Adalimumab for treating moderate to severe hidradenitis suppurativa

Clinical management

4.1 The committee noted that hidradenitis suppurativa is a chronic inflammatory skin disorder characterised by recurrent painful boils – caused by blocked hair follicles – in areas with apocrine sweat glands, such as the groin and armpits. The committee was aware of consultation comments that symptoms may reduce after the menopause, but heard from the patient experts that symptom patterns differ from person to person. The patient experts explained that hidradenitis suppurativa has a substantial effect on every aspect of their quality of life. Patients can have as many as 30 active, open abscesses in 1 area at the same time, and the pain associated with this can be so severe that they are unable to climb stairs, do housework or look after their children. The committee was aware that patient-expert submissions stated that simply walking and moving in general becomes painful. The committee heard from the clinical and patient experts that this puts a strain on intimate physical relationships, family life and work, causing many people to lose their jobs and develop clinical depression. The patient experts reported that the clinical community lacks awareness of hidradenitis suppurativa and does not appreciate the severity of the condition. They expressed frustration at the many years it took to get a correct diagnosis, and highlighted the lack of available support. The clinical experts noted that people with hidradenitis suppurativa will have repeated and extensive surgeries over their lifetime, which is burdensome. The patient experts explained that it may take months to recover from surgery and return to work, and that the procedures result in painful scarring, which affects quality of life even when the disease is under control. The clinical experts noted that scarring, which is not a feature of other skin conditions such as psoriasis, is associated with its own comorbidities. They also emphasised the substantial psychological burden of the disease and noted that hidradenitis suppurativa is associated with increased mortality, which can be a result of physical complications such as sepsis, or people taking their own lives. The committee concluded hidradenitis suppurativa has a significant physical and psychosocial impact, which can be underestimated.

4.2 The committee discussed the clinical management of hidradenitis suppurativa. It was aware that there is no standard of care and no NICE guidance; there were no medical treatments specifically licensed for hidradenitis suppurativa until adalimumab received its marketing authorisation. The committee noted the results of a survey of the UK Dermatology Trials Network and British Association of Dermatologists, presented in the company submission, which showed that the most commonly used treatments in the UK – after topical antibiotics – are oral antibiotics; first tetracycline, and then a combination of clindamycin and rifampicin. The third, fourth, fifth and sixth most commonly used interventions in the survey were acitretin, isotretinoin, dapsone and ciclosporin respectively; the choice of treatment depends on individual patient characteristics. The committee noted the company statement that if the condition has not responded to these treatments, tumour necrosis factor (TNF)‑inhibitors, including adalimumab and infliximab, are used in the UK. The clinical experts agreed that the survey results accurately reflected the treatment options for hidradenitis suppurativa and that TNF‑inhibitors are only considered if the disease is not responding to other conventional treatments. However, they noted that not all of the treatments are supported by robust evidence in this indication. The committee heard from the clinical experts that surgery is done throughout a person's lifetime. The patient experts noted that repeat surgery and ongoing pharmacological treatment are needed because surgery only treats 1 area at a time. The committee concluded that it was appropriate for the company to position adalimumab after all other conventional treatment options.

4.3 The committee questioned whether infliximab would be an appropriate comparator for adalimumab. The clinical experts explained that, although infliximab is used to treat hidradenitis suppurativa, infliximab does not have a marketing authorisation for this indication and the evidence base is very limited; there is only 1 trial of infliximab in hidradenitis suppurativa and the trial population was very small. They explained that access to biologic treatments for hidradenitis suppurativa is restricted and funding is based on individual funding requests. Therefore, the committee did not consider infliximab to be an appropriate comparator for adalimumab because it is not established practice. The committee concluded that supportive care was the most appropriate comparator for adalimumab.

