Fenfluramine for treating seizures associated with Lennox–Gastaut syndrome in people 2 years and over

Lennox–Gastaut syndrome (LGS) is a severe, lifelong and treatment-resistant form of epilepsy that begins in early childhood, generally before the age of 8 years. It is characterised by a specific electroencephalogram pattern and developmental delay or cognitive impairment. It is also characterised by frequent seizures of different types. Drop seizures result in a loss of muscle tone or stiffening of muscles, which may result in falls, serious injury, pain, hospitalisation and death. Generalised tonic–clonic seizures are particularly severe. Uncontrolled and frequent generalised tonic–clonic seizures correlate to an increased risk of death. Non-drop seizures are typically not as severe as drop seizures and do not generally result in hospitalisation. The patient carer expert noted that LGS can also result in behavioural issues such as hyperactivity, anxiety, aggression, sleep disturbances and depression. They also noted that LGS has a substantial impact on families and carers, with some reporting feelings of despair and helplessness. People with the condition need round-the-clock care, and help with almost all aspects of daily life. Families and carers may find that it prevents them from leading normal lives and prevents family activities. The anxiety that a child with LGS may be injured because of a drop seizure can also significantly affect the mental wellbeing of their family. The patient carer expert explained that they must be within catching distance of their child at all times because their child could have a drop seizure at any moment. The committee concluded that LGS severely affects the quality of life of people with the condition, their families and carers.

The NICE guideline on epilepsies in children, young people and adults (from here referred to as NG217) recommends considering sodium valproate first. If seizures are inadequately controlled, NG217 recommends considering lamotrigine as a second-line add-on treatment or by itself. If second-line treatment is unsuccessful, cannabidiol plus clobazam, clobazam alone, rufinamide and topiramate can be considered as third-line add-on treatment options. If all other treatment options are unsuccessful, add-on treatment with felbamate (unlicensed use) can be considered, under the supervision of a neurologist with expertise in epilepsy. Non-pharmacological treatment options include vagus nerve stimulation, a ketogenic diet and surgery. The clinical experts stated that the NG217 treatment pathway for LGS is broadly reflective of clinical practice in the NHS. They also noted at the third committee meeting that rufinamide is usually tried first out of the available third-line options, and then clobazam alone before trying cannabidiol as an add-on. But, they noted that the choice of treatment regime is highly individualised and based on effectiveness, adverse effects, sedative effects and drug–drug interactions. A consultation comment from the Association of British Neurologists noted that some people would not be able to have clobazam because the sedative effects may aggravate symptoms. The clinical experts explained that not everyone would be able to have cannabidiol plus clobazam because of the side effects of clobazam. The committee noted that it would be useful to see data on the proportion of people who would not have cannabidiol plus clobazam in NHS clinical practice. The company and clinical experts were unable to provide an estimate of the proportion of people who would not have cannabidiol plus clobazam. This was because of the heterogeneity of LGS and the treatment options, and the rarity of the condition. The clinical experts noted that LGS can be difficult to diagnose because not all people display the characteristic symptoms (see section 3.1) at onset or at any one time. By the time people are diagnosed they have often already had most third-line treatment options. They also stated that current treatments often do not control seizures associated with LGS. The patient carer experts noted that the currently available drugs that comprise standard care (SC) that initially work, can lose efficacy. At the third committee meeting, the clinical expert noted that the progression of LGS means that antiseizure medicines may control seizures initially, but then provide inadequate disease control over time. People may try different combinations of treatments to effectively balance seizure control with side effects. This may also include revisiting previous treatments if the current medicines do not adequately control symptoms or the side effects are too severe. The committee concluded that LGS is a heterogenous condition and there is an unmet need for treatments that reduce the number of drop seizures without markedly increasing adverse events.

At the first committee meeting, the company positioned fenfluramine as a third-line add-on therapy, in line with the positioning of cannabidiol plus clobazam, and this was in line with its marketing authorisation. Based on this positioning, the comparator included in the company submission was cannabidiol plus clobazam (plus SC). The company also provided a scenario comparing fenfluramine with SC alone. SC comprised a basket of treatments that included:

• clobazam

• levetiracetam

• valproate

• lamotrigine

• topiramate and

• rufinamide.
  
  The EAG noted that clobazam, rufinamide and topiramate are recommended as third-line treatment options in NG217. So they should also be considered separately as comparators and not just within the basket of treatment options. The company highlighted the refractory nature of LGS and the heterogeneity of the treatment population. It noted that it is therefore not clinically or statistically meaningful to compare fenfluramine plus SC with individual or specific combinations of antiseizure medicines, except cannabidiol plus clobazam plus SC. It added that it believed that cannabidiol plus clobazam plus SC is the only treatment with enough trial data to permit a robust comparison. The company also referenced the NICE technology appraisal guidance on cannabidiol with clobazam for treating seizures associated with Lennox–Gastaut syndrome (from here referred to as TA615). In that appraisal, cannabidiol plus clobazam plus SC was compared with SC alone (referred to as 'current clinical management' in TA615 and defined as a 'basket of choices of antiepileptic drugs'). The committee recalled that the treatment pathway in LGS, particularly after second-line treatment, can be heterogeneous (see section 3.2). The committee considered that it would be helpful to have seen scenarios that considered clobazam, rufinamide and topiramate as separate comparators, if data was available. It added that data about the proportion of people with LGS using those treatments in the NHS would also be helpful. The company stated that it was unable to provide an estimate of the proportion of people using clobazam, rufinamide and topiramate in the NHS because of the heterogeneity of LGS and the treatment options. It noted that these treatments are considered within the basket of treatments in SC in the placebo arm of Study 1601. It added that the healthcare professionals who were consulted considered the proportions of these treatments in SC in the placebo arm of Study 1601 to be reflective of clinical practice. The committee acknowledged that most of the studies where these treatments are considered separately, rather than in a basket of antiepileptic drugs, were conducted over 20 years ago. So, they do not reflect current clinical practice (see section 3.6). Because of this and the heterogeneity in the treatment population, it accepted that any comparisons where these treatments are considered separately may not be robust and clinically meaningful.

