18.3.1.1 Current Standard of Care

The current UK standard of care for chronic migraine is that patients who have failed three or more oral prophylactic medications should be offered 2 cycles of botulinum toxin injections. The pooled results from the double-blind placebo-controlled PREEMPT 1 and PREEMPT 2 trials provided evidence for the effectiveness of botulinum toxin in chronic migraine prophylaxis [4]. In 2012, based on manufacturer’s data on costs and the pooled PREEMPT results, NICE performed a single technology appraisal to approve botulinum toxin injection as a cost-effective treatment in chronic migraine in patients who have already failed three or more standard prophylactic pharmacological therapies and for those appropriately managed for medication overuse [5]. NICE has, however, recommended that treatment with botulinum toxin should be stopped in patients who have converted to episodic migraine (≤15 migraines/month) termed positive stopping rule and in those who respond inadequately (≤30% reduction in headache days/month after two treatment cycles) termed negative stopping rule. So how might TMS compare with the ‘gold standard’ of botulinum toxin injection in migraine prophylaxis for both cost and clinical effectiveness?

18.3.1.2 Cost Impact Study

Using botulinum toxin injection for comparison, one group assessed the cost impact of using non-invasive sTMS in a UK NHS setting [2]. In this study, subjects were assumed to have at least 2 cycles of botulinum toxin injections on the NHS before reassessment for further 12 weekly injections. In the sTMS group, subjects were assumed to have used the device for the first quarter without charge before a reassessment for funding for further usage on the NHS. This initial quarter usage of sTMS was provided for free by the manufacturer as a risk sharing so NHS only pays for the responders. Continuation rates in the two treatments were assumed to be the same based on the responder rates on previous studies. Using this information, the authors predicted the various costs associated with the average therapy for a year, including costs associated with medication/device utilisation, costs of clinical appointments and costs of hospital admissions secondary to headache. The costing model predicted that the costs of the risk-sharing sTMS treatment over the first 12-month period (£1466) was in fact considerably lower than the cost of using the currently NICE-approved botulinum toxin injection (£2923).

There are some important limitations to this study. The model only applies for the first 12 months. The saving though remains uncalculated will be much less (if at all) in the subsequent years as both treatment arms are paid for the whole year. Although the study has compared the costs of both TMS and botulinum toxin in chronic migraine, it assumed the responder rates of both treatments to be identical to the continuation rates from past studies [6], but did not actually appraise the clinical effectiveness nor the tolerability of the treatments.

18.3.1.3 Clinical Effectiveness of sTMS in Migraine

Lipton et al. [7] were the first to report the efficacy of sTMS in patients suffering from migraine with aura. Pain-free responses were significantly higher in the active group versus sham (39% vs. 22% p = 0.0179). The effect was sustained over 24–48 h. Patients were not able to differentiate between active and sham stimulus.

The post-marketing study assessed the use of sTMS in acute migraine with and without aura as well as prophylaxis in episodic and chronic migraine [6]. Subjects chosen were either intolerable to the standard acute medications or did not find them effective, which is comparable to patients who would be considered for botulinum toxin injections in the UK. Subjects were advised to apply 2 pulses of sTMS as early as possible when their migrainous symptoms begin and then repeat the 2 pulses every 15 min for 1–2 h until their symptoms resolve. With time, the number of pulses could increase. The sTMS treatment applied was therefore not strictly prophylactic and different treatment regimens were used. In contrast, those treated with botulinum toxin were all injected on a standard prophylactic protocol. Nevertheless, in this open-label study, patients reported an improvement in pain intensity (62%), as well as improvement in associated migrainous symptoms of nausea (52%), photophobia (55%) and phonophobia (53%) over a period of 12 weeks. There was also a reduction in the median number of migraine days from 15 (IQR 10–20) at baseline to 6 (IQR 3–13) at 12 weeks which was statistically significant (p < 0.001) and the number of days requiring acute medication use reduced by a mean of 8.5 days. The main criticism of the Bhola study was being open-label; hence, a large placebo effect may have contributed to the given efficacy.

The efficacy of sTMS was further studied in another open-label prospective multicentre study in the USA. The ESPOUSE study consented 263 patients of which full analysis was available in 132 patients [8]. For prophylaxis, patients used two pulses four times a day, while acute treatment was three consecutive pulses followed by another two pulses after 15 min of no relief. The study met the primary end point in reduction of headache days in weeks 9–12 of treatment and had a 50% responder rate of 46%. There was improvement in HIT-6 score as well as reduction in acute medication use. Like the Bhola study, this was open-label and a large placebo effect was the main criticism. There are currently no controlled studies looking at sTMS in migraine prophylaxis.

18.3.1.4 Repetitive TMS

18.3.1.5 Summary

18.4.1.1 Current Standard of Care

The current UK standard of care (SoC) in treating cluster headache patients has been based on pharmacological treatment for many years. Acute treatment of cluster headache is typically subcutaneous or intranasal triptan and high flow oxygen. Standard prophylactic treatments include verapamil, corticosteroids, lithium, and topiramate.

