4 Committee discussion

The appraisal committee reviewed the data available on the clinical and cost effectiveness of belimumab, having considered evidence on the nature of active autoantibody-positive systemic lupus erythematosus and the value placed on the benefits of belimumab by people with the condition, those who represent them, and clinical experts. It also took into account the effective use of NHS resources.

4.1 The committee considered the nature of the condition, and noted evidence submitted and presented by the patient experts and clinical experts on the clinical signs and symptoms associated with systemic lupus erythematosus. The committee heard from clinical experts and patient experts how this disease is a debilitating condition, primarily affecting younger women. It affects daily life, including the ability to work and to have children. The clinical experts explained that people with systemic lupus erythematosus tend to die younger than the average population. The committee heard that there are very few licensed treatments for the disease and that people with systemic lupus erythematosus would welcome an additional treatment option specifically for this disease. Furthermore, it was highlighted that many people have to take several different drugs daily and that any treatment that might reduce this number would be welcomed. Reduced side effects of other drugs, especially corticosteroids, would also be welcome. The committee recognised the importance of the availability of treatment options for people with systemic lupus erythematosus and the need to reduce the side effects of immunosuppressants in current use.

4.2 The committee discussed the likely position of belimumab in clinical practice. The committee noted that standard care is likely to consist of non-steroidal anti-inflammatory drugs, corticosteroids, antimalarials or immunosuppressants. It also noted that the marketing authorisation for belimumab states that it should be used for people with high disease activity 'despite standard therapy'. The committee heard from clinical experts that 10–15% of people with systemic lupus erythematosus continue to have high disease activity despite standard therapy and that a proportion of these patients are currently treated with rituximab. The committee understood that rituximab is used in people with severe disease to reduce the levels of disease activity (that is, to induce remission) and to reduce the amount of corticosteroids and other immunosuppressants prescribed. The committee also heard from the clinical experts that rituximab treatment is repeated in such people when the disease shows signs of a significant increase in activity and that the re-treatment interval with rituximab varies from person to person. The clinical experts explained that they considered that rituximab would be a relevant comparator for belimumab. The committee noted that since September 2013 rituximab has been provided by NHS England through an interim clinical commissioning policy statement. The committee therefore concluded that standard care and rituximab should be comparators for belimumab.

4.3 The committee was aware that cyclophosphamide was also included as a comparator in the scope for the appraisal, but noted the company's justification that it was largely used for lupus nephritis, which was a different population to the one included in the trials of belimumab and covered by the marketing authorisation for belimumab. Furthermore, it heard from clinical experts that cyclophosphamide is used infrequently because of side effects.

4.4 The committee discussed how belimumab would be used in clinical practice. It noted that the marketing authorisation stated that treatment should be initiated and supervised by a qualified physician experienced in the diagnosis and treatment of systemic lupus erythematosus. It heard from the clinical experts that continuous use of belimumab for a long time would be very unlikely. The clinical experts explained that, as with other immunosuppressants, one of the aims of treatment with belimumab would be to work towards coming off the treatment. Once a person was in remission, belimumab treatment would be gradually stopped by reducing its frequency or dose. Serological activity would be monitored and belimumab treatment restarted if a person became symptomatic or if the serological tests signalled that this was likely. The company explained that there were no data available that reflected the scenarios described by the clinical experts, such as treatment holidays or tapering of treatment. However, the committee noted that the European Medicines Agency has requested that the company address uncertainties about the effect of stopping treatment with belimumab (treatment holidays) as well as the risk of rebound phenomena, as part of the routine pharmacovigilance programme. The committee was aware that belimumab is indicated as an add-on treatment in people with a high degree of disease activity despite standard therapy and it also observed that the European Medicines Agency's European Public Assessment Report for belimumab acknowledged that the BLISS studies were not designed for evaluating whether remission was induced, but rather for evaluating maintenance of remission. In addition the committee noted that the most recent data supporting longer‑term use of belimumab used a continuous schedule of administration over 7 years in people whose disease responded to treatment. Although the company had presented data supporting the continuous use of belimumab in people whose disease responded, the committee concluded that, in clinical practice, belimumab might be used in the same intermittent way as rituximab, although there are no efficacy data for such an approach and the likely treatment durations and discontinuation rates are not known.

4.5 The committee discussed the population in the company's decision problem. It noted that the company focused on a target population comprising a post hoc subgroup of the population covered by the marketing authorisation and the BLISS clinical trials. The target population was identified by a Safety of Estrogen in Lupus National Assessment – Systemic Lupus Erythematosus Disease Activity Index (SELENA-SLEDAI) score of greater than or equal to 10 and evidence of serological disease activity. The committee noted that although a SELENA-SLEDAI score of greater than or equal to 10 had been a pre-specified stratification factor in the BLISS clinical trials, when combined with the marketing authorisation criterion of a high degree of serological disease activity, this was not a group that had been pre-specified in the BLISS clinical trials. However, the committee heard from the clinical experts that, although the SELENA-SLEDAI score was not currently used in clinical practice to measure disease activity, people with a SELENA-SLEDAI score of greater than or equal to 10 would be those with clinically significant disease likely to be considered for treatment with belimumab. The committee also noted comments from consultation that a more routine use of the SELENA-SLEDAI score in clinical practice could improve the management of systemic lupus erythematosus. The experts also explained that the biomarkers mentioned in the marketing authorisation (that is, low complement and positive anti-double-stranded DNA antibodies), would be used for demonstrating evidence of serological disease activity and would detect changes in disease activity. The committee concluded that, although specifying a SELENA-SLEDAI score of greater than or equal to 10 may be considered arbitrary, the specified target population is clinically relevant.

Clinical effectiveness

4.6 The committee discussed the company's submission of clinical evidence, noting that most of the evidence in the company's submission was from the 2 BLISS trials (BLISS‑52 and BLISS‑76) that compared belimumab against standard care. The committee considered the composite end point of the Systemic Lupus Erythematosus Responder Index (SRI) used in the BLISS trials. It noted that this end point was developed in conjunction with the Food and Drug Administration in the US. The committee heard from the clinical experts that the SELENA-SLEDAI score, a component of the SRI, is a relatively crude tool and that the experts considered the use of the composite tool, which also includes the British Isles Lupus Assessment Group (BILAG) tool (as well as the physician's global assessment), was reasonable. The committee accepted the evidence from the clinical experts that the SRI was an appropriate end point in the trials.