At the first committee meeting, the committee concluded that the positioning of fenfluramine plus SC in the treatment pathway in line with cannabidiol plus clobazam plus SC was appropriate. It also concluded that cannabidiol plus clobazam plus SC is an appropriate comparator and that it was appropriate to define SC alone as the basket of treatments outlined in section 3.3. In response to the draft guidance, the company stated that it did not consider SC alone to be an appropriate comparator. This is because data for the SC alone arm is only available for 3 months and so extrapolation beyond this relies on assumptions. It noted that the heterogenous treatment pathway results in various SC drugs being used, with varying costs and efficacy. It added that, because of this, comparison with SC alone is much more uncertain than the comparison with cannabidiol plus clobazam plus SC. At the appeal panel meeting after the second committee meeting, clinical experts noted that there are some people for whom cannabidiol plus clobazam is unsuitable or ineffective. These people would instead have some other combination of treatments. But the clinical experts noted that, because of the highly heterogenous nature of the disease, these treatments are also very heterogeneous. So, the clinical experts at the appeal panel meeting considered that SC alone was not a relevant comparator for fenfluramine plus SC. The majority of the consultation responses to the second draft guidance stated that SC alone was not a relevant comparator in clinical practice. They noted that the treatments listed in NG217 were included because of availability of evidence, not because of clinical utility or use in practice. They also noted that if people could not have cannabidiol plus clobazam, they would have non-pharmacological therapies, such as vagus nerve stimulation, or treatments in clinical trials (see section 3.2). At the third committee meeting, the clinical expert explained that the suitability of treatments can vary at different times in the same person. This is because of changes in the LGS disease course and the impact of side effects. They explained that people may revisit previous treatments if they are more effective at controlling current symptoms (see section 3.2). The committee noted that most people who were eligible for fenfluramine plus SC would have cannabidiol plus clobazam plus SC if fenfluramine was not available. But, it considered that fenfluramine would be suitable for some people who cannot have clobazam (see section 3.2) and so cannot have cannabidiol plus clobazam. So, it agreed that SC alone is a relevant comparator for people who cannot have clobazam. The committee noted the heterogeneity of LGS and the high unmet need for treatments for the condition (see section 3.2). It also noted that it is difficult to define the population that clobazam is unsuitable for, because of changes in treatment suitability over time. So, it did not consider it appropriate to restrict the population eligible for fenfluramine to people who could have clobazam. This is because people who could not have clobazam would benefit from the availability of fenfluramine. The committee understood that the data for SC from Study 1601 was only available for 3 months and that the extrapolation from short-term studies is inherently uncertain. It also considered that SC as a comparator is not well defined and that treatment options at this point in the treatment pathway are heterogeneous because of heterogeneity in the LGS population. The committee concluded that there were people who would not have cannabidiol plus clobazam who are having SC in current clinical practice, and would be eligible for fenfluramine. So, SC is a relevant comparator for fenfluramine. But the committee acknowledged that there is substantial heterogeneity in SC and substantial uncertainty in the efficacy assumptions for SC.

The primary clinical evidence for fenfluramine plus SC came from Study 1601 and an interim analysis of the Study 1601 open-label extension (OLE) study. Study 1601 was a phase 3, double-blind, international randomised controlled trial (RCT). It compared the efficacy and safety of fenfluramine 0.2 mg/kg/day (n=89) and fenfluramine 0.7 mg/kg/day (n=87) as an add-on therapy to SC, with placebo plus SC (n=87). The trial period was 20 weeks. It recruited people aged between 2 and 35 years, with Epilepsy Study Consortium-confirmed LGS diagnoses, on stable antiseizure medicine regimens. The EAG noted that the final scope outcomes included seizure frequency (overall and by seizure type) and seizure severity. But, it noted that the company reported seizure frequency for only drop seizures and seizure severity was not collected in the trial. The primary outcome was percentage reduction from baseline in drop-seizure frequency (DSF) per 28 days in the fenfluramine 0.7 mg/kg/day arm. At week 14 of the titration and maintenance period, the median percentage change from baseline in DSF was a 26.5% reduction in the fenfluramine 0.7 mg/kg/day arm. This was compared with a 7.6% reduction in the placebo arm (p=0.001). At week 14, the proportion of people with a reduction in DSF of 50% or more was 25.3% in the fenfluramine 0.7 mg/kg/day arm and 10.3% in the placebo arm (p=0.015). Study 1601 OLE (n=247) is an ongoing flexible-dose, single-arm study to assess the safety and efficacy of fenfluramine plus SC for people who completed Study 1601. All people were initially started on 0.2 mg/kg/day fenfluramine and after 1 month were titrated by effectiveness and tolerability, which were assessed at 3-month intervals. At the latest data cut, 142 people had completed 12 months of follow up. At year 1 of the OLE, the median percentage reduction in DSF from baseline was 51.8% (p<0.0001). The committee concluded that fenfluramine as an add-on to SC is more effective at reducing DSF than SC alone. The committee also noted the adverse events reported in Study 1601 (available in the summary of product characteristics [SPC] for fenfluramine). It acknowledged that the most common treatment-emergent adverse events were decreased appetite, drowsiness and fatigue, which occurred at a higher rate in the fenfluramine 0.7 mg/kg/day arm than in the fenfluramine 0.2 mg/kg/day arm.

