Ankylosing spondylitis - adalimumab, etanercept and infliximab: appraisal consultation document

The Department of Health and the Welsh Assembly Government have asked the National Institute for Health and Clinical Excellence (NICE or the Institute) to conduct an appraisal of adalimumab, etanercept and infliximab and provide guidance on their use to the NHS in England and Wales. The Appraisal Committee has had its first meeting to consider both the evidence submitted and the views put forward by the representatives nominated for this appraisal by professional organisations and patient/carer and service user organisations. The Committee has developed preliminary recommendations on the use of adalimumab, etanercept and infliximab.

This document has been prepared for consultation with the formal consultees. It summarises the evidence and views that have been considered and sets out the preliminary recommendations developed by the Committee. The Institute is now inviting comments from the formal consultees in the appraisal process (the consultees for this appraisal are listed on the NICE website, www.nice.org.uk).

Note that this document does not constitute the Institute's formal guidance on these technologies. The recommendations made in section 1 are preliminary and may change after consultation.

The process the Institute will follow after the consultation period is summarised below. For further details, see the Guide to the technology appraisal process (this document is available on the Institute’s website, www.nice.org.uk).

The Appraisal Committee will meet again to consider the original evidence and this appraisal consultation document in the light of the views of the formal consultees.

At that meeting, the Committee will also consider comments made on the document by people who are not formal consultees in the appraisal process.

After considering feedback from the consultation process, the Committee will prepare the final appraisal determination (FAD) and submit it to the Institute.

Subject to any appeal by consultees, the FAD may be used as the basis for the Institute’s guidance on the use of the appraised technology in the NHS in England and Wales.

The key dates for this appraisal are:
Closing date for comments: 31 January 2007
Second Appraisal Committee meeting: 21 February 2007

Details of membership of the Appraisal Committee are given in appendix A and a list of the sources of evidence used in the preparation of this document is given in appendix B.

Note that this document does not constitute the Institute's formal guidance on this technology. The recommendations made in section 1 are preliminary and may change after consultation.

1 Appraisal Committee’s preliminary recommendations

1.1 Adalimumab and etanercept are recommended as treatment options for adults with severe active ankylosing spondylitis only if all of the following criteria are fulfilled.

  • The patient’s disease satisfies the modified New York criteria for diagnosis of ankylosing spondylitis (see appendix C).
  • There is confirmation of sustained active spinal disease as evidenced by activity of at least 4 units on the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) and at least 4 cm on the 0 to 10 cm spinal pain visual analogue scale on two occasions at least 4 weeks apart.
  • The severity of inflammatory activity is assessed to be at least 6 units on the BASDAI at the time treatment with a tumour necrosis factor alfa (TNF-α) inhibitor is started.
  • Conventional treatment with two or more non-steroidal anti-inflammatory drugs each taken sequentially at maximum tolerated or recommended dosage for 4 weeks has failed to control symptoms.

1.2 Infliximab is recommended as a treatment option for adults with severe active ankylosing spondylitis only if the patient has been shown to be intolerant of, or have contraindications to, treatment with etanercept and adalimumab, or has major difficulties with self-administered injection, and treatment with an anti-TNF-α agent is considered appropriate under the circumstances outlined in section 1.1.

1.3 Patients treated with adalimumab, etanercept or infliximab should be reassessed at 3 months after initiation of therapy. Treatment should be discontinued in patients who have not experienced a reduction of at least 6 units on the BASDAI from their baseline value.

1.4 Response to treatment should be monitored at regular intervals of 3 months. If the original response of at least 6 BASDAI units has not been maintained, assessment should be repeated after 6 weeks. If the original response has not been maintained on both occasions, treatment should be discontinued.

1.5 If the patient has had an initial response to treatment of less than 6 BASDAI units, or experiences loss of response later during treatment, prescription of an alternative TNF-α inhibitor is not recommended.

2. Clinical need and practice

2.1 Ankylosing spondylitis (AS) is an inflammatory disease of unknown cause. It is one of a group of conditions known as seronegative spondyloarthropathies. The principal feature of AS is inflammation of the sacroiliac joint at the base of the spine (sacroiliitis) followed by inflammation rising along the spine. The result is back pain and stiffness. Inflammation at entheses (the sites where ligaments and tendons attach to bone) can lead to new bone development and joint fixation (ankylosis). The large peripheral joints (hips, shoulders and knees) may also be involved, and the eyes and cardiovascular system can also be affected. Systemic involvement may be significant. Disease damage is progressive and irreversible. There is increased risk of spinal fracture later in life.

2.2 Although symptoms can occur at any stage of life, onset of AS is typically in the late teenage years and twenties. AS is nearly three times as common in men as it is in women, and men are also more likely to develop severe spinal disease.

2.3 The course of AS is highly variable and is characterised by mild or moderate flares of active disease, alternating with periods of near or total quiescence. The disease can be severe, causing spinal fusion, pronounced incapacity and significant deformities. There is a need for joint replacement surgery in some patients. About a third of people with AS may be unable to work altogether, with a further 15% reporting some changes to their working lives. In addition, AS is associated with an increased risk of death: it is estimated that patients have a standardised mortality ratio of 1.5 or greater. According to British Society for Rheumatology (BSR) guidelines, the excess mortality is mainly accounted for by cardiac valvular disease, amyloidosis and fractures.

2.4 The prevalence of AS is unknown, although it has been estimated to range from 0.05% to 0.23% (based on data from the UK and Hungary). The Assessment Group considered that data from a Finnish study provide more accurate estimates of the prevalence and incidence of AS for a UK population. In that study, the prevalence of ‘clinically significant AS’ was estimated to be 0.15%. The annual incidence was calculated to be 6.9 per 100,000. Using mid-2004 population figures for England and Wales, this suggests that there are approximately 2300 new cases each year.

2.5 There are three key elements to AS that are assessed in clinical trials: disease activity, physical function and structural damage. A number of assessment tools have been developed to measure these. For example, the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) is the most commonly used instrument to measure the inflammatory activity of AS. The BASDAI is a validated, composite index that records patients’ responses to six questions relating to the five major symptoms of AS: fatigue, axial and peripheral pain, stiffness and enthesopathy. Responses are recorded on 10-cm visual analogue scales (VAS).

2.6 Physical function is widely assessed through the use of the Bath Ankylosing Spondylitis Functional Index (BASFI). The BASFI is a patient-assessed, validated, composite index made up of 10 questions that address function and the patient’s ability to cope with AS. As with the BASDAI, responses are recorded on a 10-cm VAS.

2.7 Structural damage and disease progression are primarily evaluated using radiography. Two instruments used to assess structural damage are the Bath Ankylosing Spondylitis Radiology Index (BASRI) and the modified Stoke Ankylosing Spondylitis Spinal Score (mSASSS).

2.8 The Assessment in Ankylosing Spondylitis (ASAS) working group has developed a set of response criteria commonly used in AS clinical trials. The ASAS criteria relate to improvement across a set of four domains including physical function (as measured using the BASFI) and spinal pain (measured on a 100-mm VAS). An ASAS 20 response (a common primary efficacy outcome in clinical trials) is defined as an improvement of greater than 20% and an absolute change of 10 or more points on the 0–100 mm VAS in at least three of the four domains. In the fourth domain there must be no worsening by a similar amount. Other definitions of ASAS response (ASAS 50 and ASAS 70), based on improvements of 50% and 70%, respectively, are also used to measure outcomes in clinical studies.