4.7 The committee discussed whether the individual BLISS trials were representative of the UK population, in particular, whether data from the BLISS‑52 trial were as relevant to UK practice as data from the BLISS‑76 trial. The committee noted that the BLISS‑52 trial recruited people from eastern Europe, South America and Asia, and that the BLISS‑76 trial recruited people from Europe (western and eastern), the US, Canada and Israel. The clinical experts explained that, because the UK is a multi-ethnic country and systemic lupus erythematosus affects many ethnic groups more severely than white populations, data from different populations would still be relevant to the UK. Furthermore, the committee understood from the clinical experts that clinical practice varies between countries, for example, in the US higher doses of corticosteroids are used than in the UK. Therefore, there may also be issues about the relevance of the data from BLISS‑76. On balance, the committee concluded that BLISS‑76 was more representative of the population of England and Wales than BLISS‑52. However, data from BLISS‑52, and therefore from the pooled analysis would be relevant.

4.8 The committee discussed the characteristics of the patients in the BLISS trials. It noted that the patients in the BLISS trials had mainly immunological, mucocutaneous and musculoskeletal manifestations of systemic lupus erythematosus at baseline. The committee noted comments from consultation that the range of manifestations in the BLISS clinical trials was similar to those in clinical practice in the UK. Furthermore, it noted comments that serological manifestations are indicative of wider systemic disease activity. The committee discussed whether, on this basis, belimumab may be expected to also show benefits for other manifestations. The committee heard from clinical experts that if the experience of belimumab was like rituximab, then benefits for the range of manifestations may be expected. However, there remained uncertainty, and initially belimumab may be more likely to be used in people with predominantly musculoskeletal and mucocutaneous involvement. The committee concluded that currently the effect of belimumab on the full range of manifestations of systemic lupus erythematosus was uncertain.

4.9 The committee discussed baseline standard care in the 2 BLISS trials. It noted variations in treatments at baseline and that approximately 50% of people were receiving an immunosuppressant. The committee understood there was variability in clinical practice in the use of such drugs. However, it heard from the clinical experts that, in the UK, people for whom treatment with belimumab would be considered would have active disease despite standard therapy, and that standard therapy for most people would include an immunosuppressant. The committee concluded that there was uncertainty about the extent to which standard care in the belimumab trials represented UK clinical practice, for the target population for whom belimumab is intended.

4.10 The committee discussed the results of the BLISS trials and noted that, although in the individual trials the difference between the 2 arms for the primary outcome (the SRI) was statistically significant, the difference between the 2 arms for the components of the SRI were not statistically significant in BLISS‑76, with the exception of the SELENA-SLEDAI outcome. The committee also discussed the evidence of corticosteroid sparing, noting that a statistically significant reduction in corticosteroid use was observed in the pooled analysis. The committee noted the absolute reduction in use was about 1 mg per day in the economic model. The committee discussed the health-related quality of life outcomes in the clinical trials (EQ‑5D and SF‑36) and noted that, at week 52, no statistically significant differences between the treatment groups were reported in either trial for the target population. The committee also noted that the difference between the 2 arms for the functional assessment of chronic illness therapy (FACIT)-fatigue scores was not statistically significant at week 52 in the target population. The committee concluded that, compared with standard care, there was some evidence of the clinical effectiveness of belimumab. However, the evidence of effect was observed with greater consistency across outcomes in the BLISS‑52 trial. Furthermore, the relevance of both the pooled and unpooled data to a UK population was associated with a number of uncertainties in terms of the patient populations enrolled, the nature of standard care and the effects of belimumab on the full range of possible manifestations of systemic lupus erythematosus (see sections 4.7–4.9).

4.11 The committee discussed the long‑term data provided by the company from a 6‑year follow‑up report of the extension of the phase II study. The committee recognised that this study had been provided by the company primarily as additional evidence about long‑term reduction in corticosteroid dose, but noted that data from the study suggested continued clinical benefit from belimumab treatment over a 6‑year period. The committee first discussed the data for reduction in corticosteroid dose, noting that these showed an absolute reduction in corticosteroid dose of 5 mg a day at 6 years. The committee then noted the sustained improvement over 6 years in measures of disease activity (such as the SRI response rate, reduced autoantibody and complement levels) and the reduced frequency of disease flares, as well as the fact that belimumab was generally well tolerated over 6 years. The committee considered that in the absence of a control group, the phase II data were unable to definitively demonstrate the clinical benefits of continuous belimumab treatment for patients whose disease responded, but the data were suggestive of continuing benefit. The committee heard from the evidence review group (ERG) that the reduction in corticosteroid use modelled in the economic analyses showed an absolute change in corticosteroid use for belimumab that was similar to the reduction seen in the phase II extension study. The committee concluded that these data suggested, but were not definitive proof of, a reduction in corticosteroids associated with belimumab treatment. However, the committee understood the importance of reduction in corticosteroid dose for patients and recognised the positive indications of these findings.

4.12 The committee explored the comparison of belimumab with rituximab and the evidence available to support the comparison, noting that head-to-head data comparing belimumab with rituximab were not available. It discussed the available evidence including the uncontrolled observational data and the comparative data for rituximab and placebo from the EXPLORER trial. It considered whether any indirect analysis of the EXPLORER and BLISS data could be conducted. The committee heard from the clinical experts that the EXPLORER trial included patients with more severe disease (that is, in terms of corticosteroid use and dose, and existing organ damage) than those in the BLISS studies, so the trial populations were different. The committee heard from the ERG that there were 3 outcomes for which an indirect comparison could be completed (that is, BILAG, SLEDAI and SF‑36 scores), but data were only available in the public domain for the SF‑36. The ERG also highlighted the differences in the trial populations, which it considered meant that the results of an indirect comparison were not meaningful. The committee concluded that there were no data that would allow a robust calculation of the relative clinical efficacy of belimumab compared with rituximab.