Because there was no direct head-to-head evidence for fenfluramine plus SC compared with cannabidiol plus clobazam plus SC, the company did a series of network meta-analyses (NMAs). Outcomes captured between 10- and 20-week timepoints were considered. Outcomes assessed were:

• median percentage reduction in frequency of generalised tonic–clonic seizures

• reductions in DSF of:

  • 25% or more

  • 50% or more

  • 75% or more

• discontinuation due to adverse events.
  
  After the company's systematic literature review and feasibility assessment, 3 RCTs were identified (covering fenfluramine, cannabidiol and placebo only). The company did an NMA with these 3 RCTs, each with intention-to-treat (ITT) populations, referred to as the 'ITT data NMA'. But not everyone in the RCT for cannabidiol was also having clobazam. So, the company performed an additional NMA analysis using cannabidiol plus clobazam subgroup data, based on data published by the German health technology assessment body, the GBA (The Federal Joint Committee). This was referred to as the 'GBA data NMA'. The GBA data did not include sufficient data on the median reduction in frequency of generalised tonic–clonic seizures or the discontinuation due to adverse events. So, the ITT data NMA was used for these outcomes. Together, the ITT data NMA and the GBA data NMA formed the company's base-case NMA. The company stated that its base-case NMA point-estimate results at 14 weeks suggest that fenfluramine plus SC is most likely to be superior to placebo plus SC and cannabidiol plus clobazam plus SC for all outcomes assessed, except the 75% or more reduction in DSF. But, the committee noted that for all outcomes the credible intervals for fenfluramine plus SC and cannabidiol plus SC overlap. So, it is uncertain whether fenfluramine plus SC is superior to cannabidiol plus clobazam plus SC. The exact credible intervals are considered confidential by the company and cannot be reported here. The EAG disagreed with the exclusion following the feasibility assessment of 6 RCTs that included rufinamide, lamotrigine, clobazam and topiramate. It noted that rufinamide, topiramate and clobazam are recommended for consideration as third-line treatments in NG217. The company's rationale for the exclusion was that the 6 RCTs did not report all outcomes of interest or all key patient characteristics. It added that most of the excluded studies included data that was 20 to 30 years old and so do not capture improvement in LGS treatment. In its initial submission, the company considered that cannabidiol plus clobazam plus SC was the only relevant comparator (see section 3.3). It also provided a scenario analysis where SC alone was included as a comparator. Results from the NMA that comprised the 9 RCTs in the network suggested that, overall, some clinical benefits of some other third-line antiseizure medicines used as monotherapies may be numerically superior to those of fenfluramine. The committee acknowledged the challenges of robust data collection in people with LGS (see section 3.3). The committee concluded that the comparative clinical effectiveness of fenfluramine plus SC and cannabidiol plus clobazam plus SC is uncertain. This was because of the mixed direction of results for the efficacy outcomes assessed at 14 weeks (titration and maintenance phase) and the overlapping credible intervals. It also noted the lack of robust data for rufinamide, topiramate and clobazam. So, the results of the indirect comparisons including these comparators as monotherapies were very uncertain.

At the third committee meeting, the company revised its approach to assume that fenfluramine and cannabidiol plus clobazam were equally effective. The company said that this approach was supported by the results of a survey of healthcare professionals (n=14). In the survey,93% of respondents agreed that the assumption of equal efficacy between fenfluramine and cannabidiol plus clobazam was appropriate. The company highlighted that although the results from their indirect treatment comparison favoured fenfluramine (see section 3.8) the credible intervals overlap and include 1. So it suggested that equal clinical efficacy between fenfluramine and cannabidiol plus clobazam could not be ruled out. The EAG maintained its approach from the second committee meeting of using data from the treated population of the OLEs for fenfluramine and cannabidiol to estimate clinical efficacy. It explained that the company's revised approach does not account for observed clinical differences between fenfluramine and cannabidiol plus clobazam shown in the clinical trial data. But it also acknowledged that there were limitations to both the naive comparison and OLE NMA approaches, so the most appropriate approach was uncertain. The committee recalled that the results of the naive comparison favoured cannabidiol plus clobazam plus SC, while the NMA favoured fenfluramine plus SC when using either the LOCF or non-random imputation approaches. The committee acknowledged the limitations of the available evidence. It considered that, based on the available evidence, it was not possible to confirm a difference in clinical efficacy between fenfluramine and cannabidiol plus clobazam. It noted that any differences in clinical effectiveness between fenfluramine and cannabidiol plus clobazam were likely to be very small. So it concluded that using the assumption of equal efficacy between fenfluramine and cannabidiol was suitable for decision making.

