Nalmefene for reducing alcohol consumption in people with alcohol dependence

Clinical effectiveness

4.4 The Committee considered the evidence on the clinical effectiveness of nalmefene plus psychosocial support, noting that the evidence was derived from the ESENSE1, ESENSE 2 and SENSE studies. It discussed whether the population in the 3 studies reflects those seen in clinical practice in England, and whether it could allow clinicians to determine the population eligible for nalmefene. The Committee noted from the trials that patients must be diagnosed as having alcohol dependency using the Diagnostic and Statistical Manual of Mental Disorders (DSM‑IV‑TR), with an average daily alcohol consumption classed as medium risk or higher (more than 40 g [5 units] per day for men and more than 20 g [2.5 units] per day for women) with at least 6 heavy drinking days (defined as more than 60 g per day for men and more than 40 g per day for women) in the last 28 days, and 14 or fewer abstinent days in the 4 weeks before the screening visit. It heard from the clinical experts that the inclusion criteria reflected the definition in the NICE guideline on alcohol-use disorders for mild alcohol dependence and the World Health Organization's classification of drinking risk levels.The Committee noted that the 2 ESENSE studies excluded people with severe psychiatric conditions or severe medical comorbidities, but noted the company's consultation response explaining that at the UK sites of the SENSE trial, nalmefene was given to patients with stable psychiatric comorbidities and who were taking multiple medications. It also noted that none of the sites in the ESENSE trials was in the UK, and that only 5 sites in the SENSE trial were UK‑based. The Committee was aware that both the company and the Evidence Review Group (ERG) had commented that many people who have alcohol dependence also have medical conditions or psychiatric conditions. The Committee was also aware that the clinical experts agreed with this view. The Committee concluded that the baseline characteristics of the populations in the 3 studies were not wholly generalisable to clinical practice in England, but provided sufficient evidence for clinicians to determine the appropriate patient population for treatment with nalmefene plus psychosocial support, with the psychosocial support focusing on treatment adherence and reducing alcohol consumption.

4.5 The Committee discussed the psychosocial support used both in conjunction with and as a comparator to nalmefene in the ESENSE1, ESENSE2 and SENSE studies. It was aware that the psychosocial support provided in the studies was in the form of BRENDA (see section 3.1), which is not currently used in clinical practice in England, although it is used in clinical trials. The Committee considered if BRENDA, as administered in the clinical trials, is applicable to clinical practice in England. It was aware that the NICE guideline on alcohol-use disorders specifies the type and frequency of psychosocial intervention that should be offered to people with mild alcohol dependence who wish to reduce their alcohol consumption, and that both the intervention and comparator in the final scope issued by NICE specified psychological intervention 'as defined in NICE clinical guideline 115'. The Committee heard from the clinical experts that BRENDA was delivered at different intervals and intensity to both the psychosocial intervention as described in the NICE guideline on alcohol‑use disorders and that used in clinical practice in England. However, it heard from the clinical experts that although BRENDA is not used in its entirety in clinical practice, most of the components within it are currently provided in the form of brief or extended brief interventions and could be administered by healthcare professionals. The Committee accepted that BRENDA, as described in the 3 nalmefene studies, closely resembled current established practice. It concluded that the clinical effectiveness evidence based on the comparison with BRENDA was relevant to clinical practice in England.

4.6 The Committee considered the clinical‑effectiveness results of the 3 nalmefene studies. It agreed that it should only consider the post hoc subgroup analyses carried out on trial patients in the 3 nalmefene studies with a high or very high drinking risk level at baseline who maintained such a level at randomisation because these analyses formed the basis of the licensed population in the marketing authorisation for nalmefene. The Committee was aware that the subgroup analyses had not been pre‑specified but had been performed because 18% (ESENSE1), 33% (ESENSE2) and 25% (SENSE) of patients reduced drinking between screening study visits and randomisation, therefore leaving little scope for additional improvement. The Committee noted the ERG's concerns that the subgroup efficacy data may be less robust because none of the studies were powered for this analysis and initial randomisation may have been lost with the high dropout rate possibly affecting the results. It was also aware that the Scientific Advisory Group to the European Medicines Agency recognised the validity of the subgroup analyses and that these analyses formed the basis of the licensed population in the marketing authorisation for nalmefene. The Committee accepted that the post hoc subgroup analyses were sufficiently robust to use in its decision‑making. It noted that the results from the post hoc subgroup analyses suggested that people in the nalmefene plus BRENDA group had fewer heavy drinking days per month and total alcohol consumption per day compared with those who received placebo plus BRENDA. However, the Committee was concerned that the differences between the treatment groups were relatively small (13% in heavy drinking days and 11% in total alcohol consumption), suggesting that most of the treatment gain from nalmefene could be attributed to the psychosocial support (BRENDA). The Committee heard from the clinical experts that both the number of heavy drinking days and total alcohol consumption are clinically relevant outcome measures and that although the reduction in these outcomes appear modest, they are clinically significant. The Committee concluded that nalmefene plus BRENDA reduces the number of heavy drinking days and total alcohol consumption compared with BRENDA alone, although the exact magnitude of effect was uncertain because of the post hoc subgroup analyses and the trials were not powered for these analyses (see section 3.8).