Cost effectiveness

4.13 The committee discussed the economic model submitted by the company that informed both the original and revised analyses. The committee noted that short‑term outcomes from the BLISS studies were linked to long‑term outcomes, using data from the Johns Hopkins cohort. The committee considered the similarity of people in the Johns Hopkins cohort to those in the BLISS trials and noted that the people in the BLISS trials had higher SELENA-SLEDAI scores than the average SLEDAI scores in the Johns Hopkins cohort, indicating that the populations in the trials had more active disease than in the Johns Hopkins cohort. The committee noted that the SLEDAI scores from the Johns Hopkins cohort were used to inform the equation for disease activity, corticosteroid use, mortality and organ involvement, but that only the equation for disease activity was adjusted so that it more closely matched the BLISS trial populations. The committee heard from the company how the model was driven by changes in the SELENA-SLEDAI score based on data from the Johns Hopkins cohort and that cost effectiveness was not particularly driven by other factors, such as by corticosteroid use. The committee accepted that attempting to link short‑term outcomes to long‑term outcomes was appropriate and recognised that there were limited data sources available with which to do this. However, it concluded that there was uncertainty about whether the equations derived from the Johns Hopkins data could be reliably applied to the target population because of differences in study populations.

4.14 The committee again discussed the expected duration of use of belimumab in clinical practice, noting that the original model predicted continuous treatment with belimumab for some people over the course of 40 years. The committee had concluded that continuous treatment over many years may not reflect how belimumab would be used in clinical practice (see section 4.4). However, it was aware that the summary of product characteristics for belimumab describes continuous use and noted the company's statements that there were no data available to model treatment holidays or tapering of treatment. In addition, the committee noted that the only data for longer‑term use of belimumab were for a continuous schedule of administration in patients whose disease responded to treatment. Therefore the committee was unable to make recommendations taking into account intermittent treatment or alternative administration schedules because there were no efficacy data or any evidence of the cost effectiveness of such an approach, despite suggestions from clinicians that belimumab may be used this way in UK clinical practice.

4.15 The committee discussed the analyses presented by the company in its response to consultation, which assumed continuous treatment, but limited to the maximum treatment duration of 6 years. The committee heard from the company that, taking into consideration the evidence from clinical experts at the first committee meeting and from other consultations with clinicians, it was likely that in clinical practice belimumab would not be used continuously over a lifetime. The company stated that belimumab would probably be used in the same way as other immunosuppressants in systemic lupus erythematosus, that is, people would discontinue belimumab as early as possible once sustained disease control was achieved. The company explained that the only long‑term data available on which to base treatment duration were the 6‑year data from the phase II extension study, hence the choice of 6 years. The committee heard from the clinical experts that, because of the heterogeneity of systemic lupus erythematosus, some people may need treatment continuously for longer than 6 years. But for most, it was more probable that belimumab would be used for less than 6 years, once a person's disease was in remission. The committee considered the implications of stopping belimumab treatment at 6 years. The committee noted that the data from the phase II extension study suggested there could be a possibility of continued benefit with continued treatment at 6 years because approximately 50% of patients on treatment with belimumab at the end of the second year were still on it at the end of the sixth year. The results of this study therefore suggested a rationale for continued use of belimumab in a significant proportion of patients beyond 6 years. The committee was also aware that the ERG had identified 7‑year data in relation to the belimumab phase II extension study. The committee concluded that, although the 6‑year maximum treatment duration modelled in the company's revised analyses improved the cost effectiveness of belimumab, the rationale for the choice of a maximum treatment duration of 6 years could not be considered sufficiently robust for use as the basis for decision-making.

4.16 The committee discussed the annual discontinuation rates for belimumab after the first 24 weeks assumed in the original and revised economic models. The committee noted that, in the original model, the company had based the annual discontinuation rate of 8% on data from the BLISS trials and that, in the company's additional evidence, the revised base case used a rate of 13%, based on longer‑term 6‑year data provided from the phase II extension study. The committee understood that the 13% rate had been revised by the ERG to 11.7% using the 7‑year data. The committee also noted that the company had included a scenario analysis that used a variable annual discontinuation rate of 13% for the first 5 years and 30% thereafter. The committee heard from the company that it considered the variable annual discontinuation rate better reflected the expected use of belimumab in UK clinical practice, in which clinicians had suggested that belimumab would be used like other immunosuppressants and most people would not receive belimumab beyond 6 years. The committee discussed the use of the variable discontinuation rate noting that it was not presented with any evidence supporting an increased rate of discontinuation after 5 years. The committee heard from the NICE Decision Support Unit (DSU) that the results of its survey of experts in lupus in the UK did not reduce the uncertainty surrounding the discontinuation rates or have more credibility than the estimates available from the BLISS or phase II extension studies (see sections 3.63–3.64). The committee also heard from the clinical experts that, without any long‑term evidence on the use of belimumab in clinical practice, it was difficult to provide estimates for the rate of discontinuation with belimumab. The committee considered that, because the effects of stopping belimumab treatment (rebound phenomena in patients whose disease has responded to belimumab) were not fully understood and because there was no evidence on treatment holidays and the efficacy of re-treatment with belimumab, this could make clinicians less willing to stop. The committee was therefore not persuaded that the proposed variable discontinuation rate was plausible.

4.17 The committee then discussed the alternative constant annual discontinuation rates. It questioned whether the discontinuation rate in the phase II extension study may have been higher because of the lower baseline disease activity observed in the patients in the study compared with the target population from the BLISS trials. Furthermore, the committee considered that there could be an interaction effect between the annual discontinuation rate and the response criteria for continuing treatment. Therefore, because no response criteria had been applied in the phase II extension study, the appropriateness of using the rates derived from the study were uncertain given the use of response criteria in the company's model. However, based on the phase II extension study, the committee accepted that the company may have underestimated the annual discontinuation rate in the original economic model, and noted that lower rates of discontinuation increased the incremental cost-effectiveness ratio (ICER). The committee considered it preferable to use an annual discontinuation rate from the available trial data and understood that the phase II extension study was the only available long‑term data source. It accepted the error identified by the ERG in the company's estimated rate of discontinuation. The committee concluded that the analysis using the ERG's estimated rate of 11.7% annual discontinuation was the most appropriate on which to base its decision.