In the first and second committee meetings, the company presented a 6-state cohort-based Markov model with a lifetime time horizon of 86 years. The model compared fenfluramine plus SC with cannabidiol plus clobazam plus SC and SC alone. The model included 4 health states based on percentage reduction in DSF from baseline:

• state 0, for people with a less than 25% decrease in DSF

• state 1, for people with a 25% to less than 50% decrease in DSF

• state 2, for people with a 50% to less than 75% decrease in DSF

• state 3, for people with a 75% or greater decrease in DSF.
  
  It also included an additional health state for people who discontinued treatment and an absorbing death state. In the model, there were 3 main phases:

• titration and maintenance

• treatment and

• subsequent follow up.
  
  The titration and maintenance phase was modelled for a duration of 2 weeks (titration) and 3 months (maintenance). State occupancy was based on drop-seizure distribution at baseline in Study 1601. The model assumed that people remain in these health states during the titration and maintenance phase unless they either discontinue because of adverse events or die. After the titration and maintenance phase, people moved to the corresponding health state based on the efficacy data from the RCT NMA (see section 3.6). The model cycles lasted 3 months. For the SC arm, it was assumed that there was no change in state occupancy from cycle 2 onwards, except for people who die. Data informing state occupancies varied from cycles 2 to 9 for fenfluramine plus SC and cannabidiol plus clobazam plus SC (see section 3.14 and section 3.15). After cycle 9, the change in state occupancy was based on treatment waning, discontinuation and death (see section 3.16).

The EAG highlighted concerns with basing health states on the relative reductions in drop seizures. It noted that this results in people with different numbers of absolute drop seizures being in the same health state, despite having significant differences in health-related quality of life (HRQoL) and costs. It added that this model structure based on relative reduction in drop seizures deviated from other published models in LGS and from the model used in TA615. So, it would prefer a model based on absolute reduction in drop seizures. The company stated that a modelling approach using absolute reductions in drop seizures was not feasible because of the lack of absolute trial data for cannabidiol plus clobazam plus SC. It also highlighted that in its model, relative reduction in the percentage of DSF was translated to absolute DSF using the midpoint approach in Neuberger et al. (2020). This allowed the incorporation of healthcare resource use data from TA615, based on absolute DSF categories. The committee noted the very large interquartile ranges for the baseline median DSF in Study 1601 (2 to 1,761 and 7 to 1,803 for placebo and fenfluramine 0.7 mg/kg/day, respectively). It questioned the plausibility of using a relative approach, given the large difference in the absolute number of drop seizures for the treatment population. The committee noted that, as a result, it is highly uncertain to assume people in the same relative reduction in DSF health state have the same utility values and healthcare resource use. It considered that a model based on absolute reduction in DSF would be more robust. But it acknowledged other limitations that would have been present with a model with health states based on absolute DSF categories. So, although the committee had significant reservations about the appropriateness of the model structure, it agreed to use it to inform its decision making. It concluded that the model structure added uncertainty to the cost-effectiveness estimates.

The committee noted that the model only included drop seizures, and so did not include the impact of fenfluramine on other seizure types. It noted that it was unclear whether the exclusion of non-drop seizures from the model would favour fenfluramine or the comparators. It recognised that reducing non-drop seizures is important to people with LGS and their carers. But it understood that non-drop seizures are harder to measure and verify than drop seizures. It concluded that the absence of non-drop seizures in the model adds to the uncertainty around the economic analysis.

The company's model had a lifetime time horizon of 86 years. But, only 15 months of data for fenfluramine plus SC was available from Study 1601 and the OLE. So, extrapolation of treatment effect was needed beyond the trial period. For fenfluramine plus SC, the company's initial modelling approach assumed that the transition probabilities for cycles 6 to 9 equalled the transition probabilities of cycles 4 to 5, which were based on the last 3 months of the Study 1601 OLE. That is, it was assumed that the treatment effect for fenfluramine plus SC increased after the observed trial period. In contrast, the company assumed the treatment effect for cannabidiol plus clobazam plus SC was stable for cycles 6 to 9. This assumption was based on the experience of healthcare professionals using fenfluramine to treat Dravet syndrome and state occupancy data of fenfluramine and cannabidiol from the respective OLE studies. The company stated that the data suggested that the treatment effect of fenfluramine is sustained and increases, based on increasing percentages of people showing improvement in DSF reduction over time. The EAG highlighted that in NICE's technology appraisal guidance on fenfluramine for treating seizures associated with Dravet syndrome (from here referred to as TA808), a maintained treatment effect of fenfluramine was modelled based on the efficacy data. The EAG preferred to model a maintained treatment effect for fenfluramine plus SC treatment during cycle 6 to cycle 9 in its base case (in line with the assumed maintained treatment effect for cannabidiol plus clobazam plus SC). After the first committee meeting, the company updated its base case to assume a maintained treatment effect for cycles 6 to 9 for fenfluramine plus SC and cannabidiol plus clobazam plus SC. The committee noted the lack of robust evidence for an increased treatment effect for fenfluramine. So it considered it appropriate to assume a maintained treatment effect for cycles 6 to 9 for both fenfluramine plus SC and cannabidiol plus clobazam plus SC.