2.9 Therapies aim to provide symptom relief and improve spinal mobility. Conventional therapy for AS includes agents such as non-steroidal anti-inflammatory drugs (NSAIDs), drugs that modify the disease process in rheumatoid arthritis (for example, sulfasalazine and methotrexate), and non-drug interventions (for example, physiotherapy). These treatments may be used concomitantly with TNF-α inhibitors adalimumab, etanercept and infliximab. Physiotherapy, exercise and NSAIDs are often first-line therapies. From the available evidence, the benefits of sulfasalazine and methotrexate in AS are somewhat limited. Furthermore, consultees have indicated that disease-modifying anti-rheumatic drugs (DMARDs) may help with peripheral joint involvement, but not with spinal symptoms.

2.10 There is no clear evidence that NSAIDs alter the structural progression of the disease. This, and the side-effect profile of these drugs, have led clinicians to use NSAIDs for symptomatic control rather than as continuous therapy in the majority of patients.

3. The technology

3.1 Adalimumab

3.1.1 Adalimumab (Humira, Abbott Laboratories Ltd) is a human-sequence antibody that binds specifically to TNF-α and neutralises its biological function by blocking its interaction with cell-surface TNF α receptors. TNF-α is a cytokine that mediates inflammation and modulates the cellular immune response. Adalimumab has UK marketing authorisation for the treatment of severe active AS in adults who have had an inadequate response to conventional therapy.

3.1.2 According to the summary of product characteristics (SPC), common adverse events reported during adalimumab therapy include injection-site reactions and infections. Before treatment begins all patients must be evaluated for both active and inactive (latent) tuberculosis infection. Adalimumab is contraindicated in patients with moderate to severe heart failure, active tuberculosis or other active infections. The SPC specifies a number of uncommon but serious adverse events that may be related to the immunomodulatory activity. For full details of side effects and contraindications, see the SPC.

3.1.3 Adalimumab is administered every other week via subcutaneous injection. The SPC states that a clinical response is usually achieved within 12 weeks of treatment. Continued therapy should be carefully reconsidered in a patient not responding within this time period . The net price for a 40-mg prefilled syringe is 357.50 (excluding VAT; ‘British national formulary’, 52nd edition). The annual cost of adalimumab for 26 doses at a dose of 40 mg every other week is 9295. Costs may vary in different settings because of negotiated procurement discounts.

3.2 Etanercept

3.2.1 Etanercept (Enbrel, Wyeth Pharmaceuticals) is a recombinant human TNF receptor fusion protein that inhibits the activity of TNF-α. Etanercept is licensed for the treatment of adults with severe active AS who have had an inadequate response to conventional therapy.

3.2.2 The most frequent adverse events reported during etanercept therapy include injection-site reactions, infections and allergic reactions. The SPC states that etanercept is contraindicated in people with sepsis or risk of sepsis, active infections, and hypersensitivity to the active substance or excipients. The SPC specifies a number of uncommon but serious adverse events that may be related to the immunomodulatory activity. For full details of side effects and contraindications, see the SPC.

3.2.3 Etanercept is administered by subcutaneous injection either twice weekly or once weekly depending on the dose. The net price for a 25-mg vial is 89.38 and the net price of a 50-mg vial is 178.75 (excluding VAT; ‘British national formulary’, 52nd edition). The annual cost of etanercept using either 52 once weekly doses or 104 twice weekly doses is 9295. Costs may vary in different settings because of negotiated procurement discounts.

3.3 Infliximab

3.3.1 Infliximab (Remicade, Schering-Plough Ltd) is a chimeric monoclonal antibody that binds with high affinity to TNF-α, thereby neutralising its activity. Infliximab is licensed for treatment of AS in patients who have severe axial symptoms and elevated serological markers of inflammatory activity, and whose AS has responded inadequately to conventional therapy.

3.3.2 The most common adverse events reported during infliximab therapy include acute infusion-related reactions, infections and delayed hypersensitivity reactions. The SPC states that infliximab is contraindicated in people with moderate or severe heart failure, and before treatment is initiated, people must be screened for both active and inactive tuberculosis. The SPC also specifies a number of uncommon but serious adverse events related to the immunomodulatory activity. For full details of side effects and contraindications, see the SPC.

3.3.3 Infliximab is administered by intravenous infusion over 2 hours at weeks 0, 2 and 6, and thereafter every 6–8 weeks. The SPC states that if there is no response by 6 weeks (that is, after 2 doses), no additional treatment with infliximab should be given. The net price for a 100-mg vial is 419.62 (excluding VAT; ‘British national formulary’, 52nd edition). For a 75-kg adult, each dose of 5mg/kg requires four 100-mg vials at a cost of 1678. The three loading doses cost 5034 in total, with an approximate annual cost following the loading doses of between 15,100 and 11,700 depending on whether infusions are repeated every 6 or 8 weeks. Costs may vary in different settings because of negotiated procurement discounts.

4. Evidence and interpretation

The Appraisal Committee considered evidence from a number of sources (see appendix B).

4.1 Clinical effectiveness

The Assessment Group found nine randomised controlled trials (RCTs) that met the inclusion criteria for their review. All these studies were placebo controlled and all participants continued with some form of standard treatment such as NSAIDs with or without DMARDs.

4.1.1 Adalimumab

4.1.1.1 There were two RCTs comparing adalimumab with placebo. In the larger of the two studies (ATLAS) 315 patients were randomised to adalimumab or placebo. In the smaller study (Canadian) 82 patients were randomised to adalimumab or placebo. The entry criteria for both studies specified that participants must have active AS which had responded inadequately to one or more NSAIDs. People in whom one or more DMARDs had failed were also included.

4.1.1.2 Both studies were double blind in design and in both adalimumab was given at a dose of 40 mg by subcutaneous injection every other week. Treatment was for 24 weeks, after which both active and placebo groups were switched to open-label treatment with adalimumab. The primary endpoint was the proportion of ASAS 20 responders at week 12. Those who had not reached an ASAS 20 response on assessment at weeks 12, 16 or 20 were offered the option of ‘early-escape’ open-label treatment with adalimumab. Participants who chose this option were counted only as ‘non-responders’ in the analysis of primary and secondary endpoints. The manufacturers note that this has implications when comparing the results of these studies with previous studies of anti-TNF-α drugs. Earlier studies did not include this ‘early-escape’ option and therefore would include those who responded later than 12 weeks in their analyses of outcomes. For this reason the Assessment Group did not include 24-week data for adalimumab in their meta-analysis.

4.1.1.3 In the ATLAS study the primary endpoint (ASAS 20 at week 12) was reached by 58.2% of the adalimumab group and 20.6% of the placebo group (p < 0.001). In the Canadian AS study, the respective corresponding ASAS 20 response rates were 47.2% and 21.5%. The difference did not reach conventional levels of statistical significance in the smaller study (p = 0.06). Meta-analysis results showed that the pooled relative risk of an ASAS 20 response was 2.43 (95% confidence interval [CI], 1.76 to 3.35). The corresponding relative risks for ASAS 50 and ASAS 70 responses were 3.22 (95% CI, 1.98 to 5.23) and 5.47 (95% CI, 2.43 to 12.31), respectively.

