s economic model. The Committee noted from the exploratory analyses conducted by the ERG that as the starting age of the child increases, so does the ICER. The Committee discussed the starting age of treatment with tocilizumab and understood that the peak age of onset of systemic JIA is about 2 years, with the mean age being around 5 years. The Committee noted that evidence from the Childhood Arthritis Prospective Study (CAPS) suggested that the median age of onset of systemic JIA is 6.4 years. The Committee concluded that the starting age in the model should be 5 years rather than 2 years, in order to reflect the mean age of children with the disease and to allow for them to have tried and experienced treatment failure with NSAIDs, corticosteroids and methotrexate.

4.16 In summary, because the Committee did not have any clinical evidence on the comparison of tocilizumab with methotrexate, it concluded that tocilizumab is not recommended for the treatment of systemic JIA in children and young people aged 2 years and older whose condition has responded inadequately to NSAIDs and systemic corticosteroids. Because the Committee did not have appropriate cost effectiveness data it was minded not to recommend tocilizumab for the treatment of systemic JIA in children and young people aged 2 years and older whose condition has responded inadequately to NSAIDs, systemic corticosteroids and methotrexate. The Committeerecommended that the manufacturer be asked to submit a revised cost-effectiveness analysis for patients whose systemic JIA has not responded to methotrexate. This should include a revised economic model that allows transitions between ACR response categories within a treatment line and either restructures the health states defined by CHAQ scores and uses treatment response to define transition probabilities or alternatively the model should be a patient-level simulation to define the cost and utility values of a health state depending on both the starting CHAQ score and the change in CHAQ score in relation to the ACR response. The manufacturer’s revised base case should include the assumption that treatment starts at age 5 years. For the comparison of tocilizumab and anakinra a revised base case using the primary outcome (ACR30 response and no fever). The Committee requested a fully incremental cost-effectiveness analysis of tocilizumab compared with anakinra and infliximab. Sensitivity analysis should be carried out on the revised base cases, including:  the uncertainty around the adjustment factor derived from the etanercept study used to take account of the other types of JIA subgroups in the infliximab study; a ‘stopping rule’ for tocilizumab after 2 years of treatment; and a decreased frequency of administration of tocilizumab after 6 months to a 4-weekly regimen. Scenario analyses should be carried out that include a fully incremental cost-effectiveness analysis of tocilizumab compared with anakinra and infliximab with: infliximab as first-line treatment followed by tocilizumab; and infliximab as first-line treatment followed by anakinra. The Committee requested further information on: radiographic evidence of progression of joint damage for patients receiving tocilizumab; long-term follow-up data from the trials being conducted in Japan; and how CHAQ responses were elicited from children aged under 5 years.

4.17 The Committee considered whether there were any equalities considerations affecting population groups protected by equality legislation and concluded that there were no equality issues relating to this appraisal that needed addressing in the guidance.