At the first and second committee meetings, from cycle 10 onwards in the model, people in the fenfluramine plus SC and cannabidiol plus clobazam plus SC arms stayed in the same health state unless they experienced treatment waning, stopped treatment, or died. The company calculated the proportion of people experiencing treatment waning by:

• taking the proportion of people stopping because of lack of efficacy in the last 3 months of the Study 1601 OLE, which was 5.2%

• multiplying this proportion with the deteriorating transition probabilities based on all people on treatment from last 3 months of the Study 1601 OLE.
  
  This was applied to both the fenfluramine plus SC arm and cannabidiol plus clobazam plus SC arm because of a lack of treatment waning data for the latter. The EAG explained that applying this to the health-state occupancies in cycle 10 resulted in only 0.58% and 0.48% of people moving to a worse health state for fenfluramine plus SC and cannabidiol plus clobazam plus SC, respectively. The EAG considered that this was extremely low. The committee noted that in TA615 the company assumed that people on cannabidiol stayed in the same health state from cycle 10 onwards (27 months) unless they stopped treatment or died. The company in that appraisal did a scenario analysis where 10% of people in all health states (except the seizure-free health state) stopped cannabidiol. The committee in TA615 concluded that this scenario captured some, but not all, of the treatment effect diminishing over time. At the second committee meeting, the committee agreed that it was reasonable to use deteriorating transition probabilities based on the last 3 months of the Study 1601 OLE to model treatment waning for cycle 10 onwards. But, it considered that the way treatment waning had been applied in the company's base case (where these deteriorating transition probabilities were only applied to 5.2% of people each cycle) underestimated the level of treatment waning that was likely to occur in clinical practice. At the appeal panel meeting after the second committee, clinical experts stated that treatment waning is not seen in clinical practice in people having antiepileptic treatments. So, the evidence from that last 3 months of the Study 1601 OLE may not be applicable to clinical practice. At the third committee meeting, the company did not include treatment waning in the model for either fenfluramine plus SC or cannabidiol plus clobazam plus SC. The EAG preferred to maintain its previous approach, which was to apply transition probabilities from the last 3 months of the Study 1601 OLE to 5.2% of people on treatment in the model. 5.2% was the proportion of people stopping because of lack of efficacy in the last 3 months of the Study 1601 OLE. The committee noted that there was no data provided to support a lack of treatment waning in either fenfluramine or cannabidiol. But the responses to the second draft guidance consultation and the clinical expert at the third committee meeting said that, in practice, treatment waning for fenfluramine and cannabidiol was rare or not seen. The clinical expert stated that some people with LGS may experience a 'honeymoon period' when they start a new treatment. That is, an initial period of temporary partial remission of symptoms at the start of a new treatment. But they added that people who experience a durable response to a treatment are unlikely to then experience treatment waning. They explained that changes in the frequency of seizures or other symptoms are likely caused by a change in the LGS disease course, rather than because of reduction in efficacy over time. The committee considered that it was plausible that there was no treatment waning based on the clinical and patient expert opinion.

The company collected data from responses to the Quality of Life in Childhood Epilepsy-16 item questionnaire (QOLCE‑16) in Study 1601 and the OLE. But it did not use the data in its model. It stated that the QOLCE‑16 is a disease-specific measure and that long-term data was not yet available. The company used EQ‑5D utility values from Verdian et al. (2008), a vignette-based conference abstract, to inform patient utility values. It chose this because it matched NICE's EQ‑5D reporting requirements, had been used previously in LGS models and aligned with the model's relative health-state structure. The company also considered 2 other studies reporting relevant utility values (Auvin et al. 2021 and Lo et al. 2021) but these were deemed less appropriate. Auvin et al. examined various types of epilepsies, including Dravet syndrome, which did not align with the patient population. Lo et al. did not align with the model's structure because it reported utilities for health states based on the total number of drop seizures per month. The EAG noted that the vignette approach used by Verdian et al. is condition-orientated and so may not capture all aspects that influence dimensions of the EQ‑5D. Also, the values are not directly from people living with LGS. The company highlighted that vignette-based utility values may be useful in rare conditions such as LGS, where it is not possible to recruit a large enough representative sample. The EAG also considered the utility values to be relatively low and lack face validity when compared with the mean baseline QOLCE‑16 scores from Study 1601. Also, it noted that the overall quality-of-life domain and most other domains of the QOLCE‑16 showed hardly any clinically relevant change at visit 12 (end of study or end of treatment) compared with baseline. This indicates that the HRQoL of people with LGS may not be very sensitive to improvements in DSF. So, it considered that the large differences in utility values between the health states in the model seemed to lack face validity. The EAG used the Verdian et al. utility values in its base case, but considered that none of the sources of utility values in the company submission were ideal for informing HRQoL for people with LGS. The committee considered that all utility values presented in the company submission were associated with limitations. But, it recognised the challenges associated with obtaining robust utility values in rare conditions such as LGS. The committee concluded that the Verdian et al. utility values are associated with substantial uncertainty. But, they are likely the best available source of utility values given the use of health states based on relative reductions in drop seizures.

