4 Committee discussion
4.1 The evidence base for gammaCore is small. But cluster headache is a rare condition, and the quality of the studies was good. The committee concluded that gammaCore appears to be effective in some but not all people. In people whose condition responds, it can make a significant difference to their symptoms and quality of life. The clinical experts said that, in their experience, 25% to 50% of people have cluster headaches that respond to gammaCore.
4.2 The published evidence of efficacy was supported by evidence from people with cluster headache that was submitted to the committee. Sixty people with cluster headache were surveyed. They described the impact of the condition on their lives, and how it responded to gammaCore. Two patient organisations also submitted reports. The committee understood the devastating impact that this condition can have on the lives of sufferers and the desperation that can result from ineffective treatment. For example, the submission from OUCH (Organization for the Understanding of Cluster Headache) reported that on average 5 people a year in the UK end their lives because they are no longer able to live with the pain of cluster headaches. The committee noted the life-changing effects that gammaCore had had for many people in the survey.
4.3 gammaCore is the only device available that treats cluster headache by non-invasive vagus nerve stimulation. The clinical experts described the results of studies that supported the mechanism of action of gammaCore. They explained that, patho-physiologically, vagus nerve stimulation could plausibly reduce pain in people with cluster headache. The clinical experts also explained that, while a placebo effect is likely in all treatments for chronic pain conditions, it is unlikely that the benefits of gammaCore can be explained by this alone. They said that gammaCore's response rate is higher and its therapeutic benefits more sustained than would be expected for a placebo treatment. The committee was also convinced by the experts' argument that people with such a debilitating condition welcome a treatment response regardless of its mode of action as long as it is safe and has no adverse effects.
The free initial 3-month period means gammaCore is worth trying if there is a possibility it can reduce attacks and medication
4.4 Although the evidence for gammaCore is subject to some uncertainty, the committee noted that gammaCore would only be used in people who find it effective after the first 3 months of treatment. Cluster headache attacks have a profound negative impact on everyday life so any treatment that can help reduce this is worth trying. The committee also noted that gammaCore, as a non-drug treatment, is unlikely to interact with any other treatments and may help reduce the number of drugs that are prescribed to this patient group.
A doctor should decide if treatment with gammaCore has been successful after 3 months, after consulting the patient
4.5 The clinical experts stated that the definitions of a successful response to 3 months of treatment with gammaCore vary and are subjective. This was also the case in the published evidence. However, they explained that it's usually clear within 3 months if someone's cluster headaches have reduced meaningfully in frequency and severity, leading to reduced medication use that justifies continuing treatment with gammaCore. The clinical experts said that people do not generally want to continue with treatments if they're not helping. The committee concluded that the decision about whether or not to continue with gammaCore after the first 3 months of treatment should be made by a doctor after consulting the patient.
There is enough evidence of clinical benefit for people with chronic and episodic cluster headache to recommend gammaCore for both groups
4.6 The experts explained the different patterns of symptoms that people with cluster headaches have. There is a difference in particular between people with chronic and episodic cluster headache. They also explained that some people with episodic cluster headache later become chronic sufferers and vice versa. But the natural history of the condition is unpredictable. This means it's uncertain how the condition is likely to respond to an intervention. The sham-controlled randomised clinical trials (ACT1 and ACT2) were not powered to examine therapeutic benefits separately in episodic and chronic cluster headache, and they only considered acute use of gammaCore. But people with episodic cluster headache had particular benefit. In the open-label randomised controlled trial (PREVA), people with chronic cluster headache had clinical benefit. The patient survey included responses from 12 people with episodic cluster headaches, and 9 of them said they had received substantial clinical benefits from gammaCore. The clinical experts said they most often use gammaCore for people with chronic cluster headache. They said that treatment effects were more difficult to measure in episodic cluster headache. The committee concluded that overall there was enough evidence of clinical benefit for people with chronic and episodic cluster headache to recommend adopting gammaCore if treatment is successful in the first 3 months. It said it could not reliably make a therapeutic distinction between the 2 based on current evidence.
4.7 The clinical experts said that clinical follow up of people is essential because response to treatment with gammaCore is unpredictable. People should be reassessed after the first 3 months of treatment to review attack frequency and intensity, and use of treatments to stop acute attacks (oxygen, sumatriptan, zolmitriptan). The experts advised that only people whose cluster headaches respond to treatment with gammaCore should carry on using it. The experts recommended follow up again at 12 months and every year afterwards to determine long-term benefits and to give an opportunity to stop gammaCore if it's no longer effective.
