Sarilumab for moderate to severe rheumatoid arthritis

3.1 Sarilumab's marketing authorisation and the company submission covers its use at 5 points in the treatment pathway, specifically in adults with:

3.2 NICE currently recommends the use of the biological DMARDs in its technology appraisal guidance on baricitinib, and adalimumab, etanercept, infliximab, certolizumab pegol, golimumab, tocilizumab and abatacept (of which adalimumab, certolizumab pegol, etanercept, golimumab and infliximab are tumour necrosis factor (TNF)-alpha inhibitors) and tofacitinib, in combination with methotrexate, in people with severe rheumatoid arthritis that has not responded to intensive treatment with combinations of conventional DMARDs. Disease severity is assessed using the disease activity score (DAS28). A DAS28 of more than 5.1 indicates severe disease (between 3.2 and 5.1 indicates moderate disease, less than 3.2 but more than 2.6 indicates mild disease and 2.6 or less indicates disease remission). For people who meet these criteria but cannot take methotrexate, the guidance recommends that adalimumab, baricitinib, certolizumab pegol, etanercept or tocilizumab may be used as monotherapy. The guidance recommends treatment should start with the least expensive drug (taking into account administration costs, dose needed and product price per dose) and should only be continued according to EULAR response at 6 months.

3.3 For people with severe rheumatoid arthritis who have already had at least 1 TNF‑alpha inhibitor that hasn't worked, NICE technology appraisal guidance on adalimumab, etanercept, infliximab, rituximab and abatacept and golimumab recommend the biological DMARD rituximab in combination with methotrexate for treating severe active rheumatoid arthritis. But, if rituximab is contraindicated or withdrawn because of an adverse event, NICE technology appraisal guidance recommends abatacept, adalimumab, etanercept, infliximab, golimumab, tocilizumab, certolizumab pegol, baricitinib or tofacitinib in combination with methotrexate. If methotrexate is contraindicated or withdrawn because of an adverse event, NICE's guidance recommends adalimumab, etanercept, certolizumab pegol, baricitinib or tofacitinib as monotherapy. NICE technology appraisal guidance also recommends tocilizumab in combination with methotrexate when neither TNF‑alpha inhibitors nor rituximab have worked. See the NICE Pathway on rheumatoid arthritis for more details.

3.4 The patient experts explained that rheumatoid arthritis is a lifetime condition that can severely reduce quality of life. The clinical expert stated that choosing an appropriate treatment depends on disease severity and response to treatment. The clinical expert noted that the disease sometimes does not respond adequately to the first biological DMARD prescribed. When there is not an adequate response the treatment is stopped and the next available treatment in the pathway is prescribed. Both the clinical and patient experts said it would be helpful to have new treatments that can be used at various points in the treatment pathway, alongside biological DMARDs after failure of conventional DMARDs. The clinical and patient experts highlighted that sarilumab is administered by injection at home, which has major benefits for both patients and the health system. Sarilumab has a longer shelf-life when kept out of the fridge and the clinical expert emphasised that having a treatment that lasts 14 days rather than a number of hours is an important practical factor for people, especially those who travel. The committee recognised that rheumatoid arthritis significantly affects quality of life. It concluded that there is a need for new treatment options, particularly when there is an inadequate response to conventional or biological DMARDs.

3.5 The company analysed 5 distinct subgroups in which sarilumab could be used. These were people with:

3.6 The company's clinical evidence came from 5 randomised controlled trials and 1 long-term safety study. The trials included people with moderate to severe rheumatoid arthritis, as defined in section 3.2. The trials were:

3.7 In MOBILITY-A, sarilumab plus methotrexate showed a statistically significant improvement in ACR20 at 12 weeks compared with placebo plus methotrexate (sarilumab 200 mg 65%, placebo 46%; p=0.0426). In MOBILITY-B, sarilumab plus methotrexate showed a statistically significant improvement in ACR20 at 24 weeks compared with placebo plus methotrexate (sarilumab 200 mg 66%, placebo 33%; p<0.0001). In MONARCH, sarilumab plus placebo showed a statistically significant improvement in ACR20 at 24 weeks compared with adalimumab 40 mg plus placebo (sarilumab 200 mg 72%, adalimumab 40 mg 58%; p=0.0074). The committee concluded that sarilumab plus methotrexate is more clinically effective than placebo plus methotrexate, and that sarilumab alone is more clinically effective than adalimumab alone for moderate to severe rheumatoid arthritis which has responded inadequately to conventional DMARDs.