The committee recalled that caring for someone with LGS has a substantial impact on carers' quality of life (see section 3.1). It considered that capturing this in the model is appropriate. The company included carer utilities for each health state in its base case by applying the same utility values from Verdian et al. (2008) used for people with LGS (see section 3.17). The company assumed 1.8 carers per person with LGS. The company assumed that the utility value of carers equalled that of people with LGS. This was because of a lack of LGS carer utility values in the literature and the substantial impact of LGS on carers who provide round-the-clock care. The EAG considered this assumption to be unrealistic. It highlighted that Auvin et al. (2021) and Lo et al. (2021) reported higher utility values for carers compared with people with LGS. It also noted that the Zarit Caregiver Burden Inventory results in Study 1601 suggested a mild to moderate carer burden and that carer burden may not be sensitive to changes in seizure frequency. The company's carer utility approach also meant that when a person with LGS in the model died, the corresponding carer utility value is set to 0. This overestimates this impact of mortality, given that the carer does not die together with the person they care for. The company also provided a scenario analysis in which carer disutility values were used (instead of utility values). The disutility values were obtained by calculating the difference between the visual analogue scale utility value for the UK general population and the UK carer utility scores for LGS estimated in Auvin et al. The resulting disutility value was then used to calculate a decrement applied to the QALYs for each treatment. Given the limitations with the carer utility approach, the EAG preferred to use the carer disutility approach in its base case. But, the EAG preferred to use disutility values calculated from Lo et al. in its base case (rather than Auvin et al.). This was because it considered that:

• the time trade-off approach from Lo et al. is better aligned with the NICE reference case (stating that a choice-based method should be used) than the visual analogue scale approach used by Auvin et al.

• the sample size of Lo et al. (n=150) was larger than the sample size of Auvin et al. (n=30)

• the DSF categories in Lo et al. better aligned with the DSF categories in the model compared with the DSF categories Auvin et al.
  
  The committee considered that the responsibility for carers was substantial but would expect that the HRQoL for people living with the condition themselves to be lower than carers. So, it considered the company's assumption of equal utility values for patients and carers to be unrealistic and preferred to use carer utility values from Lo et al. The committee noted the limitations with applying carer utility values, rather than disutility values. But, it noted that the EAG's application of the disutility approach resulted in negative total QALYs for all treatments. It considered that this lacked face validity given that no person or carer in the model is assumed to experience negative utility. The company stated that negative QALYs were inherent to the disutility approach in this case, considering that people with LGS have very low QALYs and require more than 1 carer. It clarified that the QALY changes are spread across the patients and applied to an average of 2 caregivers, and that they do not represent a worse-than-death outcome for anyone in the cohort. The committee acknowledged the company's rationale for negative QALYs with the carer disutility approach and considered this was appropriate in this case. The committee concluded that it preferred to use the carer disutility approach in its base case, using disutility values calculated from Lo et al.

The SPC for fenfluramine recommends increasing the dose of fenfluramine as tolerated up to the recommended maintenance dosage of 0.7 mg/kg/day. The company implemented a base-case maintenance dosage for fenfluramine of 0.413 mg/kg/day. It stated the dosage was based on the mean daily dosage for fenfluramine for people in the Study 1601 OLE, in which efficacy continued to improve at lower average doses than used in Study 1601. This mean daily dosage excluded people who had dosages of more than 0.7 mg/kg/day (maximum licensed dosage) in the OLE. The company considered that the OLE dosage is more reflective of clinical practice than those in Study 1601 because dosages were titrated based on safety and tolerability in the OLE. It also suggested that the dosage was comparable to the average dosage of people with Dravet syndrome who are not on stiripentol. The EAG agreed that in clinical practice, dosages will be titrated based on tolerability, efficacy and safety. It noted that the mean daily dosage was lower than the maintenance dosage recommended in the SPC (that is, 0.7 mg/kg/day). The dosage also differed from the dosages that people had in Study 1601 (see section 3.5), which was used to inform the indirect treatment comparison. The EAG disagreed with the company's rationale for excluding people who had mean dosages of more than 0.7 mg/kg/day in the OLE from the calculation. It noted that people with a mean daily dose lower than the initial titration dosage (0.2 mg/kg/day) were included in company's calculation. And that people who had more than 0.7 mg/kg/day were included in clinical-effectiveness data used in the model. So the EAG preferred using the mean daily dosage for fenfluramine for all people in the Study 1601 OLE (including those who had more than 0.7 mg/kg/day), which was 0.416 mg/kg/day. The committee concluded that it preferred to use the mean dose from the Study 1601 OLE as this dose is likely to be most reflective of clinical practice. It agreed with the EAG's rationale that the maintenance dosage calculation should include people that had mean dosages of more than 0.7 mg/kg/day. So, the committee preferred to include a mean daily dosage for fenfluramine of 0.416 mg/kg/day.

The SPC for cannabidiol states that the dosage can be increased from a maintenance dosage of 10 mg/kg/day to 20 mg/kg/day. At the third committee meeting, the company assumed a base-case maintenance dosage for cannabidiol of 16 mg/kg/day. The company considered that 16 mg/kg/day is conservative based on clinical expert opinion and the cannabidiol OLE study. It highlighted that the mean modal dosage within the cannabidiol OLE was 24 mg/kg/day. The EAG noted that an average dosage of 12 mg/kg/day was used in TA615. It highlighted that the company also used the same data to model cannabidiol efficacy as that used in TA615. The EAG modelled 3 base cases with average maintenance dosages of 12 mg/kg/day, 14 mg/kg/day and 16 mg/kg/day for cannabidiol. But it noted that this was uncertain and considered that the range between 12 mg/kg/day and 16 mg/kg/day should be considered for decision making. The clinical experts stated that in their experience the average maintenance dosage of cannabidiol was around 12 mg/kg/day to 15 mg/kg/day. But the committee noted that cannabidiol RCTs demonstrated a statistically significant reduction in the number of drop and non-drop seizures at 10 mg/kg/day. Based on clinical expert opinion and data from the cannabidiol OLE, the committee considered that it was appropriate to consider a range of cannabidiol maintenance dosages between 12 mg/kg/day and 16 mg/kg/day. For decision making,14 mg/kg/day was used as the midpoint of this range.