Summary of Appraisal Committee’s key conclusions

TAXXX		Appraisal title:			Section
	Key conclusion
Tocilizumab is not recommended for the treatment of systemic juvenile idiopathic arthritis in children and young people aged 2 to 17 years and older whose disease has responded inadequately to non-steroidal anti-inflammatory drugs (NSAIDs) and systemic corticosteroids. The Committee is minded not to recommend tocilizumab for the treatment of systemic juvenile idiopathic arthritis in children and young people aged 2 years and older whose disease has responded inadequately to NSAIDs, systemic corticosteroids and methotrexate. The following further information on clinical and cost effectiveness should be provided by the manufacturer: ·       A revised economic model that allows transitions between ACR (American College of Rheumatology)-response categories within a treatment line and: -      restructures the health states defined by Childhood Health Assessment Questionnaire (CHAQ) scores and uses treatment response to define transition probabilities or -      uses a patient-level simulation to define the cost and utility values of a health state depending on both the starting CHAQ score and the change in CHAQ score in relation to the ACR (American College of Rheumatology) response. ·       A revised manufacturer’s base case with the assumption that treatment starts at age 5 years. ·       For the comparison of tocilizumab and anakinra a revised base case using the primary outcome (ACR30 response and no fever). ·       A fully incremental cost-effectiveness analysis of tocilizumab compared with anakinra and infliximab. ·       Sensitivity analysis on the revised base cases that includes: -      the uncertainty around the adjustment factor derived from the etanercept study used to take account of the other juvenile idiopathic arthritis subgroups in the infliximab study -      a ‘stopping rule’ for tocilizumab after 2 years of treatment -      a decreased frequency of administration of tocilizumab after 6 months to a 4-weekly regimen. ·       Scenario analyses that include a fully incremental cost-effectiveness analysis of tocilizumab compared with anakinra and infliximab with: -      infliximab as first-line treatment followed by tocilizumab -      infliximab as first-line treatment followed by anakinra. The Committee requests further information on: ·       radiographic evidence of progression of joint damage for patients receiving tocilizumab ·       long-term follow-up data from the trials being conducted in Japan ·       how CHAQ responses were elicited from children aged under 5 years.					1.1  1.2 1.3
	Current practice
Clinical need of patients, including the availability of alternative treatments			The Committee heard from the clinical specialists and patient experts that children with systemic JIA experience severe pain and fatigue, and considerable disability. The Committee heard that these children will be prescribed systemic corticosteroids over long periods of time and therefore have an increased mortality risk and experience adverse effects that can lead to chronic conditions, including infections, diabetes mellitus, cardiovascular complications and osteoporosis in later life, and risk of long-term joint damage and need for joint replacement. In addition, they may have visible side effects such as growth restriction and Cushing’s syndrome.		4.3
	The technology
Proposed benefits of the technology How innovative is the technology in its potential to make a significant and substantial impact on health-related benefits?			The Committee heard from the clinical specialists and patient experts how tocilizumab had made a dramatic difference to the majority of children who had been treated with it. The Committee heard from the patient expert how tocilizumab had made such a considerable difference to a child’s symptoms that the child now had significantly less pain and better energy levels, could take part in everyday activities and sports, and could concentrate sufficiently to participate in school. The Committee also heard that the majority of children taking tocilizumab had been able to reduce or completely stop using steroids and therefore the visible side effects of the corticosteroid treatment were no longer present. No specific claim of innovation was made.		4.3
What is the position of the treatment in the pathway of care for the condition?			The Committee discussed the clinical pathway of care for systemic JIA. The Committee heard that there are currently no treatments specifically licensed for systemic JIA, although it was noted that etanercept is licensed for polyarticular-course JIA, which would include some patients with systemic JIA. The Committee heard from the clinical specialists that in routine clinical practice in the UK, patients with systemic JIA would be treated first with NSAIDs and systemic corticosteroids. If the disease persisted, or if it was severe initially, then methotrexate would be used. The Committee heard that there is variation in the use of tocilizumab in the UK, but that in general tocilizumab is currently used for patients whose condition does not respond to methotrexate, and following either infliximab or anakinra.		4.2 4.4
Adverse effects			Upper respiratory tract infection, with typical symptoms such as cough, blocked nose, runny nose, sore throat and headache, is one of the most common side effects of tocilizumab. Other reported side effects include rash, urticaria, diarrhoea, epigastric discomfort and arthralgia. Infusion-related reactions that can be considered serious and life-threatening (such as angioedema) have also been reported.		2.2
	Evidence for clinical effectiveness
Availability, nature and quality of evidence			The Committee considered the evidence for the effectiveness of tocilizumab and noted that the manufacturer derived data from the TENDER trial, a 12-week randomised controlled trial that compared the efficacy of tocilizumab with placebo. For the population of patients whose systemic JIA had failed to respond to NSAIDs and systemic corticosteroids, the Committee noted that only 5% of the TENDER trial population were methotrexate naive. The Committee agreed that the 95% of patients in the TENDER trial who either had been given methotrexate previously or were currently being administered methotrexate could be considered to have not adequately responded to methotrexate.		4.5 4.6 4.7
Relevance to general clinical practice in the NHS			The Committee heard from the clinical specialists that the population in the TENDER trial was largely generalisable to the UK, but that the mean age of approximately 10 years in the TENDER trial was older than the population they would treat with tocilizumab in routine clinical practice.		4.5