4.4 The committee considered how clinicians assess disease severity and response to treatment in people with hidradenitis suppurativa. The clinical experts considered that the Hidradenitis Suppurativa Clinical Response (HiSCR) is a reliable and reproducible tool, which has been validated for hidradenitis suppurativa and is relevant to clinical practice, but noted that the minimum clinically important difference has not yet been established. The clinical experts were aware that according to the validation study for the HiSCR measure, response to treatment was defined as a 50% reduction in total abscess and inflammatory nodule (AN) count, with no increase in abscesses or draining fistulas from baseline. However, the clinical experts considered that the 50% threshold was too high, and stated that a 25% reduction in AN count, provided there was no increase in abscesses or draining fistulas from baseline, would reflect a response to treatment. The clinical experts suggested that if the reduction in AN count was between 25% and 50%, they would continue with the existing treatment but may prescribe additional treatments to be taken at the same time (such as anti-inflammatories, retinoids and antibiotics) to improve response. The committee heard from the clinical experts that they would stop treatment if the reduction in AN count was lower than 25%, or if there was an increase in abscesses or draining fistulas. The clinical experts stated that it was important to also use patient-reported outcomes when monitoring people with hidradenitis suppurativa (in particular, the Dermatology Life Quality Index [DLQI], the pain visual analogue scale [VAS] and SF‑36, even though they are not specific to this indication), because physician-reported and patient-reported scores do not always correlate. The clinical experts considered that the minimum clinically important difference on the DLQI is 4 points, but commented that, because some people with chronic skin conditions can develop coping mechanisms and so adjust to the effect of the disease, the DLQI may underestimate the beneficial effects of treatment. The clinical experts stated that a 50% reduction in baseline pain is usually considered to be clinically meaningful. The committee concluded that it is appropriate to use the HiSCR for assessing response to treatment, with supporting information provided by patient-reported outcomes. The committee accepted how treatment failure would be defined in clinical practice using the HiSCR.

Clinical effectiveness

4.5 The committee discussed the clinical evidence for adalimumab and noted that people treated with adalimumab were more likely to have a clinical response (the primary end point of the trials) than people treated with placebo. The committee recognised that the difference between adalimumab and placebo was significant. The committee was aware that the benefit with adalimumab was greater in PIONEER II than PIONEER I, possibly because PIONEER II patients appeared to have had less severe disease than people in PIONEER I, and had potentially had higher levels of systemic antibiotics. The company noted that only 19% of patients in PIONEER II took oral antibiotics during the trial. The committee noted that the company had not originally done a formal meta-analysis of the data and was concerned that the company had given contradictory views on whether the PIONEER trials had similar or heterogeneous baseline characteristics, but concluded that the trials were generalisable to UK clinical practice. The committee considered the open-label extension study of adalimumab and was concerned that it only had data up to 72 weeks, given that adalimumab may be used for many years, and that full data were only available for 26% of enrolled patients. The committee concluded that adalimumab provided significant benefits compared with placebo, but that these had not been shown over the long term.

4.6 The committee discussed the health-related quality-of-life benefits associated with adalimumab and understood that adalimumab was associated with significant improvements in health-related quality of life compared with placebo after 12 weeks, as measured by the EQ‑5D in PIONEER II. The committee was aware that adalimumab showed a beneficial effect on the SF‑36 (collected in PIONEER I) and DLQI (collected in both PIONEER trials) but noted that the difference between adalimumab and placebo was not significant for all components of the SF‑36, and that the difference between arms in DLQI improvement at week 12 was not greater than the minimum clinically important difference. The committee discussed the mental component of the SF‑36, acknowledging that the change from baseline was not significantly different between the trial arms. The clinical experts explained that they would not expect to see a change in psychological burden of a chronic disease after only 12 weeks of treatment. The committee considered that the DLQI may have underestimated the beneficial effects of adalimumab, based on the clinical experts' comments that people with chronic skin conditions can develop coping mechanisms, which may result in lower DLQI scores than would be expected (indicating a better health-related quality of life; see section 4.4). The committee concluded that adalimumab had a statistically significant and clinically meaningful positive effect on health-related quality of life.