4.7 The Committee noted that there were no trials directly comparing nalmefene plus psychosocial support with naltrexone plus psychosocial intervention, and the company had not presented an indirect comparison of the 2 treatments. The Committee accepted the rationale provided by the company and that the ERG had agreed it would be inappropriate to carry out an indirect comparison given the limitations of the naltrexone studies identified by the company. The Committee was aware that naltrexone plus psychosocial intervention is recommended in the NICE guideline on alcohol-use disorders (although oral naltrexone does not have a UK marketing authorisation for the same indication as nalmefene, that is for the reduction of alcohol consumption rather than abstinence or relapse prevention) for people whose alcohol dependence did not respond to psychosocial intervention, or those who have specifically requested a pharmacological intervention, and that it was included as a comparator in the final scope issued by NICE. The Committee agreed that the relative effectiveness of nalmefene plus psychosocial support and naltrexone plus psychosocial intervention was uncertain, mainly because of limitations in the available evidence base for naltrexone in people with mild alcohol dependence. The Committee noted that consultation comments from a professional group and patient and carer group suggested a comparison between naltrexone and nalmefene would be helpful, because some patients are being treated with naltrexone in a similar way to the nalmefene licence. It considered whether an indirect comparison should have been carried out (albeit an imperfect one) as it was included in the final scope issued by NICE. The Committee had heard from the clinical experts that naltrexone plus psychosocial intervention was not part of established practice for the reduction of alcohol consumption, and it agreed that naltrexone plus psychosocial intervention could not be considered an appropriate comparator. The Committee concluded that it would not consider further the comparison of nalmefene plus psychosocial support compared with naltrexone plus psychosocial intervention in its decision‑making.

4.8 The Committee considered the health‑related quality of life benefits associated with nalmefene plus BRENDA. The Committee noted that the company had collected health‑related quality of life data as measured by the EQ‑5D and SF‑36 in all 3 nalmefene trials. The Committee was aware that the reference case outlined in NICE's Guide to the methods of technology appraisal 2013 states that EQ‑5D is the preferred measure of health‑related quality of life in adults and concluded that the utility data available from the EQ‑5D was the most appropriate for its decision‑making. The Committee noted that the results from the EQ‑5D analyses (see section 3.11) suggested that nalmefene plus BRENDA improved a person's health‑related quality of life compared with placebo plus BRENDA. The Committee was also aware that it had heard from the patient experts that health‑related quality of life was important and any treatment that could have a positive impact on quality of life was considered valuable (see section 4.1). The Committee agreed that the EQ‑5D data showed that nalmefene plus BRENDA improved health‑related quality of life compared with placebo plus BRENDA.

Cost effectiveness

4.9 The Committee considered the company's economic model and the review and exploratory sensitivity analyses performed by the ERG. It discussed the company's general approach to developing the nalmefene plus psychosocial support economic model. It noted that the ERG considered the company's model to be well structured with most of the assumptions being unfavourable to nalmefene. The ERG commented that the company had not included a half‑cycle correction and that this was a limitation of the model. However, the ERG acknowledged that the impact of a half‑cycle correction in the monthly time cycles was likely to be small. The Committee concluded that the outlined structure of the model adhered to the NICE reference case for economic analysis and was accepted for assessing the cost effectiveness of nalmefene plus psychosocial intervention.

4.10 The Committee considered the company's cost‑effectiveness analyses for comparing nalmefene plus psychosocial support with psychosocial support alone. It noted that the company had provided a base‑case analysis in which the psychosocial support in both the intervention and comparator groups was BRENDA, which was an intervention of lower intensity than that recommended in the NICE guideline on alcohol-use disorders (see section 3.28). The Committee accepted the company's base‑case incremental cost‑effectiveness ratio (ICER) that nalmefene plus psychosocial support dominated psychosocial support alone. The Committee was aware of the ERG's comments that the evaluation carried out in the model does not meet the final scope issued by NICE because the scope stated psychosocial intervention as defined by the NICE guideline on alcohol‑use disorders. The Committee noted that the ERG had formulated 4 comparisons testing the robustness of the cost effectiveness of nalmefene plus psychosocial intervention relevant to the decision problem defined in the scope, that is, psychosocial intervention as defined in the NICE guideline on alcohol‑use disorders (see section 3.37). The Committee discussed which of the ERG's 4 comparisons were most appropriate for its decision‑making. The Committee was aware of its decision to accept that brief or extended brief interventions as the appropriate comparator for nalmefene and that it was satisfied that the psychosocial support used in the nalmefene studies (BRENDA, as part of the intervention and the comparator) closely represented current clinical practice in England. The Committee therefore agreed that the ERG's comparison 1 (which corresponded with the company's base‑case analysis) was the most appropriate analysis.