4.18 The committee considered the continuation rules applied in the economic model, noting that the summary of product characteristics states that discontinuation of treatment with belimumab should be considered if there is no improvement in disease control after 6 months of treatment. The committee noted that the original economic model applied a rule that patients would continue treatment after week 24 if there was an improvement in their SELENA-SLEDAI score of 4 points or more and that, after consultation, an additional analysis using a more stringent rule needing an improvement of 6 points or more on the SELENA-SLEDAI scale had been proposed by the company. The committee understood that the SELENA-SLEDAI scale was not widely used in clinical practice, but noted comments that its introduction could improve patient care. The committee heard from the clinical experts that if the person had not shown any benefit from treatment with belimumab after 6 months of treatment, then they would be likely to discontinue treatment as per the summary of product characteristics. The committee heard that the clinical experts indicated that a gain of 4 points on the SELENA-SLEDAI score was generally considered to be a reasonable improvement and that if there was some benefit of treatment at 24 weeks, but less than 4 SELENA-SLEDAI points, the patient may continue treatment with belimumab. The committee then discussed the difference between the 4 and 6 point continuation rules. It noted that the BLISS trials had included a gain of 4 points on the SELENA-SLEDAI as a secondary outcome and as part of the composite primary outcome (SRI response) because it was regarded as clinically meaningful. It heard from the clinical experts that they would prefer the lower continuation rule of an improvement of 4 points in the SELENA-SLEDAI score, but would use the higher continuation rule of 6 points if it reduced the base-case ICER to an acceptable level. The committee agreed that specifying a continuation rule using an improvement in SELENA-SLEDAI score of either 4 or 6 points at 24 weeks could be considered arbitrary, but noted that these could help identify a population for which belimumab was more cost effective. It considered that a clinician's reluctance to discontinue belimumab in a person whose disease has responded but had not reached a fall in SELENA-SLEDAI score of 4 at 24 weeks would be magnified with a continuation rule of 6 points. On balance, it was persuaded that the application of continuation rules was appropriate, but concluded that, given the uncertainties about the application of SELENA-SLEDAI in clinical practice and the specification of 4 rather than 6 points as part of the primary end point in the clinical trials, it was not appropriate to consider using the more restrictive rule of a SELENA-SLEDAI score improvement of 6 or more as the basis for the most plausible ICER. However, because the application of the SELENA-SLEDAI score improvement of 6 or more improved the cost effectiveness of belimumab, it agreed that this alternative scenario should still be considered when it examined the range of estimates of the ICER.

4.19 The committee discussed the assumption in the economic model that the effect of belimumab was maintained over time. The committee heard from the clinical experts that there were limited data available about the maintenance of treatment effect in systemic lupus erythematosus. The clinical experts explained that, in other conditions such as rheumatoid arthritis, people on biological treatments can experience a reduction in the response to treatment over time. However, the clinical experts explained that, in their experience with systemic lupus erythematosus, those people whose disease responded to rituximab and who needed re-treatment with rituximab at a later stage had shown a good response to re-treatment. The committee was aware that the only longer‑term data identified by the company and the ERG in relation to the benefit of belimumab was the open-label phase II extension study that had been reported in conference abstracts and a journal publication (Petri et al. 2011; Merrill et al. 2012a; Merrill et al. 2012b). The committee concluded that there was still some uncertainty in the evidence about whether it was appropriate to assume that treatment effect was maintained over time. If the treatment effect was not maintained over time, this would lead to an increase in the ICER.

4.20 The committee discussed the modelling of response in the economic model. The committee noted the ERG comments that, for patients receiving belimumab whose disease did not respond to treatment at 24 weeks, it was assumed that at week 52 they had the mean benefit observed in the standard care group. The ERG stated that, because the standard care group included both patients whose disease had responded and not responded to standard care, this was likely to overestimate the benefit of belimumab. The ERG stated that a more appropriate approach would have been to model the changes for the group of patients whose disease did not respond to standard care. The committee heard from the company that, rather than reflecting the trial protocol treatment, the economic model was reflecting expected clinical practice with the introduction of a continuation rule, whereby patients whose disease does not respond to belimumab are switched to standard care and therefore get the benefits associated with standard care. The committee agreed that, in patients whose disease has not responded to belimumab at 24 weeks, some response was likely when the patient was switched to standard care, but the size of this effect was uncertain. The committee concluded that the company's approach may have overestimated the treatment effect of belimumab, and that alternative scenarios exploring the impact on the ICER with respect to the assumed mean benefit experienced by the patients whose disease did not respond to treatment in the belimumab group would help to better reflect the level of uncertainty.

4.21 The committee noted that the model outputs in the original base-case analysis demonstrated a gain in survival of 2.9 years from treatment with belimumab compared with standard care. The committee considered the predicted survival from the model, noting that there was no evidence from the trials to support this modelled outcome and that in the trials there was a trend towards higher mortality in the belimumab arms compared with standard care. The company explained that the modelled benefit was expected as a result of reduced or delayed damage to organ systems, which would in turn have an effect on mortality risk. The committee heard from the clinical experts that people with higher disease activity are more likely to have organ damage and die than people with lower disease activity. However, the clinical experts stated that this was likely to be dependent on the site of organ damage. For example, treatment for people with mainly musculoskeletal or mucocutaneous damage was unlikely to result in a survival benefit. The committee was also aware that, because of the prolonged life expectancy of people treated with belimumab, the duration of damage for the other organ systems is increased, affecting cost and health-related quality of life. The committee also discussed how survival time in the model was predicted to be longer in the target population than in the overall trial population (31.9 years in the standard care arm of the target group compared with 30.5 years in the overall standard care arm in the overall pooled BLISS populations), even though the target population had more severe disease. The committee noted comments from consultation that this was because of the different baseline ages of the target and trial populations. The committee considered that, although the different ages at baseline accounted for the survival difference, it noted that the age of death remained the same for both age groups. This was considered to be an unexpected finding given the longer disease history of the younger age group. The committee considered that the company should have explored uncertainty around the estimate of survival in the model and its impact on the ICER by exploring a scenario that assumed no survival gain for treatment with belimumab compared with standard care. The committee concluded that, although gains in survival from reduced organ damage were plausible, there was considerable uncertainty around the validity of the modelled gains in survival and that alternative scenarios exploring the impact to the ICER around this parameter would better reflect the level of uncertainty.

4.22 The committee considered the standardised mortality ratios used by the company and the alternative values identified by the ERG. The committee heard from the ERG that the values it identified were unpublished data from an English cohort of patients. The committee heard from the clinical experts that they considered that the standardised mortality ratios provided by the company appeared more appropriate, but highlighted in both sets the very high mortality ratios for the youngest ages (for people aged 24 years or younger). The committee noted that the model was only modestly sensitive to the use of alternative standardised mortality ratios. The committee concluded that it was appropriate to use the mortality ratios provided by the company in its decision-making.