Cost effectiveness

4.7 The committee considered the structure of the company model and noted the company's justifications for modelling 4 health states according to the level of HiSCR response (see section 3.10). The committee considered it appropriate that the company had developed a more granular model than might have been expected given the dichotomous primary end point in the trials, because it reflected the how treatment success is defined in the clinical management of hidradenitis suppurativa. The committee was aware that response to treatment at 12 weeks determined whether people in the company's model continued having adalimumab, and understood that this reflected the marketing authorisation for adalimumab. The committee discussed the company's assumption that anyone with a partial response or higher at 12 weeks, defined as at least a 25% reduction in AN count with or without an increase in abscesses or draining fistulas from baseline, would continue adalimumab treatment. The clinical experts confirmed that it was reasonable to assume a 25% reduction in AN count would support treatment continuation (see section 4.4). However, they reiterated that if they saw an increase in abscesses or draining fistulas, which are very painful and troublesome complications indicating that adalimumab is not working, they would stop treatment. The committee concluded that the model structure was broadly appropriate for its decision-making, but would have preferred to see a model in which people stopped adalimumab treatment if abscesses or draining fistulas increased from baseline. After a request by the appraisal committee in the appraisal consultation document, the company submitted a scenario analysis in which this assumption was applied, through redefining the partial response and non-response health states.

4.8 The committee discussed the company's assumption, in its original model, that people in the non-response health state at 36 weeks or later would continue treatment for an additional 12 weeks, and so would not stop treatment until 48 weeks. The clinical experts disagreed with the assumption that treatment would be continued in people whose disease is not responding (see section 4.4), because this exposes people to a risk of adverse effects without giving any health benefits. The committee concluded that it was not appropriate to assume that people would continue having treatment if their disease is not responding to treatment (that is, if there is less than a 25% reduction in AN count, or an increase in abscesses or draining fistulas). In response to the appraisal consultation, the company submitted a revised base-case analysis in which people stopped treatment immediately if they were in the non-response health state at 36 weeks or later.

4.9 The committee discussed the company's application of clinical trial data in its original model. It considered that the company's use of an integrated arm-based summary to pool data from the 2 PIONEER trials, to inform the transition probabilities up to week 12 in the model, was inappropriate because it breaks the randomisation in the clinical trials and may have introduced bias in the analysis. The committee was also concerned that the transition probabilities from weeks 12–36 used different trial data depending on the treatment arm; the transition probabilities for the adalimumab arm came from pooled data, whereas only PIONEER II data were used for the supportive care arm. The company explained that this was a result of the clinical trial design (see section 3.3), but the committee was concerned that the approach created uncertainty and may have introduced bias in the model. The committee concluded that it would have preferred the company to do a formal random effects meta-analysis of both periods of the PIONEER trials to calculate the efficacy estimates in the model. In response to consultation, the company did a formal random effects meta-analysis of the 4 levels of HiSCR response from 2 PIONEER trials which was incorporated into the revised base case.

4.10 The committee considered the company's extrapolation of long‑term data in its original base-case analysis, beyond week 36. The committee heard the evidence review group's (ERG) concerns that the long‑term transition probabilities were not robust because they were based on a very small sample of data from the open-label extension study. The committee acknowledged that this could introduce a risk of bias and confounding in the model because of the study design and the inclusion of a select group of people who did not reflect the modelled population. The committee was also concerned that the company's use of different imputation methods (to account for missing data) for different arms of the model had the potential to introduce bias into the model. The committee concluded that the long‑term transition probabilities in the model would be more robust if extrapolation was based on data from the PIONEER trials and missing data were handled consistently. In response to consultation, the company submitted a scenario analysis in which long‑term transition probabilities were based on weeks 12–36 of the PIONEER trials instead of the open-label extension study.

4.11 The committee discussed the utility values in the company's model. The committee was satisfied with the company's rationale for not including adverse-event-related disutilities in the model. The committee considered it appropriate to use trial-based EQ‑5D data for utility values, in line with the NICE reference case, and agreed that the utility values for each health state seemed appropriate. However, the committee was concerned that the company had only used EQ‑5D data from PIONEER II and had not used any quality-of-life data from PIONEER I in the model, particularly because the benefit of adalimumab was lower in PIONEER I. In response to consultation, the company gave more information about how it calculated the utility values, including the number of patients used to inform the calculations for each level of response. The committee was aware that few patients in PIONEER II completed the EQ‑5D questionnaire at week 36, and noted an imbalance in the proportion of patients in each response category at this time point. The committee heard from the ERG that this could lead to bias in the utility values applied to the model, but the ERG was unsure of the size of the impact. The committee concluded that the utility estimates generated uncertainty in the model, but it was broadly satisfied with the company's approach given that the estimates came directly from trial-based EQ‑5D data.