4.11 The Committee considered the ERG's exploratory amendments in comparison 1. It noted the amendments made by the ERG to the company's base case (see sections 3.38–3.39). The Committee noted that the changes did not include amending the company's assumption that people would remain on treatment (regardless of drinking risk level) for the full year. It discussed whether the company's assumption that patients would remain on treatment for 12 months regardless of drinking level and response was reasonable. The Committee heard from both the clinical experts and the ERG that it is unlikely that GPs would allow a patient to continue treatment and continue drinking at a high drinking risk level for up to 1 year. The Committee understood that the length of treatment time would be decided on an individual basis between the clinician and patient but that 12 months of treatment was possible. The Committee was aware that it was unclear to the ERG if such changes to the duration of treatment would be favourable or unfavourable to nalmefene plus psychosocial support. The ERG had commented that it was highly unlikely to change the cost‑effectiveness results from comparison 1. The Committee considered the 7 exploratory analyses carried out by the ERG and the ERG's exploratory base case, which combined 4 of the ERG'S exploratory analyses: medium‑risk drinkers relapsed to high or very high risk, all of the patients who withdrew for nalmefene‑related responses also withdrew from psychosocial support, the average cost of medically assisted withdrawal was £645 per patient and that the cost of an expert psychosocial support appointment was £119. With these assumptions taken into account, the ERG's exploratory base case indicated that nalmefene plus psychosocial support still dominated psychosocial support alone (that is, was less expensive and more effective). The Committee also discussed that when the ERG presumed no second‑line treatments were available, the incremental cost‑effectiveness ratio (ICER) increased to £5100 cost per quality‑adjusted life year (QALY) gained for nalmefene plus psychosocial support compared with psychosocial support alone. It concluded that based on the analyses provided by the ERG the ICER would lie somewhere between nalmefene plus psychosocial support being dominant and £5100 per QALY gained compared with psychosocial support alone.

4.12 The Committee also discussed whether any other factors should be taken into account when considering the cost effectiveness of nalmefene plus psychosocial support. It noted that adopting a wider perspective than the NHS and personal social services, as included in the remit from the Department of Health, resulted in nalmefene plus psychosocial support still dominating psychosocial support alone. The Committee considered whether the utility values used in the economic model incorporated all the health‑related quality‑of‑life benefits associated with a reduction in alcohol consumption. The Committee was aware that it had heard from patient experts that reducing alcohol consumption was of considerable importance to family members and carers (see section 4.1). The Committee agreed that the utility values used in the economic model may have underestimated the true benefit of nalmefene plus psychosocial support. Although aware of the uncertainty about whether the results from the 3 nalmefene clinical studies are generalisable to patients seen in practice in England (see section 4.4) and the uncertainty associated with the post hoc subgroup analyses (see section 4.6), taking into account the wider societal perspective and the possible underestimation of the utility values, the Committee agreed that the most plausible ICER was likely to be lower than £5100 per QALY gained. The Committee therefore concluded that nalmefene given in conjunction with psychosocial support was a cost‑effective use of NHS resources compared with psychosocial support alone for treating people with alcohol dependence who have a high drinking risk level, without physical withdrawal symptoms and who do not require immediate detoxification.

4.13 The Committee discussed the issue of adherence to nalmefene treatment in clinical practice, given that it should only be prescribed in conjunction with psychosocial support focusing on treatment adherence and reducing alcohol consumption. The Committee was aware that the summary of product characteristics for nalmefene indicates that physicians should continue to assess the patient's progress in reducing alcohol consumption and treatment adherence and that physicians must take this into consideration when prescribing nalmefene plus psychosocial support. The clinical experts commented that although some patients in clinical practice may be less likely to adhere to treatment because of the need to document their drinking level, or to attend their scheduled psychosocial intervention sessions, there are many who would be sufficiently motivated to adhere to all aspects of the treatment. The Committee heard from the clinical experts that patients taking nalmefene would usually be given information to ensure that they understand why adherence to treatment (in terms of when they take their medication, recording of alcohol consumption and attendance at psychosocial support sessions) is important. The Committee concluded that treatment adherence for both nalmefene and psychosocial support is an important consideration for physicians when prescribing treatment.

4.14 The Committee noted the concerns raised during both its meetings and the consultation regarding difficulties that may be encountered complying with the implementation period in which to provide funding for nalmefene. It was aware of the requirement for the relevant health bodies (clinical commissioning groups, NHS England and local authorities) to provide funding to ensure the technology is available within 3 months, from the date the recommendation is published by NICE. The Committee noted that the provision of psychosocial intervention differs throughout England, and the licence for nalmefene mandates that treatment should be given in combination with psychosocial support. The Committee highlighted that it would be reasonable for NICE to reflect on whether the standard 3 month implementation period is appropriate.

4.15 The Committee noted the potential equality issue raised by a patient expert and a Committee member in the meeting that families may be stigmatised for having a family member with alcohol dependence. It also noted the equality issue raised in a clinical expert statement, suggesting that there could be issues with consent of treatment in certain populations in terms of cognitive decline and learning disability. The Committee considered that healthcare professionals should be mindful of the need to ensure equality of access to treatment for patients with disabilities. The Committee concluded that its recommendation on the use of nalmefene plus psychosocial support does not have a particular impact on any group with a protected characteristic in the equality legislation and that there was no need to alter or add to its recommendations.

Summary of Appraisal Committee's key conclusions