4.23 The committee discussed the administration costs used in the economic model. It noted that in the original model a cost of £126 had been used, based on 2 hours of specialist nurse time. The committee noted that that this may be an underestimate of the costs of administration and noted that the ERG had completed a number of scenario analyses using values based on day-case codes and also values used in previous appraisals of intravenous monoclonal antibodies for rheumatoid arthritis (£154). Furthermore, the committee noted comments from consultation that pharmacy preparation time had not been included in the economic analyses. The committee noted that the company included the updated administration cost of £154 in its revised submission. The committee concluded that the administration cost of £154 was appropriate to use in its consideration of the most plausible ICER.

4.24 The committee discussed the costs and utilities in the model. The committee heard from clinical experts that some of the costs and disutilities may not be accurately captured, specifically the difference in costs associated with renal disease (£1,765 in the first year and £2,453 in the second year) compared with those associated with pulmonary disease (£9,679 and £9,603 respectively). The committee also noted, for example, that the disutility multiplier in the first year of organ involvement for the serious consequence of renal involvement was 0.97, whereas for musculoskeletal organ damage the corresponding figure was 0.67. The committee expected that the disutility multiplier for renal involvement would be lower than 0.97. The clinical experts further highlighted that the assumption that disutilities and costs were the same in second and subsequent years may underestimate the effects of reducing or delaying organ damage because some types of damage, such as renal damage, were associated with increasing costs and reduced health-related quality of life, as damage progresses and people need haemodialysis. The committee concluded that deriving cost data from different sources may have led to some inconsistencies in the estimates and that the company may have underestimated some of the benefits associated with delaying certain types of organ damage.

4.25 The committee considered the cost effectiveness of belimumab in comparison with standard care. The committee recognised that a scenario reflecting lifetime continuous treatment may not accurately capture how belimumab would be used in clinical practice. It noted that the company presented updated base-case ICERs with 3- and 5‑year maximum treatment durations that incorporated the patient access scheme, and were based on the committee's previously preferred assumptions. It observed that they were within the range normally considered a cost-effective use of NHS resources, and that this was also the case when using the company's non-responder scenario. However, the committee considered the assumption that all patients stopped at the appropriate time, regardless of the benefit they were receiving, to be unlikely and that it did not reflect clinical practice. The committee noted that there is no evidence of the effectiveness of such a strategy. It also noted that the ERG's exploratory analysis based on the company's non-responder scenario with a lifetime maximum treatment duration increased the ICER beyond the range normally considered to be cost effective. The committee considered that the estimated ICERs with the revised patient access scheme may have been underestimated because of a number of uncertainties. These included linking the short‑term trial outcomes to long‑term data with differing study populations, the annual discontinuation rate, the maintenance of the treatment effect, the mean benefit assumed for patients receiving belimumab whose disease did not respond to treatment at 24 weeks, and validity of the modelled gains in survival that remained in the economic modelling (see sections 4.13, 4.16–4.21). The committee agreed that, because of the considerable uncertainty that remained in the economic modelling, it was unable to conclude that the true value of the ICER incorporating the revised patient access scheme was within or outside a range in which belimumab could be considered a cost-effective use of NHS resources compared with standard care.

4.26 The committee considered the cost effectiveness of belimumab compared with rituximab. The committee had previously discussed the clinical effectiveness of rituximab in comparison with belimumab (see section 4.4) and concluded that no reliable data were available to demonstrate the relative efficacy of belimumab in comparison with rituximab. It recalled that although rituximab was not licensed for treating systemic lupus erythematosus, it was available in the NHS because it was funded by NHS England via an interim clinical commissioning policy statement. The committee concluded that without any comparison of the clinical effectiveness of belimumab with rituximab, it could not reach a conclusion as to the cost effectiveness of belimumab compared with rituximab as an add-on therapy in adults with active autoantibody-positive systemic lupus erythematosus with a high degree of disease activity (for example, positive anti-double-stranded DNA and low complement) despite standard therapy. It further concluded that collecting data for belimumab and rituximab via the BILAG registry could provide data for a future comparison of the cost effectiveness of the 2 treatments when the NICE technology appraisal guidance on belimumab for systemic lupus erythematosus is reviewed.

4.27 The committee discussed the innovative nature of belimumab. It specifically noted the comments from clinical experts and patient experts that few drugs are licensed for treating systemic lupus erythematosus, and the comment from the company that belimumab was developed to target the underlying pathology of this disease. The committee also discussed whether any health-related quality-of-life benefits may not have been captured in the calculation of the quality-adjusted life years (QALYs). It was aware that disease flares had not been fully included in the economic modelling and that the company stated that this could underestimate the benefits of treatment. The committee noted that in the BLISS trials differences in EQ‑5D were demonstrated between treatment groups but that this was not statistically significant at 52 weeks, the longest follow‑up time for which quality-of-life data are available for the target population. Furthermore, there were no statistically significant differences at week 52 for FACIT-fatigue scores in the target population in people receiving belimumab compared with people receiving standard care. The committee was not persuaded that the clinical evidence submitted strongly indicated that the changes in health-related quality of life from belimumab had not been adequately captured. The committee concluded that the issues identified around innovation did not change its conclusions about the cost effectiveness of belimumab.

4.28 The committee was aware of a potential equality issue relating to the lower response rates observed in the clinical trials for the subgroup of patients of African American or African origin. The committee also noted comments received during consultation that systemic lupus erythematosus predominantly affects women of child-bearing age from ethnic minority groups. Given that the recommendations do not differentiate between any groups of people, the committee concluded that its recommendations do not limit access to the technology for any specific group, compared with other groups.

Proposed managed access arrangements

4.29 The committee discussed the general concept of a managed access arrangement, which described the arrangements and responsibilities for the use of, and further data collection for, belimumab for treating active autoantibody-positive systemic lupus erythematosus, and agreed it would be useful. It reviewed the structure of the proposed managed access arrangement that had been developed. It concluded that this approach was a useful tool that should be used to achieve consensus among the key stakeholders in the event of a positive recommendation with further evidence collection in the NHS in England.

4.30 The committee heard from the company that, if belimumab was recommended by the committee, the NHS would be able to recruit people with active autoantibody-positive systemic lupus erythematosus to the UK BILAG registry and collect real-world evidence on the safety and efficacy of belimumab to address the key uncertainties, including rates of discontinuation. The committee noted that the data could inform a future comparison of belimumab and rituximab. The committee considered the nature of the evidence that could be collected from the BILAG registry according to the company's proposal and whether this would be useful in future decision-making. The clinical experts acknowledged that observational registries have limitations, and that randomised controlled trials are the gold standard. However, the committee heard from a clinical expert that there are challenges to conducting randomised controlled trials in this therapy area, such as the small number of patients in the UK and ethical constraints in balancing corticosteroid use, high-quality care and demonstrating a treatment effect. The clinical experts emphasised the value of the BILAG registry in providing real-world evidence from UK clinical practice. The committee heard that few of the patients in the ongoing clinical trials are in the UK, and several of the studies have assessments that are less robust than the registry (for example, the physician's global assessment of improvement). The committee concluded that collecting data via the BILAG registry had the potential to provide additional data that could be useful in a future technology appraisal of belimumab, including a comparison with rituximab.