4.1.1.4 The mean reduction in BASDAI score at 12 weeks in the ATLAS study was 2.55 (based on 0–10 cm VAS) in the adalimumab group compared with 0.86 in the placebo group (p < 0.001).

4.1.1.5 Adjusted BASFI score at 12 weeks in the ATLAS study was reduced by a mean of 1.72 (scale of 0–10) in the adalimumab group compared with 0.55 (8.0% reduction from baseline) in the placebo group (p < 0.001).

4.1.1.6 The manufacturer’s submission reports that potential biomarkers of synovitis (matrix metalloproteinase-3) and type II collagen degradation (urinary type II collagen C-telopeptide) were significantly suppressed by adalimumab in the Canadian AS study. This may suggest inhibition of structural damage. However, further research is required to demonstrate the relationship of these markers with structural damage.

4.1.2 Etanercept

4.1.2.1 There were five studies of etanercept versus placebo included in the Assessment Group review. The studies compared etanercept at a dose of 25 mg twice weekly with placebo. One of the studies also had a third arm in which etanercept was given at a dose of 50 mg once a week. All the studies were double blind, and ASAS 20 response at 12 weeks was the stated primary endpoint with the exception of one study where the primary endpoint was 50% improvement in disease activity using BASDAI scores. The duration of the studies varied from 6 weeks to 24 weeks.

4.1.2.2 Meta-analysis results showed that the pooled relative risk of an ASAS 20 response was 2.13 at 12 weeks (95% CI, 1.73 to 2.63) and 2.53 at 24 weeks (95% CI, 1.80 to 3.57). The corresponding relative risks for ASAS 50 and ASAS 70 responses at 12 weeks were 3.53 (95% CI, 2.50 to 4.98) and 3.38 (95% CI, 2.10 to 5.45), respectively, and at 24 weeks the relative risks for ASAS 50 and ASAS 70 responses were 3.96 (95% CI, 2.37 to 6.63) and 4.59 (95% CI, 2.32 to 9.07), respectively.

4.1.2.3 Four of the five etanercept studies reported BASDAI scores. According to meta-analysis based on the three studies that reported BASDAI score at 12 weeks, the additional reduction in BASDAI score (on a scale of 0–10) achieved with etanercept versus placebo (weighted mean difference) was 1.67 (95% CI, 1.24 to 2.10). At 24 weeks the additional reduction was 2.0 (95% CI, 1.39 to 2.61), based on the study by Davis and colleagues only.

4.1.2.4 The Assessment Group meta-analysis found an additional lowering in BASFI score of 1.48 (95% CI, 1.13 to 1.83) with etanercept compared with placebo at 12 weeks and 1.42 (95% CI, 0.95 to 1.89) at 24 weeks (on a scale of 0–10).

4.1.2.5 Radiographic outcome data are only available from observational follow-up. Patients in one study (n = 84) were eligible to enter a 2-year open-label follow-up study to provide longer-term data including radiographic outcomes. X-rays taken at baseline in the double-blind study were compared with films at week 48 of the extension or at early termination. There appeared to be no change in mSASSS over this period of time. In another extension to a study, 18 etanercept-treated patients had X-rays at baseline and after 1 year.

4.1.3 Infliximab

4.1.3.1 The Assessment Group included two randomised, placebo-controlled studies of infliximab in their review. In the larger of the two studies, (ASSERT), participants were randomised to placebo or infliximab infusion at a dose of 5 mg/kg at weeks 0, 2, 6 and every 6 weeks thereafter. The primary endpoint was the proportion of ASAS 20 responders at week 24. The smaller study included 70 patients randomised to placebo or infliximab The dose of infliximab was 5 mg/kg at weeks 0, 2 and 6, and the primary endpoint was the proportion of patients achieving a 50% reduction in BASDAI at 12 weeks. In both studies the inclusion criteria specified that patients had to have a BASDAI score of 4 or greater (scale of 0–10) and a spinal pain score of 4 or greater measured on a 10-cm VAS.

4.1.3.2 In the ASSERT study the primary endpoint (ASAS 20 at week 24) was reached by 61.2% of the infliximab group and 19.2% of the placebo group (p < 0.001). Meta-analysis showed that the pooled relative risk of an ASAS 20 response with infliximab versus placebo was 4.11 at 12 weeks (95% CI, 2.62 to 6.44). At 24 weeks the relative risk of an ASAS 20 response was 3.18 (95% CI, 1.99 to 5.08), based on ASSERT only.

4.1.3.3 The mean reduction in BASDAI score (0–10) at 12 weeks in the ASSERT study was 2.9 in the infliximab group versus 0.4 in the placebo group. The corresponding figures from the smaller study were 3.2 and 0.6. In the Assessment Group meta-analysis, the additional reduction in BASDAI score achieved with infliximab versus placebo (weighted mean difference) was 2.46 at 12 weeks (95% CI, 1.97 to 2.95).

4.1.3.4 The mean additional reduction in BASFI score (0–10) at 12 weeks in the ASSERT study was 1.7 in the infliximab group compared with the placebo group.

4.1.3.5 The manufacturer’s submission refers to one study that showed less radiographic progression in people treated with infliximab for 2 years than in people receiving conventional treatment. The change in mSASSS was not statistically significant. Findings on magnetic resonance imaging from the ASSERT study may indicate potential for an effect on structural progression. A 73% reduction in spinal inflammation score (using a specially developed scoring system) at 24 weeks was observed in the infliximab group versus no change in the placebo group.

4.1.4 Pooled results

4.1.4.1 The Assessment Group also assessed the TNF- a inhibitors as a class versus placebo. ASAS20 data indicated a significant advantage of anti-TNF- a therapy over placebo at 12 weeks and 24 weeks (relative risk 2.52; 95% CI, 2.14 to 2.98, and relative risk 2.80; 95% CI, 2.11 to 3.71, respectively). At 12 weeks, the results of meta-analysis showed an additional mean reduction in BASDAI score of 1.89 (95% CI, 1.55 to 2.23), and an additional mean reduction in BASFI score of 1.46 (95% CI, 1.24 to 1.69). The Assessment Group also undertook a longitudinal meta-analysis based on correlation data between two time points. This produced near-identical results to the standard meta-analysis.

4.1.5 Indirect comparison

4.1.5.1 The Assessment Group attempted indirect comparisons of the agents. No statistically significant differences were found in ASAS response rates. For BASDAI, the differences both between infliximab and adalimumab and between infliximab and etanercept were statistically significant at 12 weeks. In both cases the comparison favoured infliximab, but the differences were marginal. For infliximab and etanercept, a 24-week comparison was also possible, and this found that the difference in BASDAI was no longer statistically significant. For BASFI there was a significant difference between infliximab and adalimumab at 12 weeks and between infliximab and etanercept at 24 weeks. Again, both differences favoured infliximab, but were judged to be marginal.

4.1.6 Adverse events

4.1.6.1 There were few serious adverse events reported in the clinical studies. For the drugs given subcutaneously (that is, adalimumab and etanercept), injection-site reactions were the most commonly reported adverse events.