4.12 The committee understood that the cost of surgical-inpatient admissions was a key cost driver in the model, and noted that the company did not change its assumptions about surgical-inpatient admissions in the revised model submitted in response to consultation. The committee was aware that the company had estimated the cost of inpatient surgeries using an online survey in which physicians were asked to estimate resource use for each of the 4 HiSCR health states in the model, based on the average baseline characteristics of patients in the trial. The committee was concerned that this would have been extremely difficult for physicians to estimate. In addition, the committee did not consider it appropriate to estimate resource use based on the level of HiSCR response in the absence of data from the clinical trials, because each health state would comprise patients with varying disease severity and different surgical needs. The committee heard from the ERG that it agreed with the company's estimate of total lifetime surgeries for the supportive care arm (33.87 surgeries). The ERG also considered that, based on clinical advice, it was not physically possible for a patient to have 34 wide excision procedures in their lifetime, as assumed by the company, and that most of these 34 procedures would be minor. The committee was aware that the ERG had estimated that someone with hidradenitis suppurativa would have 2 wide excisions in their lifetime. The clinical experts agreed that the company had overestimated the surgery-related resource use, and stated that most surgeries are minor procedures; wide excisions are less common. However, the clinical experts suggested that the ERG's alternative assumptions about surgical procedures may have underestimated the costs. The committee noted that comments received during consultation estimated that a patient might have 3–4 wide excisions in a lifetime; the ERG explored the impact of these new estimates on the incremental cost-effectiveness ratio (ICER). In the second appraisal committee meeting, the clinical experts suggested that the average number of wide excisions in a person's lifetime may exceed 4, based on research which shows that wide excisions are often associated with poor outcomes and disease recurrence, meaning that many people have repeated surgeries. The committee concluded that the company had overestimated resource-use costs for supportive care and adalimumab, and that the true values were closer to the ERG's estimates, but that the ERG may have underestimated the average cost of surgical-inpatient admissions in all of its exploratory analyses.

4.13 The committee discussed the company's assumption that adalimumab reduced the number of surgical-inpatient admissions compared with supportive care. The ERG and the clinical experts stated that there is no clinical evidence to support this assumption. However, the committee was aware of consultation comments suggesting that the disease control gained with adalimumab, combined with surgery, might lead to disease-freedom in some areas of the body and so reduce the need for major surgery in the long term. The committee noted clinical trial data showing that adalimumab reduces the number of minor surgeries, such as narrow margin excisions and incision and drainage procedures, and noted the consultation comments supporting this. The committee concluded that adalimumab reduces the need for some types of surgical procedure, but it could not make any definite conclusions about adalimumab's effect on the need for surgical-inpatient admissions in the absence of robust evidence.

4.14 The committee discussed the company's revised base-case cost-effectiveness analysis and noted that the company included the 2 amendments requested by the committee in the appraisal consultation document, which were that:

4.15 The committee was aware that the company's revised base case did not include its preferred definitions of partial response and non-response and that this was addressed in one of the company's scenario analyses. The committee noted that redefining the response health states in line with its preferred assumptions reduced the ICER substantially (see section 3.26). However, the committee heard from the ERG that the results of this scenario analysis were not reliable because the company had not accounted for the impact of redefining partial response and non-response on all relevant model parameters, such as utility values, adalimumab discontinuation caused by adverse events, and costs for the partial response and non-response health states. The committee concluded that the results of the company's scenario analysis were unreliable.

4.16 The committee discussed the company's second scenario analysis provided in response to consultation, in which the transition probabilities beyond week 36 were based on the PIONEER trials instead of the open-label extension study. The committee heard from the ERG that the company did not use the new transition probabilities estimated from the network meta-analysis done at the request of the committee. Instead, the company applied the transition probabilities for weeks 12–36 from the integrated arm-based summary in its original model, which the committee had already concluded were not robust (see section 4.9). The committee concluded that the results of this scenario analysis were also unreliable.