4.31 The committee discussed the potential sample size and duration of treatment in the registry. The committee heard that if belimumab were recommended, the clinical experts anticipated that 10–12 new patients receiving belimumab and 6–8 receiving rituximab would join the registry each month. It heard from NHS England that it would expect a licensed, NICE-recommended treatment to be prescribed over one used outside its marketing authorisation, but that the decision would ultimately come down to physician and patient choice. The committee noted that if the research were carried out for 3 years, and 10–12 patients were recruited each month, very few patients would have 3‑year follow‑up data at 3 years. The committee concluded that a minimum of 3 years would be needed for data collection from the registry and that it was likely that up to 5 years would be needed to collect a meaningful dataset.

4.32 The committee questioned the benefit of obtaining data from the registry over and above the existing ongoing trials and whether it was needed to inform a review of the guidance. It noted the DSU's opinion that, in relation to most uncertainties, the proposed evidence collection is ongoing in other existing studies. The committee recalled that the company's economic modelling included a treatment continuation rule that specified belimumab would only be continued in people whose SELENA-SLEDAI score dropped by 4 points or more. It agreed that it was appropriate to introduce this continuation rule in clinical practice (see section 4.18). The committee understood that belimumab discontinuation practice would therefore be expected to differ in UK clinical practice compared with non‑UK trials, and agreed that the data captured should include dose and timing data, discontinuations and adherence to the treatment continuation rule. It agreed that the registry would provide useful data on the treatment continuation rule and would be unique in providing efficacy and safety data for the population affected by it. It also heard that, unlike the clinical trials, the registry would provide lupus-specific data on health-related quality of life. The committee concluded that although the ongoing clinical trials could potentially address many of its uncertainties, the BILAG registry would provide some unique data, particularly on the treatment continuation rule.

4.33 The committee queried whether there were any irrecoverable NHS costs associated with establishing the registry. It noted that there would be clinician and administrator time involved in recruiting and following-up patients and in adding data to the database. However, it heard from the clinical experts that the registry has established funding and has already been set up with NHS England to provide data on existing treatments such as rituximab and standard care. The committee concluded that there were no significant irrecoverable NHS costs associated with starting the belimumab research.

4.34 The committee considered what might happen if belimumab were initially recommended (with further evidence collection) but, after a subsequent review of the technology appraisal guidance, was later found not to be cost effective. If belimumab was subsequently not recommended, the committee considered that it would be clinically unacceptable for belimumab treatment to be withdrawn from all patients after 3 or 5 years. It noted the company's proposed exit strategy, which is confidential and cannot be included here. It emphasised the importance of ensuring high-quality care for patients, and considered that withdrawing belimumab, at an arbitrary time, from people whose disease was responding to the drug, would be both inappropriate and impractical. The committee concluded that NHS England needed reassurance from the company about how a person's care would continue in these circumstances.

4.35 The committee discussed whether belimumab met the criteria for recommending a treatment with research according to the guide to the methods of technology appraisal 2013. In considering each of these criteria, the committee agreed the following:

  • Data collected using the BILAG registry would inform the future development of NICE guidance and clinical practice on using belimumab.

  • Reconsidering the decision in light of research findings would be likely to reduce the uncertainty in the economic model.

  • The research will be feasible even if belimumab is recommended for NHS use outside the context of research.

  • No significant irrecoverable costs would be incurred when introducing the technology.

    The committee concluded that this meant belimumab was a potential candidate for recommending with further evidence collection, but had reservations about the risks involved.

4.36 The committee then explored whether the potential value to the NHS of the recommended research is likely to represent good value in the context of limited research resources. It noted that the company proposed funding the research and that NHS England would fund the provision of the technology in the NHS. The committee considered the 'budget impact' cost comparison between belimumab and rituximab. It heard from the clinical expert that people tend to receive rituximab approximately every 12–15 months on average, and was aware that belimumab's marketing authorisation recommends continuous use. The committee noted that it had been estimated that there was an additional cost of using belimumab rather than rituximab over 5 years, assuming 1 dose of rituximab per year and using belimumab continuously (that is, as recommended in belimumab's summary of product characteristics). The committee recalled that it had heard from the clinical experts that belimumab could be used in a similar intermittent way to rituximab in clinical practice (see section 4.4). It considered that this would reduce the costs of belimumab compared with administering it as specified in the summary of product characteristics. The committee agreed that the company's calculations could overestimate the cost of belimumab, depending on how it is used in clinical practice. The additional cost of using belimumab rather than rituximab is confidential and cannot be presented here. The committee concluded that in the context of limited research resources, the proposed research would represent good value, only if NHS England finds the financial arrangements and cost to the NHS acceptable compared with current alternative biological treatment options.

4.37 The committee deliberated over making a recommendation on belimumab:

  • It agreed that it was unable to conclude if the true value of the ICER incorporating the revised patient access scheme for belimumab compared with standard care was within or outside a range in which belimumab could be considered a cost-effective use of NHS resources (see section 4.25).

  • It observed that the clinical and cost effectiveness of belimumab compared with rituximab could not be determined because of a lack of clinical evidence. The committee recalled that rituximab, which does not have a marketing authorisation for systemic lupus erythematosus, was being routinely funded in the NHS via an interim clinical commissioning policy statement. It acknowledged that belimumab does have a marketing authorisation for treating systemic lupus erythematosus (see section 2.1). It considered that a comparison of the 2 treatments could be made when the guidance is reviewed, informed by evidence collected via the BILAG registry, which would include some unique data that would not be provided through ongoing clinical trials (see section 4.26 and section 4.32).

  • It considered that collecting real-world data from UK patients had additional benefits to support the data generated in the global belimumab clinical trial programme, such as better generalisability of the patient population and standard care that more closely represents clinical practice in England (see section 4.7, section 4.9 and section 4.30).