At the first committee meeting, clinical experts stated that there may be treatment wastage caused by bottle breakages or leftover liquid medicine in the bottle. Because the company's and EAG's initial analyses all assumed no wastage, the committee requested scenarios accounting for the expected wastage costs associated with both cannabidiol and fenfluramine. The company provided scenarios in which it assumed:

• 5% wastage for both treatment arms

• 5% wastage for fenfluramine and 10% wastage for cannabidiol

• 0% wastage for fenfluramine and 10% wastage for cannabidiol.
  
  The EAG noted that the assumed wastage percentages provided by company were not justified and so it was uncertain whether any of the scenarios were reflective of clinical practice. A patient carer expert stated that wastage of liquid treatments for LGS is often caused by the person having the treatment knocking it out of a carer's hand, which is not specific to the drug used. The clinical experts stated that some drug wastage does happen for both fenfluramine and cannabidiol, but that this is relatively small. A clinical expert estimated that they would typically lose 1 bottle of cannabidiol per year due to accidents or breakages, in their cohort of 45 adults. The committee noted that cannabidiol is an oily substance that is provided in glass bottles and that fenfluramine is a liquid that is provided in plastic bottles. So, it considered that there may be more treatment wastage of cannabidiol than of fenfluramine. At the third committee meeting, the company presented different treatment-wastage values for cannabidiol and fenfluramine based on the results of a survey of healthcare professionals (these wastage values are confidential and cannot be reported here). The company preferred to use the mean wastage values based on the survey, while the EAG preferred to use the median results. The committee acknowledged that there may be small differences in wastage between cannabidiol and fenfluramine. It considered that, based on the available evidence, it was appropriate to use treatment-wastage values from the survey. It preferred to use the mean wastage values. But it noted that the differences in the company's and the EAG's preferred approaches for applying treatment wastage were minimal and only had a small impact on the cost-effectiveness results.

In its submission the company stated that most people will need residential care. The company did not include the impact of residential care in its initial base-case model but provided a scenario analysis including residential-care costs applied to 10% of people who reach age 18. This approach was similar to that used in TA615. In that appraisal, 10% of people experiencing seizures were assumed to need residential care by the time they were 18 compared with 2% for people who had no drop seizures. The EAG preferred to include the cost of residential care in its base case. It used the residential-care rate of 10% provided by the company, but noted that it was uncertain whether this figure was representative of NHS clinical practice. The EAG also considered that the impact of residential care on carer HRQoL should be modelled. In its base case it assumed that people who need residential care will need 0.7 carers (rather than 1.8). This was calculated based on the proportion of days per year that people who need residential care are expected to be at home. The patient carer experts explained that they would expect that most carers would prefer to look after people with LGS themselves rather than opting for residential care. The committee considered that some carers may not be able to provide adequate care because of their own health and so residential care may be the only option. The committee concluded that it was appropriate to assume 10% of people with LGS reaching 18 years old will need residential care. It also concluded that it was appropriate to include residential-care costs and to assume 0.7 carers for people needing residential care, to account for the reduced carer responsibility. The company updated its base-case model after the first committee meeting to align with the committee's preferences for residential care.

The marketing authorisation for fenfluramine does not specify a stopping rule. But the company initially proposed a stopping rule whereby treatment is stopped if DSF has not reduced by at least 25% from baseline, assessed every 3 months. The EAG noted that in TA808, the committee recommended a stopping rule for people who had less than 30% reduction in DSF over a period of 6 months. This stopping rule was also in line with current practice for cannabidiol plus clobazam in LGS. At the clarification stage, healthcare professionals consulted by the company considered it reasonable to stop treatment if the reduction in DSF was less than 25% to 30%. They also agreed it would be reasonable to assess outcomes every 6 months. The EAG preferred to apply the stopping rule applied in TA808. But, it noted that the stopping rule at 6 months appeared to be incorrectly implemented in the model. It explained that all people from health state 0 stopped treatment every 6 months, instead of only the people who were in health state 0 for 6 months. As a result, people who were in health state 0 for only 3 months also stopped treatment. In response to the draft guidance, the company stated that tracking people in the model in the cannabidiol plus clobazam plus SC arm would not be possible without transition probabilities, because patient-level data would be needed. The company implemented a revised stopping rule but the EAG considered this also had limitations and so preferred the company's initial approach. The committee concluded a stopping rule whereby fenfluramine is stopped if the DSF has not reduced by at least 30% from baseline, assessed every 6 months is reasonable. The company updated its base case after the first committee meeting to align with the committee's preferred stopping rule, whereby fenfluramine is stopped if the DSF has not reduced by at least 30% from baseline, assessed every 6 months.