Uncertainties generated by the evidence			The Committee also noted that the manufacturer had used a post-hoc analysis to compare patients receiving tocilizumab with those patients in the placebo group receiving methotrexate and that this was not methodologically acceptable. The Committee therefore concluded that there was no evidence to allow them to further consider the clinical or cost effectiveness of tocilizumab compared with methotrexate. The Committee agreed that the 95% of patients in the TENDER trial who either had been given methotrexate previously or were currently being administered methotrexate could be considered to have not adequately responded to methotrexate. The Committee therefore concluded that only data for these patients should be considered in any comparison of tocilizumab with TNF-alpha inhibitors or anakinra. The Committee noted that the manufacturer used the TENDER and ANAJIS trials for an indirect comparison analysis of tocilizumab with anakinra. The Committee noted that patients on tocilizumab were significantly more likely to reach an ACR30 response than patients on anakinra. The Committee also noted that there was no significant difference between tocilizumab and anakinra in terms of ACR30 response plus absence of fever.		4.6 4.7 4.9
Are there any clinically relevant subgroups for which there is evidence of differential effectiveness?			No clinically relevant subgroups were identified for which there was differential effectiveness.
Estimate of the size of the clinical effectiveness including strength of supporting evidence			The Committee noted that there were statistically significant improvements in the primary efficacy endpoint (ACR30 response and no fever) and all secondary endpoints at 12 weeks with tocilizumab compared with placebo. The Committee concluded from the 12-week results of the TENDER trial that tocilizumab was efficacious for the treatment of patients whose systemic JIA had not responded to NSAIDs, corticosteroids and methotrexate. The Committee noted that the manufacturer had used data from the TENDER trial and the NCT00036374 trial to perform an indirect comparison between tocilizumab and infliximab, and had used infliximab to represent the class effect of the TNF-alpha inhibitors. The Committee noted that patients on tocilizumab were significantly more likely to reach an ACR30 response than patients on anakinra. The Committee also noted that there was no significant difference between tocilizumab and anakinra in terms of ACR30 response plus absence of fever. The Committee concluded that tocilizumab was clinically effective compared with anakinra and infliximab.		4.8 4.9
	Evidence for cost effectiveness
Availability and nature of evidence			The manufacturer submitted a Markov model to evaluate the cost effectiveness of tocilizumab as part of a sequence of treatments. In the tocilizumab versus methotrexate model, patients progressed to anakinra, etanercept and then adalimumab; in the tocilizumab versus anakinra model, patients progressed to etanercept, adalimumab and then abatacept.		3.15
Uncertainties around and plausibility of assumptions and inputs in the economic model			The Committee noted that the model structure meant that the health states were not homogeneous, and did not allow transition from one ACR response to another in the same line of treatment. The Committee considered that after each treatment cycle, patients could move from one health state to another, which was represented by a change in CHAQ score. An improvement in the health state did not necessarily translate to a specific improvement in ACR response. The Committee disagreed with the manufacturer’s assumption of an absolute change in CHAQ score but a relative change in ACR response. The Committee concluded that the manufacturer’s economic model did not accurately represent the natural history of systemic JIA and its response to treatment. The manufacturer had made the assumption that the CHAQ score of a child was equal to the HAQ score of an adult and the adult EQ-5D is equal to the health-related quality of life of a child. The Committee expressed concern about the methods and assumptions that had been used by the manufacturer but, given that no other health-related quality of life measure was used in the trial, the Committee noted that the resulting utility values may represent plausible surrogate values. The Committee concluded that the starting age in the model should be 5 years rather than 2 years, in order to reflect the mean age of children with the disease and to allow for them to have tried and experienced treatment failure with NSAIDs, corticosteroids and methotrexate.		4.10 4.11 4.15
Incorporation of health-related quality-of-life benefits and utility values Have any potential significant and substantial health-related benefits been identified that were not included in the economic model, and how have they been considered?			The Committee noted that the manufacturer had identified the limitations of the CHQ (which had been used in the TENDER trial to elicit patients’ health-related quality of life) and therefore had instead used data from the CHAQ. The manufacturer had made the assumption that the CHAQ score of a child was equal to the HAQ score of an adult and the adult EQ-5D is equal to the health-related quality of life of a child. The Committee expressed concern about the methods and assumptions that had been used by the manufacturer but, given that no other health-related quality of life measure was used in the trial, the Committee noted that the resulting utility values may represent plausible surrogate values. The Committee noted that the CHAQ had not been validated for children aged under 5 years and therefore it did not know how CHAQ responses were elicited from  the 21 children aged under 5 years.		4.11
Are there specific groups of people for whom the technology is particularly cost effective?			Due to the uncertainty in the economic model, there are no specific groups for whom the technology is particularly cost effective.
What are the key drivers of cost effectiveness?			The Committee considered the starting age of 2 years in the manufacturer’s economic model. The Committee noted from the exploratory analyses conducted by the ERG that as the starting age of the child increases, so does the ICER. The Committee discussed the starting age of treatment with tocilizumab and understood that the peak age of onset of systemic JIA is about 2 years, with the mean age being around 5 years.		4.15
Most likely cost-effectiveness estimate (given as an ICER)			The Committee noted that the manufacturer’s base-case ICER for the comparison of tocilizumab with anakinra was £23,200 per QALY gained. However, given the Committee’s concerns around the model structure, it could not accept this as a reliable ICER.		4.13
	Additional factors taken into account
Patient access schemes (PPRS)			No patient access scheme was submitted for the technology being appraised.
End-of-life considerations			The supplementary advice was not relevant to this appraisal.
Equalities considerations and social value judgements			No equalities issues were raised in the appraisal.