18.4.1.2 Cost-Effectiveness Study

Morris et al. [3] analysed the cost effectiveness over 1 year of non-invasive vagal nerve stimulation (nVNS) when used as an adjunctive therapy to SoC in a German statutory health insurance model. Their analysis was based on results from the PREVA trial, which compared adjunctive nVNS plus SoC with just SoC alone [16].

The cost-effective analysis found estimated mean annual costs of €7096.69 for nVNS plus SoC versus €7511.35 for SoC alone as well as mean quality-adjusted life years (QALY) of 0.607 for nVNS plus SoC versus 0.522 for SoC alone [3]. It is important to note that their modelling did not consider the potential economic impact of fewer follow-up appointments which could result in further cost reductions for nVNS adjunctive therapy. The same study also applied the same analysis, but from a UK perspective and found estimated mean costs of £5409.83 for nVNS plus SoC versus £5393.31 for SoC and mean QALY of 0.538 for nVNS plus SoC versus 0.438 for SoC alone.

There are, however, some important limitations to these assumptions. The cost benefit is only for the first year of treatment and long-term saving remains unknown. The clinical data used come only from single study, that is, PREVA, which itself has limitations discussed in later section.

The device gammaCore is currently priced at around £750 for a 3-month treatment. Over this period, a patient may administer up to 30 stimulations per day, six of which are administered as prophylactic dosing. The additional 24 stimulations may be administered as acute therapy if required. Its use has the potential to significantly decrease the use of both triptans and/or oxygen during acute attacks, from where there are both potential economic and quality of life benefits to be realised for payers and patients. There may also be potential for a reduction in outpatient and or telephone consultations with overall improvement. However, this has to be formally evaluated through health economic analysis.

18.4.1.3 Clinical Effectiveness of nVNS in Cluster Headache

A summary of the trials investigating nVNS in cluster headache is shown in Table 18.3. The PREVA trial compared SoC with adjunctive nVNS treatment on chronic cluster headaches [18]. The study had three phases, namely a 2-week baseline followed by a 4-week randomised period (SoC plus nVNS versus SoC alone), and a 4-week extension phase (SoC plus nVNS). Subjects undergoing nVNS treatment had a mandatory prophylaxis of three 2-min stimulations twice daily but also had the option to administer acutely at pain onset.

Table 18.3

Summary of results for studies investigating nVNS in cluster headaches

Study	n	Methodology	Summary of results
Nesbitt et al. [17]	19	Open-label study 52-week follow-up	Prophylactic treatment with nVNS resulted in a significant reduction mean attack frequency per day from 4.5 to 2.6 (p < 0.0005). nVNS treatment terminated the cluster headache attack within an average of 11 ± 1 min for all the attacks treated
Gaul et al. [18] PREVA trial	97	Randomised open-label study nVNS plus SoC versus SoC Chronic cluster patients only Prophylactic stimulations twice daily plus abortive stimulations 4 weeks randomised phase 4-weeks open-label phase	There was a significant difference in the number of cluster attacks/week for nVNS plus SoC com vs. SoC alone (−5.9 vs. −2.1 respectively, p = 0.02) There was a significant difference in the 50% responder rate between SoC plus nVNS and SoC alone (40% vs. 8.3%, respectively, p < 0.001) There was a greater change from baseline for visual analogue scale for pain (p = 0.039) Use of nVNS as abortive therapy had no significant effect on the duration or the intensity of the attack (data not shown by authors)
Silberstein et al. [19] ACT1 trial	150	Randomised double-blind control nVNS vs. sham Acute stimulations only 1-month double-blind phase 3-month open-label phase	The response rate was not different between nVNS and sham (p = 0.10) In episodic cluster cohort only, a significant difference in response rate with nVNS over sham (34.2% vs. 10.6 respectively, p = 0.008). In chronic cluster cohort, there was no significant difference in response rate with nVNS and sham (13.6% vs. 23.1% respectively, p = 0.48) Sustained response rates were significant higher in nVNS over sham for all patients (p = 0.008) and for those with episodic cluster (p = 0.008)
Goadsby et al. [20] ACT2 trial	102	Randomised double-blind control nVNS vs. sham Acute stimulations only 1-week run in 2-week double-blind phase	The pain free rate 15 min after initiation of treatment not different between nVNS and sham (14% vs. 12%) In episodic cluster cohort, nVNS was superior to sham at achieving pain free status at 15 min (48% vs. 6%, p < 0.01) In chronic cluster cohort, no significance difference between nVNS and sham (5% and 13%)
Gaul et al. [16] Post hoc analysis of PREVA	97	Post hoc analysis of PREVA [18]	Compared to SoC alone, nVNS plus SoC led to a significantly lower mean weekly attack frequency by week 2 of the randomised phase and this remained significant until week 3 of the extension phase (p < 0.02) 25%, 50%, 75% responder rates were significantly greater for nVNS plus SoC than for SoC alone

nVNS non-invasive vagal nerve stimulation, SoC standard of care

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

No related posts.

Stay updated, free articles. Join our Telegram channel

Join