4.2 Cost effectiveness

4.2.1 Published economic evaluations

4.2.1.1 In their literature search, the Assessment Group identified six full economic evaluations that met the inclusion criteria. Only one study compared anti-TNF - a drugs with each other (etanercept with infliximab). None of the economic evaluations considered adalimumab. Only one of the studies was UK based and that study compared infliximab with placebo. Three of the studies extended beyond 1 year using observational data. The cost-effectiveness results were difficult to compare because of the varying approaches adopted and limited information available from the abstracts.

4.2.1.2 The Assessment Group also reviewed studies that examined the cost impact of AS. The Assessment Group found that the majority of the total costs of AS were indirect (that is, costs due to lost work productivity).

4.2.2 Submitted economic evaluations

4.2.2.1 All three manufacturers provided economic evaluations. All evaluations adopted an NHS perspective, although Abbott and Schering-Plough also considered costs from a societal perspective. Costs and benefits in all cases were discounted at 3.5%. None of the models attempted to make any direct or indirect comparisons between the
TNF- a inhibitors.

4.2.3 Adalimumab – manufacturer's model

4.2.3.1 The Abbott economic evaluation – structured as a patient-based transition-state model – compared the use of adalimumab plus NSAIDs versus treatment with NSAIDs alone. This model incorporated patient-level data from the Canadian AS and ATLAS RCTs, and aimed to simulate treatment decisions based on the BSR guidelines. The trial populations included patients who would not have met BSR eligibility criteria; for example, study patients included those who were intolerant to or whose AS had responded inadequately to fewer than two NSAIDs.

4.2.3.2 The model consisted of two components. The first used short-term trial data (first 48 weeks) as noted above. The second component simulated long-term outcomes for responders for up to 30 years.

4.2.3.3 In the short-term component of the model, response to treatment was defined as a reduction of BASDAI to 50% of the pre-treatment value or a fall of 2 units or more, plus a reduction of the spinal pain VAS score (over the last week) by 2 cm or more. From week 48 onwards, the BASDAI and spinal pain VAS scores of each patient (including those on standard therapy) were assumed to remain stable. In contrast, BASFI scores of patients on standard therapy were assumed to increase by 0.05 units per year. For adalimumab-treated patients, BASDAI and BASFI scores remained stable as long as the patient remained on adalimumab therapy. A separate scenario analysis was conducted to assess the impact of assuming no effect of adalimumab on BASFI progression.

4.2.3.4 After 48 weeks it was assumed that patients would discontinue adalimumab treatment at a rate of 10% per year. When patients on adalimumab discontinued therapy, it was assumed that their BASDAI and BASFI scores would return to the average values of those patients in the model managed by conventional therapy. In the economic evaluation the rebound occurred rapidly (that is, within the next measurement period).

4.2.3.5 Disease-specific costs were based on ordinary least squares (OLS) regression of BASDAI and BASFI data from OASIS (the Outcome in Ankylosing Spondylitis International Study). Only BASDAI measurements were used to predict costs in the base-case analysis. In a secondary analysis, BASFI scores only were used to estimate costs. Costs of managing adverse events were included in the model.

4.2.3.6 In the base-case analysis, BASDAI and BASFI scores were used jointly to estimate health-related quality of life. A mapping exercise was undertaken, which used health utility index scores collected in the Canadian AS and ATLAS studies.

4.2.3.7 In the base case, the incremental cost-effectiveness ratio (ICER) over a 30-year time horizon was about 23,000. Univariate sensitivity analyses on a number of parameters including annual discontinuation rates were undertaken; ICERs varied from 18,000 to around 27,000 (over 30 years).

4.2.4 Etanercept – manufacturer's model

4.2.4.1 The Wyeth model compared the use of etanercept plus NSAIDs with NSAIDs alone. The model generated a hypothetical patient population based on patient-level data from two RCTs and an open-label extension. The principal RCT evidence used in the model was drawn from a single study (n = 356). The time horizon was up to 25 years.

4.2.4.2 Response to treatment was determined by BSR criteria, and was evaluated at 12 and 24 weeks within the model. Response rates in the placebo arms of the trials were used to model outcomes for the comparator in the economic evaluation. Responders obtained an initial health gain in BASDAI/BASFI scores. Patients were assumed to remain at the new BASDAI/BASFI levels during the period they responded to treatment.

4.2.4.3 It was assumed in the base case that responders to etanercept did not progress in terms of BASDAI and BASFI scores while on treatment, and that the rate of disease progression in non-responders was 0.3 units per year for both BASDAI and BASFI. In terms of the rate of withdrawal from etanercept treatment, it was assumed that initial responders to treatment would discontinue etanercept at a rate of 10% per year. In the model, patients who withdrew from etanercept would continue to receive NSAIDs, and it was therefore assumed that disease progression would mirror that of the comparator arm.

4.2.4.4 Assumptions were made concerning the impact of withdrawal from treatment on BASDAI and BASFI scores. In the base case, BASDAI and BASFI scores returned to baseline levels (base case). In an alternative scenario, the disability of the person was assumed to have progressed while on treatment. On withdrawal from etanercept, scores rebounded by the same magnitude as the initial improvement to a state worse than baseline.

4.2.4.5 Changes in BASDAI and BASFI scores drive changes in both predicted disease costs and utility. Regression analyses of EQ-5D (n = 356) and SF-36 (n = 511) data collected from AS patients were used to establish a relationship between changes in BASDAI and BASFI and utility. In the base-case analysis, the output from the EQ-5D regression was used to estimate quality-adjusted life years (QALYs).

4.2.4.6 Disease-related costs were based on a retrospective analysis of resource use by 147 AS patients attending the Staffordshire Rheumatology Centre. Adverse events were not considered in the modelling because the trials did not report statistically significant differences in adverse events between the two arms.

4.2.4.7 In the base case, the ICER was reported to be around 13,200 over a 25-year time horizon. A number of univariate sensitivity analyses were undertaken. When a utility model based on SF-36 data was used, ICERs were found to vary between 17,000 and 70,000 per QALY. Probabilistic sensitivity analysis indicated that over a 25-year time period, etanercept has an 88% probability of being cost effective at a threshold willingness to pay of 15,000.

4.2.5 Infliximab – manufacturer's model

4.2.5.1 The Schering-Plough model is based on a combined decision tree and Markov chain structure, and compares infliximab versus ‘standard therapy’. Two analyses were described, one based on the 24-week outcomes of the ASSERT trial and the other on a smaller study of up to 12 weeks. The placebo groups in these studies were assumed to have received standard therapy as these studies allowed the concomitant use of NSAIDs.

4.2.5.2 In the base case, infliximab dosing was modelled as per the administration schedule in the clinical trials; that is, 5 mg/kg every 6 weeks, with an additional infusion after 2 weeks. The SPC allows dosing to take place every 6–8 weeks. The drug dose was calculated based on the mean body weights reported in the two included studies.

4.2.5.3 Patient-level data from the two RCTs informed the decision-tree component of the model. At 12 or 24 weeks (depending on the randomised period of the trial), continuation on infliximab treatment was determined by BSR criteria relating to BASDAI response at 12 weeks. (In the case of the ASSERT data, 24-week data were used.) All patients then entered the Markov model, which has three health states (on treatment, off treatment and dead) and an annual cycle.