4.17 The committee attempted to identify the most plausible ICER for adalimumab compared with supportive care. The committee considered that the resource-use assumptions in the ERG's new exploratory analyses, provided after consultation (see section 3.30), were more realistic than the assumptions in the company's revised base-case model. The committee also preferred the ERG's assumption that there is no lifelong difference in prognosis between people who previously had adalimumab and then stopped treatment, and those who had never had the drug. It agreed with the ERG's corrected discontinuation rate for cycle 5 (see section 3.29 and section 3.30). Based on the ERG's exploratory analyses, the committee concluded that the maximum possible ICER for adalimumab compared with supportive care was between £28,500 and £33,200 per quality-adjusted life year (QALY) gained (based on the probabilistic analysis). However, the committee considered that the most plausible ICER would be lower than this for several reasons. First, the ERG's assumption of an average of 4 wide excisions over a patient's lifetime may be an underestimate, and the committee understood that the ICER reduced as the number of wide excisions increased. Second, the committee acknowledged that adalimumab may be associated with short‑term improvements in psychological wellbeing after treatment is stopped, and so considered that the ERG's assumption about prognosis was possibly pessimistic and may have overestimated the ICER. The committee also considered that if its preferred definitions of partial response and non-response had been incorporated in the ERG's exploratory analyses the ICER would have reduced, because continued treatment in people for whom a drug is not effective would be minimised. Taking these factors into account, the committee was certain that the most plausible ICER for adalimumab compared with supportive care was below the ERG's estimate of £28,500–£33,200 per QALY gained.

4.18 The committee heard from the patient experts that adalimumab was innovative in its potential to make a significant and substantial effect on health-related benefits. The committee understood that adalimumab is the only medical treatment with a marketing authorisation for hidradenitis suppurativa, and no other treatments offer effective long‑term disease control. The committee considered whether any gains in health-related quality of life were excluded from the QALY calculations. It understood that improvements in the psychological burden of hidradenitis suppurativa may not be captured in the QALY calculations, given the clinical experts' view that there is a time lag between reducing disease activity and seeing a benefit on patient-reported outcomes (see section 4.6). The committee also heard from patient experts that adalimumab might give enough disease control to allow people to return to work, which has an important positive impact on psychological wellbeing and feelings of self-worth. The committee heard from clinical and patient experts that the benefits associated with reducing the wound-care regimen needed during active disease, such as the time spent on wound care and the effect on work and family life, as well as the cost of dressings, were not captured in the model. The committee concluded that adalimumab is an effective treatment option for an extremely burdensome condition and may provide additional gains in health-related quality of life over those already included in the QALY calculations. Although the committee could not quantify the additional benefits of adalimumab, it considered that they would reduce the ICER compared with best supportive care. Taking this into account, alongside the committee's certainty that the ICER for adalimumab compared with supportive care was below £28,500 per QALY gained (section 4.17), the committee concluded that adalimumab is a cost-effective use of NHS resources in people with moderate to severe hidradenitis suppurativa whose disease has not responded adequately to conventional systemic therapy. The committee further concluded that the response to adalimumab treatment should be assessed after 12 weeks, in line with the marketing authorisation, and that treatment should only continue if there is clear evidence of response (defined as a reduction of 25% or more in the total abscess and inflammatory nodule count, with no increase in abscesses and draining fistulas; see section 4.7).

4.19 The committee considered whether its recommendations were associated with any issues related to the equality legislation and the requirement for fairness. The committee discussed comments from patient and professional organisations indicating that prevalence is greater in people of African family origin and in women, and some people with hidradenitis suppurativa have other disabilities; these characteristics are protected under the Equality Act 2010. The committee agreed that, because all people would be affected equally by its recommendations, there was no unfairness to any protected group.

4.20 The committee was aware of NICE's position statement on the Pharmaceutical Price Regulation Scheme (PPRS) 2014, and in particular the PPRS payment mechanism. It accepted the conclusion 'that the 2014 PPRS payment mechanism should not, as a matter of course, be regarded as a relevant consideration in its assessment of the cost effectiveness of branded medicines'. The committee heard nothing to suggest that there is any basis for taking a different view about the relevance of the PPRS to this appraisal. It therefore concluded that the PPRS payment mechanism was not relevant in considering the cost effectiveness of the technology in this appraisal.

Summary of appraisal committee's key conclusions