3.8 The TARGET trial showed that sarilumab plus methotrexate led to a statistically significant improvement in ACR20 compared with methotrexate at 24 weeks (sarilumab 200 mg 61%, placebo 34%; p<0.001). The company did not report comparative statistics for ASCERTAIN because the trial was powered for safety rather clinical effectiveness. The committee concluded that sarilumab is more clinically effective than conventional DMARDs for moderate to severe rheumatoid arthritis which has responded inadequately to TNF-alpha inhibitors.

3.9 In the MOBILITY trials the rate of adverse events was higher in the sarilumab 200 mg group (ranging from 65% to 78%) compared with the methotrexate group (ranging from 47% to 62%). In the MONARCH trial, adalimumab and sarilumab had similar adverse event rates (63.6% and 64.1% respectively). The committee concluded that sarilumab plus methotrexate has a slightly higher rate of adverse events compared with methotrexate.

3.10 Because the only direct evidence available on the comparative effectiveness of sarilumab and the biological DMARDs was with adalimumab, the company did a network meta-analyses. The company did separate analyses for patients whose disease responded inadequately to either conventional or biological DMARDs, using ACR20/50/70 score, HAQ-DI, European League Against Rheumatism (EULAR) responses, DAS28 remission, mTSS, serious infections and serious adverse events. In the conventional DMARD group the analyses were further split into combination therapy and monotherapy. EULAR responses were not identified for all the relevant comparators in the NICE scope because of a lack of reporting in the trials. The company transformed ACR responses into EULAR responses to inform treatment effectiveness in the economic model (section 3.13). At 24 weeks' follow-up, for patients whose disease responded inadequately to conventional DMARDs, the network meta-analysis showed:

3.11 The company was not able to identify any evidence on sarilumab monotherapy in patients whose disease responded inadequately to biological DMARDs. The company assumed that the efficacy of sarilumab monotherapy would be equal to that of sarilumab in combination with conventional DMARDs. The ERG noted that this assumption was reasonable but recognised the considerable uncertainty caused by the absence of direct data. The committee agreed that, in the absence of direct evidence, it is reasonable to assume sarilumab monotherapy has similar effectiveness to sarilumab plus conventional DMARDs in people whose disease has responded inadequately to biological DMARDs.

3.12 The ERG stated that there was some uncertainty in the methods used by the company in the network meta-analysis. Following a request by the ERG at the clarification stage, the company provided updated results using the ERG's preferred assumptions which had been used by the assessment group in NICE technology appraisal guidance on adalimumab, etanercept, infliximab, certolizumab pegol, golimumab, tocilizumab and abatacept for rheumatoid arthritis for patients whose disease responded inadequately to conventional DMARDs. The company noted that the results of the new network meta-analysis were comparable to those of its original analyses and that the conclusion that sarilumab is comparable to other biological DMARDs was unchanged. The ERG noted that the statistically significant result for sarilumab compared with other biological DMARD treatments (both as combination therapy and monotherapy) should be treated with caution as it may be a consequence of underestimating the uncertainty in treatment effects resulting from the use of a fixed-effects model. The committee reviewed both analyses and concluded that the methods used by the company in the network meta-analysis were in line with previous NICE technology appraisal guidance, and that the network meta-analysis was therefore suitable for decision-making.

3.13 The company used a patient-level Markov model for its economic evaluation. The model categorised patients based on their EULAR response (good, moderate or no response) at 6 months. Response rates were based on the company's network meta-analysis which transformed ACR 20/50/70 response to EULAR response. The company analysed cost effectiveness for each of the subgroups described in section 3.5. The ERG stated that the model was not an individual patient-based discrete event simulation as used by the assessment group in NICE technology appraisal guidance on adalimumab, etanercept, infliximab, certolizumab pegol, golimumab, tocilizumab and abatacept for rheumatoid arthritis. The committee noted the difference between the company's model structure and the structure accepted in the previous guidance but accepted that the company's model was appropriate for its decision-making.