  • It agreed that belimumab met the criteria for recommending a treatment with research according to the guide to the methods of technology appraisal 2013 (see section 4.35).

  • It believed that the proposed research would represent good value in the context of limited research resources, as long as NHS England found the financial arrangements and cost to the NHS acceptable compared with those for current alternative biological treatment options (see section 4.36).

    Taking all of these factors into account, the committee considered that it was appropriate to grant access to belimumab to enable further evidence collection to inform a comparison with rituximab and reduce the uncertainty in the company's economic modelling when the technology appraisal guidance is reviewed. The committee concluded that it recommended belimumab as an option, with further evidence collection, for treating active antibody-positive systemic lupus erythematosus in people with evidence for serological disease activity (defined as positive anti-double-stranded DNA and low complement) and SELENA-SLEDAI score of greater than or equal to 10.

4.38 The committee discussed the review of the technology appraisal guidance. It heard that it could take over 6 years to complete evidence collection, data analysis and re-appraisal. The committee was aware that guidance is normally considered for review 3 years after publication. The committee recalled that several ongoing trials would provide useful information to inform a re-appraisal, and that potentially many current areas of uncertainty could be addressed without the registry data. The committee concluded that although the BILAG registry would provide some unique data, NICE should consider the technology appraisal guidance for belimumab for review 3 years after publication, as is standard, to determine whether the main uncertainties could be addressed using clinical trial data, as well as any early registry data, available at that time. It further concluded that if a review was not considered appropriate at 3 years, a mandatory review should be done no later than 5 years after publication.

4.39 The committee discussed the implications of belimumab getting a negative recommendation following re-appraisal. The committee considered that issuing a positive recommendation in the context of a proposed managed access arrangement meant that there was a significant risk of NHS England paying for a treatment that was not cost effective, and noted that this risk could be borne for up to 6 years after guidance publication, depending on when additional data to inform a review become available. It believed that it was unacceptable for NHS England to bear all of this risk and considered it reasonable that NHS England and the company should establish a risk-sharing strategy in the proposed managed access arrangement. It further concluded that NHS England and the company should agree an appropriate level of numbers of patients needing to be treated during the data collection period and an exit strategy for patients still receiving belimumab in the NHS, in the event that belimumab was not shown to be cost effective at the end of the data collection period.

Summary of appraisal committee's key conclusions

TA397

Appraisal title: Belimumab for the treatment of active autoantibody-positive systemic lupus erythematosus

Section

Key conclusion

Belimumab is recommended as an option as add-on treatment for active autoantibody-positive systemic lupus erythematosus in adults only if all of the following apply:

  • There is evidence for serological disease activity (defined as positive anti-double-stranded DNA and low complement) and a Safety of Estrogen in Lupus National Assessment – Systemic Lupus Erythematosus Disease Activity Index (SELENA-SLEDAI) score of greater than or equal to 10 despite standard treatment.

  • Treatment with belimumab is continued beyond 24 weeks only if the SELENA-SLEDAI score has improved by 4 points or more.

  • The company provides belimumab with the discount agreed in the patient access scheme.

  • Under the conditions for data collection, monitoring, patient eligibility and consent, ongoing treatment, cost to the NHS, and review by NICE as laid out in sections 5 and 6 of this document.

As a condition of its positive recommendation, the committee instructed that data on efficacy, safety and quality-of-life should be collected using the BILAG registry to resolve uncertainties in a future review of this technology appraisal guidance.

The committee agreed that, because of the considerable uncertainty that remained in the economic modelling, it was unable to conclude that the true value of the incremental cost-effectiveness ratio (ICER) incorporating the revised patient access scheme was within or outside a range in which belimumab could be considered a cost-effective use of NHS resources compared with standard care.

There are no data that would allow a robust calculation of the relative clinical efficacy of belimumab compared with rituximab. For the comparison of belimumab with rituximab the committee concluded that, without any comparison of the clinical effectiveness of belimumab with rituximab, it could not reach a conclusion as to the cost effectiveness of belimumab compared with rituximab as an add-on therapy in adults with active autoantibody-positive systemic lupus erythematosus with a high degree of disease activity despite standard therapy.

1.1, 4.37, 6.1–6.3, 4.25, 4.12, 4.26

Current practice

Clinical need of patients, including the availability of alternative treatments

Systemic lupus erythematosus is a debilitating condition, primarily affecting younger women. It affects daily life, including the ability to work and to have children. People with systemic lupus erythematosus tend to die younger than the average population. There are very few licensed treatments for the disease and patients would welcome a new treatment option.

4.1

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 treatment might be corticosteroid sparing and may reduce the side effects of other drugs, especially corticosteroids.

Few drugs are licensed for treating systemic lupus erythematosus. Belimumab was developed to target the underlying pathology of this disease. However, the committee was not persuaded that the clinical evidence submitted strongly indicated that the changes in health-related quality of life from belimumab had not been adequately captured.

4.1, 4.10, 4.11, 4.27

What is the position of the treatment in the pathway of care for the condition?

Between 10% and 15% of people with systemic lupus erythematosus have high disease activity despite standard therapy. The committee therefore concluded that standard care and rituximab should be comparators for belimumab.

4.2

Adverse reactions

Adverse reactions were not a key factor in this appraisal.

Evidence for clinical effectiveness

Availability, nature and quality of evidence

Most of the evidence in the company's submission was from the 2 BLISS trials (BLISS‑52 and BLISS‑76) that compared belimumab against standard care.

There are no data that would allow a robust calculation of the relative clinical efficacy of belimumab compared with rituximab.

4.6, 4.12

Relevance to general clinical practice in the NHS

The committee concluded that, although BLISS‑76 was more representative of the population of England and Wales than BLISS‑52, data from BLISS‑52, and therefore from the pooled analysis, would be relevant.

4.7

Uncertainties generated by the evidence

The relevance of both the pooled and unpooled data from the BLISS trials to a UK population was associated with a number of uncertainties in terms of the patient populations enrolled, nature of standard of care and effects of belimumab on the full range of possible manifestations of systemic lupus erythematosus.

4.7–4.10

Are there any clinically relevant subgroups for which there is evidence of differential effectiveness?

The company focused on a target population comprising a subgroup of the marketing authorisation population and BLISS clinical trials. The target population was identified by a Safety of Estrogen in Lupus National Assessment – Systemic Lupus Erythematosus Disease Activity Index (SELENA-SLEDAI) score of greater than or equal to 10 and evidence of serological disease activity. The committee concluded that, although specifying a SELENA-SLEDAI score of greater than or equal to 10 may be considered arbitrary, the specified target population is clinically relevant.