There were no cases of pulmonary arterial hypertension or valvular heart disease reported at any point in Study 1601 and its OLE. But, the committee were aware of a previous study by Souza et al. (2008). In that study, which analysed a cohort of fenfluramine-associated pulmonary hypertension cases, there was a median of 4.5 years between exposure and onset of symptoms. The committee questioned whether pulmonary arterial hypertension could be a cumulative dose-related adverse event and could potentially be an issue after using fenfluramine for more than 5 years. It considered whether the cost of treating pulmonary hypertension should be included in the model. The company highlighted that fenfluramine, when previously used as a weight-loss medicine, was prescribed at 60 mg/day, with dosages as high as 220 mg/day. And the association with heart disease was complicated by the lack of pretreatment echocardiograms and consideration of other risk factors. In contrast, the maximum daily dose of fenfluramine for LGS is 26 mg. The company explained that, as of August 2024, over 8,000 people have had fenfluramine. Pulmonary arterial hypertension was reported in 1 child having fenfluramine (at a dosage of 10.12 mg/day) for Dravet syndrome, which is described in the SPC for fenfluramine. When the child stopped taking fenfluramine, the reaction resolved. The company noted the Medicines and Healthcare products Regulatory Agency's controlled access programme. It states that people must have an echocardiogram every 6 months for the first 2 years on fenfluramine and annually thereafter. If an abnormality is detected, then fenfluramine would be stopped. The committee concluded that, based on the latest available data, it is appropriate not to model the cost of treatment for pulmonary arterial hypertension.

NICE's manual for health technology evaluations notes that judgements about the acceptability of a technology as an effective use of NHS resources will take into account the degree of certainty around the incremental cost-effectiveness ratio (ICER). The committee will be more cautious about recommending a technology if it is less certain about the cost-effectiveness results. The committee highlighted the high level of uncertainty, specifically in:

• including SC alone as a comparator, given the substantial heterogeneity in SC and substantial uncertainty in the efficacy assumptions for SC (see section 3.4)

• the lack of clinical-effectiveness data for fenfluramine on seizure severity and frequency of seizure types other than drop seizures (see section 3.5)

• the comparative clinical effectiveness of fenfluramine plus SC and cannabidiol plus clobazam plus SC (see section 3.6)

• the appropriateness of only using drop seizures in the modelling, and not other seizure types (see section 3.12)

• the appropriateness of assuming no treatment waning for fenfluramine plus SC and cannabidiol plus clobazam plus SC (see section 3.16)

• the appropriateness of the patient utility values presented in the company submission (see section 3.17)

• the cannabidiol maintenance dose used in clinical practice (see section 3.20)

• the appropriateness of basing treatment-wastage assumptions on the results of the company's survey of healthcare professionals (see section 3.21).
  
  The committee considered how the nature of LGS affected the ability to generate high-quality evidence for an evaluation. It noted that LGS is a rare and heterogenous condition that impacts children. This is likely to have affected the available data to model SC, the clinical efficacy and utility data. The committee also took into account the high unmet need for treatments in LGS (see section 3.2) and the severity of the condition (see section 3.22) in its decision making.

In response to the second draft guidance, the company updated its base case to assume equivalent clinical efficacy for fenfluramine plus SC and cannabidiol plus clobazam plus SC (see section 3.9). It only presented results compared with cannabidiol plus SC with this cost comparison approach, and did not present results for comparisons with SC alone. Because of confidential commercial arrangements for fenfluramine, the comparators and other treatments in the model, the exact cost-effectiveness estimates are confidential and cannot be reported here. In the company's deterministic base case fenfluramine plus SC was cost saving when compared with cannabidiol plus clobazam plus SC. The EAG preferred to model state occupancies for cycles 2 to 5 for fenfluramine plus SC and cannabidiol plus clobazam plus SC based on the treated population in the fenfluramine and cannabidiol OLEs, respectively. In its deterministic base case with a cannabidiol maintenance dosage of 14 mg/kg/day, fenfluramine plus SC was cost effective against cannabidiol plus clobazam plus SC with an ICER in the south west quadrant (indicating that fenfluramine plus SC is cheaper and less effective than the cannabidiol plus clobazam plus SC). The EAG's base case results for the comparison with SC alone were higher than the range normally considered an acceptable use of NHS resources.

The clinical experts highlighted that people with LGS have learning disabilities, so support is needed at appointments. A clinical expert also considered fenfluramine treatment will be started by specialists. But, because adults with LGS may not be under the care of a specialist, they may not have access to new treatments. A patient carer expert noted that some of the tests potentially needed to start fenfluramine may be traumatic for people with LGS who have sensory issues. The committee was aware of the need for equitable access to fenfluramine if it is recommended, but noted that access to treatments is an implementation issue that cannot be addressed by a technology appraisal recommendation. It was also aware of monitoring requirements for fenfluramine and noted that these should be considered before starting fenfluramine. Following the second draft guidance consultation, a response from the Epilepsy Society highlighted that fenfluramine is available in Dravet syndrome, but the committee agreed that a treatment being recommended in a different indication was not an equality issue.

The committee also considered potential benefits of fenfluramine that were not included in the economic model. The company stated that there are a number of benefits not captured in the economic model, such as:

• reductions in:

  • duration of drop and non-drop seizures

  • losses to work productivity, which may also provide wider societal benefit

• improvements in:

  • the quality of life of siblings and other family members of people with LGS

  • the intellectual development of children with LGS, due to fewer seizures

  • motor function and

  • executive function.
    
    The committee considered that it was unclear whether including these in the model would favour fenfluramine or the comparators. It acknowledged that there may be potential benefits of fenfluramine that were not captured in the economic model, some of which may also apply to cannabidiol plus clobazam.