4.8 There are no published reports of serious adverse events with gammaCore. The experts and manufacturer representative said none had been reported to them. The patient survey confirmed that the device is well tolerated and easy to use. People with cluster headache risk side effects from conventional pharmacological treatment. Sometimes they need invasive treatment such as implanted stimulators. gammaCore could help some people avoid or delay the need for these treatments. The safety and efficacy of using gammaCore has not been evaluated in people with an implanted medical device, people with heart conditions, people who are pregnant, lactating or aged under 18 years.
4.9 The clinical experts described their experience of using gammaCore in their NHS practice. They reported that it was usually used as a second or third-line treatment option to prevent chronic cluster headache attacks after verapamil, and possibly lithium, had been tried. They also explained that verapamil and lithium can have adverse effects, they need careful monitoring, and they may be contraindicated in some people. The experts said that if people benefit therapeutically from gammaCore – and around 25% to 50% do – they usually carry on benefiting from it. They said that some of their patients have been using the technology for 3 years or more. It has been difficult for clinical services to get funding for gammaCore. The experts explained that they get it through individual patient funding requests to commissioners.
4.10 People are trained to use gammaCore by specialist headache nurses. The manufacturer provides training resources free of charge. The clinical experts explained that training was simple, and that most people were able to learn how to use the device in one session. The manufacturer representative explained that if someone cannot use the device themselves because they have problems with manual dexterity, someone else can give treatment.
4.11 The company's cost model showed a potential £450 saving per patient over 1 year if gammaCore is used with standard treatment in people with chronic cluster headache. It noted that the cost savings were largely from a reduced need for sumatriptan to stop the symptoms of acute attacks. It also noted that they depended on the first 3 months of treatment being free of charge.
4.12 The data used in the cost modelling was from one open-label randomised controlled trial (PREVA) that included patients from the UK. The committee discussed uncertainties in the cost modelling but acknowledged that they cannot be resolved by the available evidence. For example, the impact of gammaCore on other outpatient, community or inpatient services – such as occipital nerve blocks, intravenous dihydroergotamine or implanted nerve stimulators – is unclear. Reduced cluster headache symptoms could mean other treatments or care could be reduced or stopped. But with no evidence to support this, it could not be considered in the cost modelling.
The study definition of a successful response to gammaCore may differ from a clinically meaningful response
4.13 The trial used to inform the cost model classified a successful response to gammaCore as cluster headache frequency reduced by 50% or more. The experts explained that, in clinical practice, several treatments may be needed to reduce cluster headache attacks to this degree. The clinical experts also advised that people classified as non-responders on this basis may still receive clinically meaningful benefits from gammaCore that would not be captured in the cost modelling.
4.14 gammaCore is offered for a free 3‑month initial period. The external assessment centre (EAC) identified this as a key driver of the cost savings. The company representatives assured the committee that the free period is a fixed part of its business model which would not change. They also clarified that there are no extra costs for conductive gel, training resources, or replacing the gammaCore device if it is broken, lost or stops working. The committee concluded that the technology costs used in the model are accurate.
4.15 The experts explained to the committee the importance of the treatments that people use to stop the symptoms of an acute attack of cluster headache. These include inhaled oxygen, sumatriptan and zolmitriptan. The committee considered it plausible that, if gammaCore reduces the frequency and severity of attacks, then medication use is also likely to reduce. More than half the people in the patient survey had reduced their medication use since starting treatment with gammaCore. Reduced sumatriptan use was another key driver of the cost savings for gammaCore in the model. The committee concluded that the clinical evidence would support this.
Cost modelling for gammaCore has limitations but cost savings are likely if it is only used by people who it is effective for
4.16 The committee accepted the EAC's rationale for not changing the model because there was no relevant additional evidence. The model had a 1‑year time horizon because this was the duration of the relevant study. The annual cost of gammaCore treatment increases after the first year because the free period no longer applies. But the committee considered that this could continue to be offset by savings from reduced medications used to stop the pain of an attack, as well as reduced or avoided care or treatments not captured in the cost modelling. The committee noted that it was important that only people who benefited clinically from gammaCore should use it. People who do not benefit should stop treatment with it to avoid incurring additional costs. Overall, the committee concluded that the model's cost saving of £450 per patient in the first year is plausible.