4.2.5.4 The age at the start of the economic evaluation was fixed at 40 years and the analysis adopted a time horizon of 70 years. In addition, it was assumed that neither AS nor its treatments would affect mortality. Costs and utilities were assigned based on regression models controlling for BASDAI, BASFI, age and gender, and bootstrapping for the probabilistic analysis.

4.2.5.5 It was assumed that infliximab non-responders and placebo patients would have a natural disease progression of 0.07 BASFI units per year. (However, in the submitted model this was not implemented for infliximab patients who withdrew from treatment.) In terms of withdrawal rate from treatment with infliximab, the analysis assumed an annual drop-out rate of 15%. Patients who withdrew from infliximab treatment were assigned the BASFI and BASDAI scores of the no-treatment group, reduced by the underlying natural progression of BASFI during the years of treatment. In other words, the rebounded BASFI score remains lower than for those patients in the comparator arm.

4.2.5.6 In the base case, the reported ICERs in the original submission were under 20,000. However, this was based on an inaccurate model which in part allowed patients who withdrew from infliximab treatment to avoid being assigned an ‘off-treatment’ disease progression. On correcting this error, the manufacturer reported base-case 70-year ICERs of approximately 27,000 to 28,000 per QALY (depending on which of the two studies is used to inform the calculation). In contrast, the Assessment Group found that on correcting the model within an Excel replica, the lifetime ICERs were between 41,000 and 50,000 per QALY. The manufacturer also reported corrected ICERs for the scenario in which disease progression while on treatment is assumed to be 50% of natural history (that is, 0.035 units per year), and the ICERs rise to between 34,000 and 35,000 per QALY.

4.2.6 The Assessment Group model

4.2.6.1 The Assessment Group examined the use of adalimumab, etanercept and infliximab compared with ‘conventional treatment’. ‘Conventional treatment’ was defined in terms of the placebo arms of two adalimumab RCTs. The group explored cost effectiveness of these interventions over the short term (1 year) and over a time horizon of up to 20 years.

4.2.6.2 The Assessment Group assumed that all three interventions were of equal effectiveness. Short-term effectiveness was modelled using response rates (based on BSR criteria) from the pooled week-12 data from the adalimumab and etanercept trials, and week-24 response rates after pro-rata imputation of the missing data from the Wyeth study. The 12-week response rate for infliximab was slightly lower than those reported for the other drugs, so adopting the pooled estimates for all three drugs did not disadvantage infliximab.

4.2.6.3 Under base-case assumptions, from week 30 onwards it was assumed that spontaneous recovery without treatment would occur at a rate of 17.1% as identified in the patient-level analysis of two adalimumab RCTs supplied in the Abbott submission. This assumption was explored in univariate and multivariate sensitivity analyses. In univariate sensitivity analyses in which it was assumed there was no spontaneous recovery in the placebo arm, the ICERs for adalimumab and etanercept decreased from 92,000 (base case) to 57,000 (over a 20-year time horizon). The ICER for infliximab decreased from 168,000 (base case) to 109,000.

4.2.6.4 The Assessment Group model assumed that patients withdraw from TNF‑ a inhibitor treatment at a rate of 15% per year, although in sensitivity analyses, rates of 6% and 24% were also explored. The annual withdrawal rate (after the first 12 months) was applied to the difference in response rate between the two arms of the evaluation, rather than the absolute number of responders.

4.2.6.5 The Assessment Group model took into account the cost of adverse events. Disease-related costs were estimated by fitting an exponential cost model to the weighted aggregate OASIS data. BASFI was used as the major predictor of costs because it was considered to reflect long-term disease progression.

4.2.6.6 Health-related quality of life was estimated by using the utility model provided by Schering-Plough on the grounds that it used a comparatively larger sample of UK AS patients, and also because it incorporated age and gender variables.

4.2.6.7 The Assessment Group adopted a long-term increase in BASFI scores of 0.07 units per year for the conventional treatment comparator arm of the model. This progression rate is applied for all periods after week 20 in the model. The model explored two alternative scenarios for BASFI progression on treatment:

  • BASFI progression continues while on TNF - a inhibitor treatment
  • there is no increase in BASFI score while on TNF - a treatment.

4.2.6.8 In the first year of treatment, base-case ICERs for adalimumab and etanercept were essentially the same (around 55,000). In contrast, the ICER for infliximab was over 124,000.

4.2.6.9 With respect to modelling beyond 12 months, the results for adalimumab were considered as representative of etanercept, and only the former were provided. In contrast with the manufacturer models, ICERs increase steadily from year 2 onwards. This is because the costs of treatment accrue at a greater rate than the accumulation of QALYs for patients continuing on treatment with a TNF - a inhibitor. This is due primarily to assumptions made by the Assessment Group regarding spontaneous recovery of patients given conventional therapy.

4.2.6.10 Univariate and multi variate sensitivity analyses were undertaken. Probabilistic sensitivity analysis was not performed because the Assessment Group considered that the results from such an analysis would be misleading. Multivariate sensitivity analyses identified scenarios in which adalimumab/etanercept could be considered cost effective, with ICERs ranging from 12,000 to 118,000. Important factors influencing the long-term cost effectiveness of these two drugs included assumptions about spontaneous recovery, withdrawal rate from treatment and the BASFI progression rate. Multivariate sensitivity analyses on the infliximab results identified no scenario in which the ICER dropped below 35,000.

4.2.7 Additional Assessment Group analysis

4.2.7.1 Using patient-level data from the adalimumab trials, the Assessment Group carried out additional analysis that explored the cost effectiveness of hypothetical treatment strategies in which only those patients who experience a defined minimal level of early response would continue to receive anti-TNF- a treatment beyond the first 12 weeks. ICERs were calculated for several subgroups, defined by the level of BASDAI response to treatment at 4, 8 and 12 weeks. In each scenario, patients’ baseline BASDAI score is expected to be at least as high as the maximum level of response within the first 12 weeks of treatment. The Assessment Group considered that measurements at 4 weeks would be inappropriate because they either have no influence on patient selection, or would lead to premature discontinuation of treatment.

4.2.7.2 Results of the extra analysis over a 12-month time horizon showed that baseline ICERs for adalimumab and etanercept were consistently below 20,000 in scenarios in which patients had experienced a response of at least 6 BASDAI points at 12 weeks with no consideration of measurements at 4 or 8 weeks. ICERs below 27,000 were observed in scenarios in which patients had a BASDAI response of at least 5 at 12 weeks, and met an additional criterion of a response of at least 2 at 8 weeks with no consideration of the measurement at 4 weeks. ICERs for infliximab were substantially higher than for adalimumab and etanercept in all scenarios over the 12-month time horizon.

4.2.7.3 Cost effectiveness over a 20-year time horizon was consistently maintained only when applying the following long-term assumptions:

  • there is no disease progression while the patient is on treatment;
  • the rate of patient withdrawal from treatment is 7% or less; and
  • there is no spontaneous remission of underlying disease.

When these conditions are fulfilled, the proportion of patients that can be expected to be on treatment in the long term varies between 6% and 18% (depending on the chosen threshold BASDAI values). ICERs for infliximab were substantially higher than for adalimumab and etanercept in all scenarios over the 20-year time horizon.

4.2.7.4 The Assessment Group stated that the cost-effectiveness results are only applicable where patients undergo a single ‘efficacy test’ period of treatment. In the event that patients start on a second or third treatment because of failure of the first, the probability of the treatment being cost effective is markedly reduced.