3.14 The company stated that that EQ-5D utility data were not available for all comparators across all patient populations. Following a request by the ERG at the clarification stage, the company used Hernandez et al. (2013) to estimate EQ-5D data based on patient characteristics (HAQ score, pain on a visual analogue scale, age and sex). The ERG stated that Hernandez et al. was used in NICE technology appraisal guidance on adalimumab, etanercept, infliximab, certolizumab pegol, golimumab, tocilizumab and abatacept for rheumatoid arthritis. The committee concluded it was appropriate for decision-making.

3.15 The company's model included costs associated with drug acquisition, drug administration and monitoring, and hospitalisation. The model treatment sequences included a TNF-alpha inhibitor bundle in the base case. The TNF-alpha inhibitor bundle used the pooled efficacy of etanercept, an etanercept biosimilar, adalimumab, infliximab, an infliximab biosimilar, golimumab and certolizumab pegol with the price weighted according to the estimated market share of each. The company did not include the cost of biosimilars for rituximab or adalimumab, but stated that at the time of submission the rituximab biosimilar was not available and the adalimumab biosimilar had not received its marketing authorisation. The ERG stated that in the company's original base case the incremental cost-effectiveness ratio (ICER) for sarilumab compared with rituximab was more than £100,000 per quality-adjusted life year (QALY) gained, and so reducing the price of rituximab would not change the conclusion for this population. Sarilumab and several of the other biological DMARDs have patient access schemes. The company had incorporated the patient access scheme prices for sarilumab, certolizumab pegol and golimumab in the model, but not the confidential patient access schemes for abatacept and tocilizumab. The committee concluded that the costs used in the company's model were appropriate.

3.16 During the clarification stage the ERG requested that the company update the sequences in which treatments are given for each subgroup to match those agreed in NICE technology appraisal guidance on adalimumab, etanercept, infliximab, certolizumab pegol, golimumab, tocilizumab and abatacept for rheumatoid arthritis. The company provided updated sequences, but the ERG noted that in the group of patients whose disease has responded inadequately to biological DMARDs, a second line of biological treatment had been added in error. This error was corrected in the ERG's additional analyses and the committee accepted this.

3.17 Following a request by the ERG at the clarification stage, the company updated the model to allow patients treated for moderate disease to progress to treatment for severe disease. The ERG explained that this progression, although not used in previous NICE technology appraisal guidance on adalimumab, etanercept, infliximab, certolizumab pegol, golimumab, tocilizumab and abatacept for rheumatoid arthritis, was requested because it provided a more accurate representation of clinical practice. The company updated its analyses but the ERG identified 2 issues with the company's methods:

3.18 In the company's model patients were assessed for response to treatment at 6 months. Patients who had a moderate or good EULAR response were assumed to have an associated reduction in HAQ-DI score, which was assumed to be independent of treatment, and the treatment was stopped if the patient did not have at least moderate EULAR response at 6 months. The ERG stated that the company had used a linear approach to HAQ-DI score progression, which would have had a favourable effect for sarilumab when compared with conventional DMARDs. It explained that a non-linear method was more appropriate and had been accepted in previous NICE technology appraisal guidance on adalimumab, etanercept, infliximab, certolizumab pegol, golimumab, tocilizumab and abatacept for rheumatoid arthritis. The ERG had used this method in its additional analyses. The committee concluded that the non-linear approach to HAQ-DI progression was appropriate and therefore accepted the ERG's additional analyses.

3.19 In the ERG's additional analysis for the population with moderate active rheumatoid arthritis that has responded inadequately to conventional DMARDs, the ICER for sarilumab compared with conventional DMARDs was £63,438 per QALY gained. The ERG also calculated new ICERs using the confidential patient access scheme prices for abatacept and intravenous tocilizumab; this comparison produced similar estimates of cost effectiveness. The committee considered that sarilumab plus conventional DMARDs was not cost effective in people with moderate rheumatoid arthritis whose disease has responded inadequately to conventional DMARDs.