5 Implementation

5.1 The Secretary of State and the Welsh Assembly Minister for Health and Social Services have issued directions to the NHS in England and Wales on implementing NICE technology appraisal guidance. When a NICE technology appraisal recommends use of a drug or treatment, or other technology, the NHS must usually provide funding and resources for it within 3 months of the guidance being published. If the Department of Health issues a variation to the 3-month funding direction, details will be available on the NICE website. When there is no NICE technology appraisal guidance on a drug, treatment or other technology, decisions on funding should be made locally.

5.2 NICE has developed tools to help organisations put this guidance into practice (listed below). These are available on our website (www.nice.org.uk/guidance/TAXXX). [NICE to amend list as needed at time of publication]

• Slides highlighting key messages for local discussion.
• Costing template and report to estimate the national and local savings and costs associated with implementation.
• Implementation advice on how to put the guidance into practice and national initiatives that support this locally.
• A costing statement explaining the resource impact of this guidance.
• Audit support for monitoring local practice.

6 Related NICE guidance

Published

• Guidance on the use of etanercept for the treatment of juvenile idiopathic arthritis.NICE technology appraisal guidance 35 (2002). Available from www.nice.org.uk/guidance/TA35

7 Proposed date for review of guidance

7.1 NICE proposes that the guidance on this technology is considered for review by the Guidance Executive in December 2014. NICE welcomes comment on this proposed date. The Guidance Executive will decide whether the technology should be reviewed based on information gathered by NICE, and in consultation with consultees and commentators.

Gary McVeigh
Vice-Chair, Appraisal Committee
August 2011

Appendix A: Appraisal Committee members, and NICE project team

A. Appraisal Committee members

The Appraisal Committees are standing advisory committees of NICE. Members are appointed for a 3-year term. A list of the Committee members who took part in the discussions for this appraisal appears below. There are four Appraisal Committees, each with a chair and vice chair. Each Appraisal Committee meets once a month, except in December when there are no meetings. Each Committee considers its own list of technologies, and ongoing topics are not moved between Committees.

Committee members are asked to declare any interests in the technology to be appraised. If it is considered there is a conflict of interest, the member is excluded from participating further in that appraisal.

The minutes of each Appraisal Committee meeting, which include the names of the members who attended and their declarations of interests, are posted on the NICE website.