4.3 Consideration of the evidence

4.3.1 The Committee reviewed the data available on the clinical and cost effectiveness of adalimumab, etanercept and infliximab for AS, having considered evidence on the nature of the condition and the value placed on the benefits of these drugs by people with AS, those who represent them, and clinical specialists. It was also mindful of the need to take account of the effective use of NHS resources.

4.3.2 The Committee discussed the evidence of clinical effectiveness of TNF- a inhibitors. It noted that statistically significant improvements in measures of disease activity, functional disability and composite outcomes had been observed at 12 weeks and in some studies at 24 weeks. The Committee heard from patient experts that some AS patients who have received TNF- a inhibitors have experienced dramatic improvements in their functional ability, including the ability to work. The Committee noted that in many studies patients continued to receive NSAIDs and DMARDs alongside the study medication. The Committee understood that any effect due to the concurrent therapy with NSAIDs or DMARDs during the trials was likely to be similar in both treatment arms. The Committee concluded that there is clear evidence for the clinical effectiveness of TNF- a inhibitors in patients with severe active AS compared with placebo in the short term.

4.3.3 The Committee noted that TNF- a inhibitors have been associated with adverse events, some of which are serious, including infections, recrudescence of latent tuberculosis, hypersensitivity reactions, lymphoma and worsening of heart failure. The Committee heard from clinical specialists that physicians and patients carefully weigh the potential risks and benefits when considering whether to start anti-TNF- a therapy and whether to continue treatment in the longer term.

4.3.4 The Committee considered the differences in clinical effectiveness among adalimumab, etanercept and infliximab from the results reported in the principal RCTs. It noted that no head-to-head trials had been identified, and discussed the indirect comparisons undertaken by the Assessment Group. The Committee noted that although there was indirect evidence suggesting superiority of infliximab over both adalimumab and etanercept treatment in terms of BASDAI and BASFI endpoints at 12 weeks, this was of marginal statistical significance and the advantage was no longer apparent at 24 weeks.

4.3.5 The Committee further considered the potential for the TNF- a inhibitors to differ in their efficacy in treating features of AS other than spinal disease such as peripheral arthritis, and related conditions such as inflammatory bowel disease and uveitis. The Committee was aware that potential differences in these additional features would be important to patients with AS but noted that differences between TNF- a inhibitors have not been demonstrated in the RCTs reviewed. In addition the Committee was aware that the TNF- a inhibitors could affect long-term outcomes related to structural damage within the spine and the effects of ankylosis, but that evidence for this effect is currently lacking. The Committee concluded that there was no compelling evidence on which it could reliably distinguish between TNF- a inhibitors on the basis of clinical effectiveness in either the short or long term when making recommendations.

4.3.6 The Committee considered the evidence for the cost effectiveness of TNF- a inhibitors. The Committee discussed the varying structures, assumptions and input data in each of the manufacturers’ economic models and the Assessment Group model. It noted that over short time horizons (up to 12 months) none of the models indicated acceptable cost effectiveness of the TNF- a inhibitors for patients with AS. The Committee considered that there were significant uncertainties relating to the assumptions used to estimate ICERs over longer time horizons (20–70 years). These uncertainties include:

  • the long-term effectiveness of TNF - a inhibitors in controlling disease activity
  • whether TNF - a inhibitors slow or halt the progression of structural damage in the spine and functional disability related to ankylosis
  • the proportion of patients who may experience a significant improvement in their condition without anti-TNF - a treatment
  • the rate at which the disease progresses in those who respond to treatment, those who do not respond and those on placebo treatment
  • the rate at which patients are expected to withdraw from anti-TNF - a treatment
  • the degree to which a patient’s condition might be expected to rebound if therapy is withdrawn.

4.3.7 The Committee also noted that the long-term treatment strategy assumed in the models of continuous therapy with a single agent may not reflect clinical practice. Thus it heard from clinical specialists that there is uncertainty about the balance of risks and benefits of continuing therapy in the long term, and that it may be advisable to stop anti-TNF- a therapy after a period of time and resume therapy only in the event of deterioration of the condition. The Committee concluded that over a long time horizon, all these factors introduced considerable uncertainty into the estimates of cost effectiveness in all the models.

4.3.8 The Committee considered the relative merits of the manufacturers’ economic models and the Assessment Group model. It discussed the issues raised by consultees about the Assessment Group model, such as: the assumption of equivalent clinical effectiveness of the three anti-TNF- a drugs; the use of patient-level data from only one manufacturer; and the base-case assumption that 17.1% of patients not receiving anti-TNF- a drugs may achieve and maintain a clinically important improvement in BASDAI score in the long term. It also discussed issues raised about the manufacturer’s models in the Assessment Report and the related responses provided by the manufacturers during consultation. Overall, the Committee was of the opinion that the Assessment Group model adequately reflected all features relevant for appraising the cost effectiveness of TNF- a inhibitors in AS and that the assumptions had been widely explored in sensitivity analyses. The Committee concluded that the Assessment Group model provided the best available approach for considering the impact of essential differences in assumptions due to the wide range of sensitivity analyses presented.

4.3.9 The Committee discussed the univariate and multivariate sensitivity analyses explored in the Assessment Group model. It considered that an optimistic reflection of the situation in practice was the scenario in which the following long-term assumptions were made: patients were assumed to experience no disease progression while on anti-TNF- a treatment; no patients in the placebo arm were assumed to experience improvements in their condition that met the BSR criteria for a response; and the rate of withdrawal from anti-TNF- a therapy was assumed to be only 7%. The Committee discussed other assumptions in the base case of the Assessment Group model, including the estimates of utility, and the rate of BASFI progression of 0.07 points per year on a scale of 0–10, and considered these to be reasonable. It discussed the methods used in the Assessment Group model to project costs and health outcomes over the 20-year time horizon, and considered the methods and results to be reasonable. The Committee therefore considered that a cost of 42,000 per QALY over a 20‑year time horizon was a plausible estimate of the true ICER of adalimumab or etanercept compared with placebo. The corresponding ICER for infliximab was 82,000 per QALY. The Committee concluded that the use of TNF-α inhibitors in AS was not cost effective for the treatment of all patients with AS.

4.3.10 The Committee discussed situations in which a subgroup of patients could be identified in whom the anti-TNF- a agents would be more cost effective. The Committee discussed the various factors in the models influencing the ICER, and also noted that in clinical practice patients experienced a wide range of responses to anti-TNF- a agents. Some have a dramatic improvement in their condition whereas others’ disease does not respond to treatment or improves only slightly. The Committee heard from clinical specialists that it was not possible to identify factors that would reliably predict in advance of treatment which patients were most likely to benefit from anti-TNF- a therapy or might benefit in the longer term from reduction in disease progression. The Committee heard from the Assessment Group that it had been observed in patient-level data that patients who experienced a marked response to anti-TNF- a treatment tended to do so within the first few weeks of treatment. It heard from clinical specialists that responses often occur at a later stage, but that a large proportion of responses occur within the first 12 weeks. The Committee was also conscious of the BSR guidelines that treatment should not be stopped within 12 weeks because of ineffectiveness.