4.5

Estimate of the size of the clinical effectiveness including strength of supporting evidence

The committee concluded that, compared with standard care, there was some evidence of the clinical effectiveness of belimumab. However, the evidence of effect was observed with greater consistency across outcomes in the BLISS‑52 trial. Further, the relevance of both the pooled and unpooled data from the 2 BLISS trials to a UK population was associated with a number of uncertainties in terms of the patient populations enrolled, nature of standard care and effects of belimumab on the full range of possible manifestations of systemic lupus erythematosus.

4.10

Evidence for cost effectiveness

Availability and nature of evidence

The company submitted an economic model in which short‑term outcomes from the BLISS studies were linked to long‑term outcomes, using data from the Johns Hopkins cohort.

4.13

Uncertainties around and plausibility of assumptions and inputs in the economic model

The committee accepted that attempting to link short‑term outcomes to long‑term outcomes was appropriate and recognised that there were limited data sources available with which to do this. However, it concluded that there was uncertainty about whether the equations derived from the Johns Hopkins data could be reliably applied to the target population because of differences in study populations.

The committee understood that continuous treatment over many years was unlikely to reflect how belimumab would be used in clinical practice. However, the summary of product characteristics for belimumab describes continuous use as the model for administration. Although the 6‑year maximum treatment duration modelled by the company in its revised analyses improved the cost effectiveness of belimumab, the rationale for the choice of 6 years could not be considered sufficiently robust for use as the basis for decision-making.

The committee concluded that the company may have underestimated the annual discontinuation rate in the original economic model, and therefore overestimated the ICER, and that a higher rate of annual discontinuation as observed in the phase II extension study may be more appropriate. The committee was not persuaded that the proposed variable discontinuation rate of 13% for the first 5 years and 30% thereafter, as presented by the company, was plausible. The committee heard from the NICE Decision Support Unit that the results of its survey of experts in lupus in the UK did not reduce the uncertainty surrounding the discontinuation rates or have more credibility than the estimates available from the BLISS or phase II extension studies. The committee also heard from the clinical experts that, without any long‑term evidence on the use of belimumab in clinical practice, it was difficult to provide estimates for the rate of discontinuation with belimumab. The committee considered it preferable to use an annual discontinuation rate from the available trial data and understood that the phase II extension study was the only available long‑term data source. The committee concluded that it was more appropriate to use the ERG's rate of 11.7% annual discontinuation.

There was still some uncertainty in the evidence about whether it was appropriate to assume that treatment effect was maintained over time. If treatment effect was not maintained over time, this would lead to an increase in the ICER.

The committee noted the ERG comments that, for patients receiving belimumab whose disease did not respond to treatment at 24 weeks, it was assumed that at week 52 they had the mean benefit observed in the standard care group. The committee concluded that the company's approach may have overestimated the treatment effect of belimumab.

Although gains in survival from reduced organ damage were plausible, there was considerable uncertainty around the validity of the modelled gains in survival.

Deriving cost data from different sources may have led to some inconsistencies in the estimates and the company may have underestimated some of the benefits associated with delaying certain types of organ damage.

4.13–4.17, 4.19–4.21, 4.24

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 also discussed whether any health-related quality-of-life benefits may not have been captured in the calculation of the quality-adjusted life years (QALYs). It was aware that disease flares had not been fully included in the economic modelling and that the company stated that this could underestimate the benefits of treatment. However, the committee was not persuaded that the clinical evidence submitted strongly indicated that the changes in health-related quality of life from belimumab had not been adequately captured.

4.27

Are there specific groups of people for whom the technology is particularly cost effective?

The company focused on a target population, comprising a subgroup of the population covered by the marketing authorisation and the BLISS clinical trials. The target population was identified by a SELENA-SLEDAI score of greater than or equal to 10 and evidence of serological disease activity.

4.5

What are the key drivers of cost effectiveness?

The committee considered that the estimated ICERs with the revised patient access scheme may have been underestimated because of the uncertainties associated with the linking of short‑term trial outcomes to long‑term data with differing study populations, the discontinuation rate, the treatment effect, the mean benefit assumed for patients receiving belimumab whose disease did not respond to treatment at 24 weeks, and validity of the modelled gains in survival that remained in the economic modelling.

4.25

Most likely cost-effectiveness estimate (given as an ICER)

The committee agreed that, because of the considerable uncertainty that remained in the economic modelling, it was unable to conclude that the true value of the ICER incorporating the revised patient access scheme was within or outside a range in which belimumab could be considered a cost-effective use of NHS resources compared with standard care.

The committee concluded that, without any comparison of the clinical effectiveness of belimumab with rituximab, it could not reach a conclusion as to the cost effectiveness of belimumab compared with rituximab as an add-on therapy in adults with active, autoantibody-positive systemic lupus erythematosus with a high degree of disease activity (for example, positive anti-double-stranded DNA and low complement) despite standard therapy.

4.25, 4.26

Additional factors taken into account

Patient access schemes (PPRS)

The company has agreed a patient access scheme with the Department of Health that would provide a simple discount to the list price of belimumab. The level of the discount is commercial in confidence.

2.3

End-of-life considerations

End-of-life considerations were not discussed.

Equalities considerations and social value judgements

The committee was aware of equalities issues relating to the lower response rates observed in the clinical trials for the subgroup of patients of African American or African origin, and that systemic lupus erythematosus predominantly affects women of child-bearing age from ethnic minority groups. Given that the recommendations do not differentiate between any groups of people, the committee concluded that its recommendations do not limit access to the technology for any specific group, compared with any other group.

4.28

Other

The committee concluded that collecting data via the BILAG registry had the potential to provide additional data that could be useful in a future technology appraisal of belimumab, including a comparison with rituximab.

The committee concluded that a minimum of 3 years would be needed for data collection from the registry and that it was likely that up to 5 years would be needed to collect a meaningful dataset.

The committee concluded that although the BILAG registry would provide some unique data, NICE should consider the technology appraisal guidance for belimumab for review 3 years after publication, as is standard, to determine whether the main uncertainties could be addressed using clinical trial data, as well as any early registry data, available at that time. It further concluded that if a review was not considered appropriate at 3 years, a review must be done no later than 5 years after publication.

4.30, 4.31, 4.38

  • National Institute for Health and Care Excellence (NICE)