Dr Daniele Bryden
Consultant in Intensive Care Medicine and Anaesthesia, Sheffield Teaching Hospitals NHS Trust

Dr Andrew Burnett
Director for Health Improvement and Medical Director, NHS Barnet, London

Dr Mary Cooke
Lecturer, School of Nursing, Midwifery and Social Work, University of Manchester

Richard Devereaux-Phillips 
Director, Public Policy and Advocacy North West Europe, Becton, Dickinson (BD) and Company, Oxford

Professor Rachel A Elliott
Lord Trent Professor of Medicines and Health, University of Nottingham

Dr Wasim Hanif
Consultant Physician and Honorary Senior Lecturer, University Hospital Birmingham

Dr Alan Haycox
Reader in Health Economics, University of Liverpool Management School

Dr Peter Jackson
Clinical Pharmacologist, University of Sheffield

Professor Gary McVeigh
Professor of Cardiovascular Medicine, Queens University Belfast and Consultant Physician, Belfast City Hospital

Professor Eugene Milne
Deputy Regional Director of Public Health, North East Strategic Health Authority, Newcastle upon Tyne

Professor Stephen O’Brien
Professor of Haematology, Newcastle University

Professor Katherine Payne
Professor of Health Economics, University of Manchester

Dr Danielle Preedy
Lay member

Dr Peter Selby
Consultant Physician, Central Manchester University Hospitals NHS Foundation Trust

Professor Andrew Stevens
Chair of Appraisal Committee C, Professor of Public Health, University of Birmingham

Dr Matt Stevenson
Technical Director, School of Health and Related Research, University of Sheffield

Dr Judith Wardle
Lay member

B. NICE project team

Each technology appraisal is assigned to a team consisting of one or more health technology analysts (who act as technical leads for the appraisal), a technical adviser and a project manager.

Alfred Sackeyfio
Technical Lead(s)

Joanna Richardson
Technical Adviser

Lori Farrar
Project Manager

Appendix B: Sources of evidence considered by the Committee

A. The Evidence Review Group (ERG) report for this appraisal was prepared by Kleijnen Systematic Reviews Ltd:

• Riemsma R, Al MJ, Lhachimi SK et al. Tocilizumab for the treatment of systemic juvenile idiopathic arthritis:June 2011)

B. The following organisations accepted the invitation to participate in this appraisal as consultees and commentators. They were invited to comment on the draft scope, the ERG report and the appraisal consultation document (ACD). Organisations listed in I were also invited to make written submissions. Organisations listed in II and III had the opportunity to give their expert views. Organisations listed in I, II and III also have the opportunity to appeal against the final appraisal determination.

I Manufacturer/sponsor:

• Roche Products 

II Professional/specialist and patient/carer groups:

• Arthritis Care
• National Rheumatoid Arthritis Society
• South Asian Health Foundation
• British Health Professionals in Rheumatology
• British Society for Paediatric and Adolescent Rheumatology
• British Society for Rheumatology
• Royal College of Nursing
• Royal College of Paediatrics and Child Health
• Royal College of Pathologists
• Royal College of Physicians

III Other consultees:

• Department of Health
• Welsh Government

IV Commentator organisations (did not provide written evidence and without the right of appeal):

• British National Formulary
• Commissioning Support Appraisals Service
• Department of Health, Social Services and Public Safety for Northern Ireland
• Healthcare Improvement Scotland
• Abbott
• Pfizer/Wyeth
• Schering-Plough
• Swedish Orphan Biovitrum
• Arthritis Research UK
• Kleijnen Systematic Reviews
• National Coordinating Centre for Health Technology Assessment

C The following individuals were selected from clinical specialist and patient expert nominations from the non-manufacturer/sponsor consultees and commentators. They gave their expert personal view on tocilizumab by attending the initial Committee discussion and providing written evidence to the Committee. They are invited to comment on the ACD.

• Dr Jeremy Camilleri, Consultant Rheumatologist, nominated by Welsh Government – clinical specialist
• Dr Mark Wood, Consultant Paediatric Rheumatologist, nominated by British Society of Paediatric and Adolescent Rheumatology – clinical specialist
• Dr Gavin Cleary, nominated by British Society of Paediatric and Adolescent Rheumatology – clinical specialist
• Helen Copeland, nominated by National Rheumatoid Arthritis Society – patient expert
• Sarah Gebbie, nominated by National Rheumatoid Arthritis Society – patient expert

D Representatives from the following manufacturer/sponsor attended Committee meetings. They contributed only when asked by the Committee chair to clarify specific issues and comment on factual accuracy.

• Roche Products