3.20 In the ERG's additional analysis for the population with severe rheumatoid arthritis that has responded inadequately to conventional DMARDs, tocilizumab (intravenous and subcutaneous) was dominated by sarilumab plus conventional DMARDs (that is, sarilumab was both less costly and more effective). The ICERs for the TNF-alpha inhibitor bundle plus methotrexate and abatacept plus methotrexate compared with sarilumab plus methotrexate were both over £100,000 per QALY gained. The ERG also calculated ICERs using the confidential patient access scheme prices for abatacept and intravenous tocilizumab, which produced very similar estimates of cost effectiveness. The committee therefore recommended sarilumab plus conventional DMARDs for people with severe rheumatoid arthritis whose disease has responded inadequately to conventional DMARDs, in line with the NICE recommendations on adalimumab, etanercept, infliximab, certolizumab pegol, golimumab, tocilizumab and abatacept.

3.21 In the ERG's additional analysis for the population with severe rheumatoid arthritis that has responded inadequately to biological DMARDs for whom rituximab is a treatment option, the ICER for a sequence where sarilumab plus conventional DMARDs replaced rituximab plus conventional DMARDs was over £100,000 per QALY gained. The committee concluded that sarilumab plus conventional DMARDs was not a cost-effective use of NHS resources for people with severe rheumatoid arthritis whose disease has responded inadequately to biological DMARDs if rituximab is a treatment option.

3.22 In the ERG's additional analysis for the population with severe rheumatoid arthritis that has responded inadequately to biological DMARDs for whom rituximab is contraindicated, the incremental analysis for sarilumab compared with the TNF-alpha inhibitor bundle resulted in an ICER of £34,979 per QALY gained. Subcutaneous abatacept plus methotrexate was dominated by sarilumab plus methotrexate (that is, sarilumab was both less costly and more effective) and the ICERs for tocilizumab (intravenous and subcutaneous) plus methotrexate compared with sarilumab plus methotrexate were over £100,000 per QALY gained. The ERG updated its additional analysis using the confidential patient access scheme prices, which produced similar estimates of cost effectiveness. The committee therefore recommended sarilumab plus conventional DMARDs for people with severe rheumatoid arthritis whose disease has responded inadequately to biological DMARDs and for whom rituximab is not a treatment option.

3.23 In the ERG's additional analysis for the population with severe rheumatoid arthritis population that has not responded adequately to rituximab and other biological DMARDs, the ICER for intravenous or subcutaneous tocilizumab plus methotrexate compared with sarilumab plus methotrexate was over £100,000 per QALY gained. The ERG also updated its additional analysis using the confidential patient access scheme prices, which produced similar estimates of cost effectiveness. The committee therefore recommended sarilumab plus conventional DMARDs for people with severe rheumatoid arthritis whose disease has not responded adequately to rituximab and other biological DMARDs.

3.24 In the ERG's additional analysis for the population with severe rheumatoid arthritis that has responded inadequately to conventional DMARDs for whom methotrexate is contraindicated, sarilumab alone compared with a TNF-alpha inhibitor bundle alone in the incremental analysis had an ICER of £34,422 per QALY gained. In the incremental analysis subcutaneous tocilizumab was extendedly dominated by sarilumab alone (that is, the ICER for subcutaneous tocilizumab was higher than the next more effective alternative, which was intravenous tocilizumab). The ICER for intravenous tocilizumab compared with sarilumab was over £100,000 per QALY gained. The ERG updated this additional analysis using the confidential comparator patient access scheme prices, which produced similar estimates of cost effectiveness. The committee concluded that sarilumab monotherapy is cost effective for severe active rheumatoid arthritis after conventional DMARDs if methotrexate is not suitable.

3.25 In the ERG's additional analysis for the population with severe rheumatoid arthritis that has responded inadequately to biological DMARDs for whom methotrexate is contraindicated, the ICER for sarilumab compared with the TNF-alpha inhibitor bundle was £31,433 per QALY gained. The ERG stated that this analysis was subject to considerable uncertainty because the clinical effectiveness of sarilumab monotherapy was assumed to be equal to sarilumab in combination with conventional DMARDs. The committee acknowledged its decision that in the absence of direct evidence sarilumab monotherapy may have a similar clinical effectiveness to sarilumab plus conventional DMARDs in people whose disease has responded inadequately to biological DMARDs. The committee concluded that its recommendations for sarilumab plus conventional DMARDs in people whose disease has responded inadequately to biological DMARDs should also apply to sarilumab alone.