4.3.11 The Committee discussed the results of the extra analysis provided by the Assessment Group, and the issues raised by consultees and commentators. It considered the merits and limitations of the BASDAI as a measure to identify those who had responded best to therapy and noted in particular that there was considerable variability in scores between patients with apparently similar disease severity. It heard from patient experts that those who experience a marked response to therapy have substantial improvements in their quality of life. It heard from clinical specialists that although all patients have the potential to benefit from anti-TNF- a drugs, there could be a clinical rationale to continue therapy in the long-term only in those who benefit most. The Committee considered measurement of the BASDAI response in the first 12 weeks to be a reasonable approach to identifying those most likely to benefit in the long term. The Committee noted from the additional modelling that when treatment is continued only in patients with a response of 6 or more BASDAI units at 12 weeks and when this improvement is maintained, the use of adalimumab and etanercept is cost effective with ICERS below 20,000 per QALY over a 20-year time horizon, regardless of the response before 12 weeks. The Committee concluded that adalimumab and etanercept should be recommended for the treatment of AS, but that treatment should only be continued beyond 12 weeks in patients who have had a response of at least 6 units on the BASDAI.

4.3.12 The Committee discussed the differences in cost effectiveness among adalimumab, etanercept and infliximab. The Committee considered specifically the assumption of equivalence between the drugs, the use of pooled adalimumab and etanercept response rates and the use of adalimumab patient-level data in the Assessment Group model. The Committee agreed with the assumption in the Assessment Group model of equivalent clinical effectiveness for the three drugs as this was consistent with the Committee’s considerations set out in paragraphs 4.3.4 and 4.3.5. The Committee therefore accepted that the differences in cost effectiveness between the three drugs was influenced mainly by the resource use and cost associated with each treatment. The Committee was therefore persuaded that infliximab is less cost effective than adalimumab and etanercept for the treatment of AS.

4.3.13 The Committee discussed eligibility criteria for starting therapy with TNF- a inhibitors. It was in agreement with eligibility criteria set out in the BSR guidelines for prescribing TNF- a inhibitors in adults with AS in terms of diagnostic criteria, confirmation of sustained active spinal disease and failure of conventional treatment to control symptoms. In addition, it considered that for the cost-effective use, the severity of inflammatory activity should be assessed to be at least 6 units on the BASDAI at the time treatment with a TNF- a inhibitor is started.

4.3.14 The Committee discussed patient preference between drugs, which may be influenced by the route and frequency of administration. It heard from patient experts that patients value having a choice of therapy. The Committee noted that ICERs for infliximab were substantially higher than for adalimumab and etanercept in all scenarios. The Committee therefore concluded that infliximab should only be recommended for the treatment of AS if the patient has been shown to be intolerant of, or have contraindications to, treatment with etanercept and adalimumab, or has major difficulties with subcutaneous injection.

4.3.15 The Committee discussed long-term treatment with TNF- a inhibitors. It discussed the importance of monitoring with regard to discontinuing treatment in patients who were not experiencing a continued response to anti-TNF- a therapy. The Committee concluded that response to treatment should be monitored at regular intervals of 3 months. If the original response of at least 6 BASDAI units has not been maintained, there should be a repeat assessment after 6 weeks. If the original response has not been maintained on both occasions, treatment should be discontinued.

4.3.16 The Committee discussed whether it was appropriate to consider treatment with a second anti-TNF- a drug for a patient who has had an initial response to treatment of less than 6 BASDAI units, or experiences loss of response later during treatment. The Committee considered that it would be inappropriate to recommend sequential treatment in the absence of evidence demonstrating that such a strategy would be clinically or cost effective. The Committee concluded that if the patient has had an initial response to treatment of less than 6 BASDAI units, or experiences loss of response later during treatment, treatment with an alternative TNF- a inhibitor is not recommended.

5. Implementation

5.1 The Healthcare Commission assesses the performance of NHS organisations in meeting core and developmental standards set by the Department of Health in ‘Standards for better health’ issued in July 2004. The Secretary of State has directed that the NHS provides funding and resources for medicines and treatments that have been recommended by NICE technology appraisals normally within 3 months from the date that NICE publishes the guidance. Core standard C5 states that healthcare organisations should ensure they conform to NICE technology appraisals.

5.2 'Healthcare Standards for Wales’ was issued by the Welsh Assembly Government in May 2005 and provides a framework both for self-assessment by healthcare organisations and for external review and investigation by Healthcare Inspectorate Wales. Standard 12a requires healthcare organisations to ensure that patients and service users are provided with effective treatment and care that conforms to NICE technology appraisal guidance. The Assembly Minister for Health and Social Services issued a Direction in October 2003 which requires Local Health Boards and NHS Trusts to make funding available to enable the implementation of NICE technology appraisal guidance, normally within 3 months.

5.3 NICE has developed tools to help organisations implement this guidance (listed below). These are available on our website (www.nice.org.uk/TA XXX). [Note: tools will be available when the final guidance is issued]

6. Proposed recommendations for further research

The Committee considered that further research into the effectiveness of TNF- a inhibitors in ankylosing spondylitis should include the following.

6.1 Studies to investigate the long-term effects of TNF- a inhibitors in patients with AS, including effects on disease activity, functional status, structural damage, quality of life and adverse effects.

6.2 Studies to establish the appropriate duration and pattern of long-term treatment with TNF- a inhibitors.

6.3 Studies to examine whether patients’ AS would respond to more than one TNF- a inhibitor given sequentially.

7. Related NICE guidance

  • Guidance on the use of etanercept and infliximab for the treatment of rheumatoid arthritis. NICE technology appraisal no. 36 (May 2002). Available from: www.nice.org.uk/TAxxx
  • Etanercept and efalizumab for the treatment of adults with psoriasis NICE technology appraisal no. 103 (July, 2006). Available from: www.nice.org.uk/TAxxx
  • Etanercept and infliximab for the treatment of psoriatic arthritis NICE technology appraisal no. 104 (July, 2006). Available from: www.nice.org.uk/TAxxx

NICE is developing the following technology appraisal guidance.

  • Adalimumab, etanercept and infliximab for the treatment of rheumatoid arthritis (publication expected April 2007).
  • Adalimumab for the treatment of psoriatic arthritis (publication TBC).

8. Proposed date for review of guidance

The review date for a technology appraisal refers to the month and year in which the Guidance Executive will consider whether the technology should be reviewed. This decision will be taken in the light of information gathered by the Institute, and in consultation with consultees and commentators.

It is proposed that the guidance on this technology is considered for review in June 2009. The Institute would particularly welcome comment on this proposed date.

David Barnett
Chair, Appraisal Committee
December 2006

Appendix A. Appraisal Committee members and NICE project team

A. Appraisal Committee members

The Appraisal Committee is a standing advisory committee of the Institute. Its 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. The Appraisal Committee meets three times a month except in December, when there are no meetings. The Committee membership is split into three branches, each with a chair and vice-chair. Each branch considers its own list of technologies and ongoing topics are not moved between the branches.

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 Jane Adam
Radiologist, St George's Hospital, London

Professor A E Ades
MRC Senior Scientist, MRC Health Services Research Collaboration, Department of Social Medicine, University of Bristol

Dr Amanda Adler
Consultant Physician, Addenbrooke’s Hospital, Cambridge

Dr Tom Aslan
General Practitioner, Stockwell, London

Professor David Barnett (Chair)
Professor of Clinical Pharmacology, University of Leicester

Mrs Elizabeth Brain
Lay Member

Dr Karl Claxton
Health Economist, University of York

Dr Richard Cookson
Senior Lecturer in Health Economics, School of Medicine Health Policy and Practice, University of East Anglia

Mrs Fiona Duncan
Clinical Nurse Specialist, Anaesthetic Department, Blackpool Victoria Hospital, Blackpool

Professor Christopher Eccleston
Director Pain Management Unit, University of Bath

Dr Paul Ewings
Statistician, Taunton & Somerset NHS Trust, Taunton

Professor John Geddes
Professor of Epidemiological Psychiatry, University of Oxford

Mr John Goulston
Director of Finance, Barts and the London NHS Trust

Ms Linda Hands
Consultant Surgeon, John Radcliffe Hospital

Dr Rowan Hillson
Consultant Physician, Diabeticare, The Hillingdon Hospital

Professor Philip Home (Vice Chair)
Professor of Diabetes Medicine, University of Newcastle upon Tyne

Dr Catherine Jackson
Clinical Senior Lecturer in Primary Care Medicine, University of Dundee

Dr Terry John
General Practitioner, The Firs, London

Professor Richard Lilford
Professor of Clinical Epidemiology, Department of Public Health and Epidemiology, University of Birmingham

Dr Simon Maxwell
Senior Lecturer in Clinical Pharmacology and Honorary Consultant Physician, Queens Medical Research Institute

Dr Simon Mitchell
Consultant Neonatal Paediatrician, St Mary’s Hospital, Manchester

Ms Judith Paget
Chief Executive, Caerphilly Local Health Board, Wales

Dr Katherine Payne
Health Economist, The North West Genetics Knowledge Park, The University of Manchester

Dr Ann Richardson
Independent Research Consultant

Dr Stephen Saltissi
General Manager, Clinical Support Services, Cardiff and Vale NHS Trust

Mr Mike Spencer
General Manager, Clinical Support Services, Cardiff and Vale NHS Trust

Dr Debbie Stephenson
Head of HTA Strategy, Eli Lilly and Company

Professor Andrew Stevens
Professor of Public Health, University of Birmingham

Dr Cathryn Thomas
General Practitioner & Associate Professor, Department of Primary Care & General Practice, University of Birmingham

Simon Thomas
Consultant Physician, General Medicine and Clinical Pharmacology, Newcastle Hospitals NHS Trust

Mr David Thomson
Lay Member

Dr Norman Vetter
Reader, Department of Epidemiology, Statistics and Public Health, College of Medicine, University of Wales, Cardiff

Professor Mary Watkins
Professor of Nursing, University of Plymouth

Dr Paul Watson
Medical Director, Essex Strategic Health Authority

B. NICE project team

Each appraisal of a technology 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.

Helen Chung and David Chandiwana
Technical Leads

Janet Robertson
Technical Adviser

Alana Miller
Project Manager


Appendix B. Sources of evidence considered by the Committee

A The assessment report for this appraisal was prepared by Liverpool Reviews and Implementation Group (LRiG).

McLeod C, Bagust A, Boland A et al. Adalimumab, etanercept and infliximab for the treatment of ankylosing spondylitis, May 2006.

B The following organisations accepted the invitation to participate in this appraisal. They were invited to make submissions and comment on the draft scope and assessment report. They are also invited to comment on the appraisal consultation document (ACD), and consultee organisations are provided with the opportunity to appeal against the final appraisal determination (FAD).

I Manufacturers/sponsors:

  • Abbott Laboratories Ltd
  • Wyeth Pharmaceuticals
  • Schering-Plough Ltd

II Professional/specialist and patient/carer groups:

  • Arthritis and Musculoskeletal Alliance
  • Arthritis Care
  • Back Care
  • British Brain and Spine Foundation
  • British Health Professionals in Rheumatology
  • British Orthopaedic Association
  • British Pain Society
  • British Society for Rheumatology
  • British Society of Rehabilitation Medicine
  • Changing Faces
  • National Ankylosing Spondylitis Society
  • Pain Concern
  • Physiotherapy Pain Association
  • Primary Care Rheumatology Society
  • Royal Association for Disability and Rehabilitation
  • Royal College of General Practitioners
  • Royal College of Nursing
  • Royal College of Ophthalmologists
  • Royal College of Physicians
  • Royal College of Physicians of Edinburgh

III Commentator organisations (without the right of appeal):

  • Arthritis Research Campaign
  • British National Formulary
  • Liverpool Reviews and Implementation Group, University of Liverpool
  • Medicines and Healthcare Products Regulatory Agency
  • National Coordinating Centre for Health Technology Assessment
  • NHS Confederation
  • NHS Purchasing and Supplies Agency
  • NHS Quality Improvement Scotland
  • Society for Back Pain Research

C The following individuals were selected from clinical specialist and patient advocate nominations from the professional/specialist and patient/carer groups. They participated in the Appraisal Committee discussions and provided evidence to inform the Appraisal Committee’s deliberations. They gave their expert personal view on ankylosing spondylitis by attending the initial Committee discussion and/or providing written evidence to the Committee. They are invited to comment on the ACD.

  • Richard Bridgeman, patient expert, nominated by the National Ankylosing Spondylitis Society
  • Dr Karl Gaffney, clinical specialist, Consultant Rheumatologist, Norfolk and Norwich University Hospital, nominated by the Royal College of Physicians
  • Dr Andrew Keat, clinical specialist, Consultant Rheumatologist, Northwick Park Hospital, nominated by the British Society for Rheumatology
  • Fergus Rogers, patient expert, nominated by the National Ankylosing Spondylitis Society
  • Professor Roger Sturrock, clinical specialist, Professor of Rheumatology, University of Glasgow, nominated by the British Society of Rheumatology
  • Jane Skerret, patient expert, nominated by the National Ankylosing Spondylitis Society
  • Terry Orsler, patient expert, nominated by the National Ankylosing Spondylitis Society

Appendix C: Modified New York criteria for diagnosis of ankylosing spondylitis

A definite diagnosis of ankylosing spondylitis requires the radiological criterion and at least one clinical criterion to be satisfied as defined below.

Radiological criterion

Sacroiliitis at least grade 2 bilaterally or grade 3 or 4 unilaterally.

Clinical criteria

  • Low back pain and stiffness for more than 3 months that improves with exercise but is not relieved by rest.
  • Limitation of motion of the lumbar spine in both the sagittal and frontal planes.
  • Limitation of chest expansion relative to normal values correlated for age and sex.

(All reasonable measures should be taken to ensure that symptoms are due predominantly to AS and that alternative causes, including spinal fracture, disc disease and fibromyalgia, are excluded.)

This page was last updated: 29 June 2010

Accessibility | Cymraeg | Freedom of information | Vision Impaired | Contact Us | Glossary | Data protection | Copyright | Disclaimer | Terms and conditions

Copyright 2014 National Institute for Health and Care Excellence. All rights reserved.

Selected, reliable information for health and social care in one place

Accessibility | Cymraeg | Freedom of information | Vision Impaired | Contact Us | Glossary | Data protection | Copyright | Disclaimer | Terms and conditions

Copyright 2014 National Institute for Health and Care Excellence. All rights reserved.