FormalPara Key Summary Points

Why carry out this study?

Medication overuse (MO) is a common complication in patients with chronic migraine, resulting from excessive use of acute headache medication and leading to MO headache

Fremanezumab targets the calcitonin gene-related peptide, which should significantly reduce the need for acute headache medication; however, studies are lacking on this effect, especially in Asian populations

What was learned from the study?

We learned from this post hoc analysis that Japanese patients with chronic migraine treated with fremanezumab had, on average, significantly fewer headache days of moderate or greater severity (HDs) and migraine days per month, regardless of whether they had MO or not

Furthermore, significantly greater numbers of fremanezumab-treated patients had ≥ 50% reduction in HDs than placebo, regardless of MO status, and were more likely to revert to no MO status

Although MO remained an issue for many patients, it appeared that fremanezumab was at least as effective as a preventive treatment in chronic migraine patients with MO and may be beneficial at reducing the impact of MO to some extent

Introduction

Patients with chronic migraine commonly take medications, such as simple and combination analgesics, triptans and opioids, for acute headache treatment. This can potentially lead to medication overuse (MO) [1, 2], which may contribute to the development and/or maintenance of MO headache [3], which often leads to reduced quality of life [4,5,6].

The initial management of MO headache is typically focused on patient education regarding the risks of acute medication overuse, effective prophylaxis and supported discontinuation of the overused medication, including via medication breaks. Regarding prophylaxis, studies have clearly shown that preventive migraine therapies targeting the calcitonin gene-related peptide (CGRP) significantly reduce the extent to which patients need to take acute headache medication [7,8,9,10,11,12,13]; however, to date, few studies have explored the efficacy of these treatments in patients who have MO [14,15,16,17]. Most of the participants in these trials were Caucasian, and efficacy in Asian populations has not been established.

Fremanezumab, a fully humanized monoclonal antibody targeting CGRP, has been extensively trialed for chronic or episodic migraine [7, 18,19,20]. HALO, an international large-scale phase 3 trial, showed that fremanezumab monthly or quarterly significantly reduced the average number of headache days of moderate or greater severity and produced a greater response rate than placebo in patients with chronic migraine [7]. Most recently, a phase 2b/3 placebo-controlled trial in Japanese and Korean patients with chronic migraine demonstrated that subcutaneous monthly or quarterly fremanezumab administration was effective and safe for migraine prevention in this patient population [18].

To determine the influence of MO on the outcomes observed with fremanezumab, this post hoc analysis used data from the phase 2b/3 trial to assess outcomes in Japanese patients with chronic migraine with and without MO at baseline.

Methods

Study Design

The design, patient population, inclusion and exclusion criteria of this multicenter, double-blind, placebo-controlled, randomized, parallel-group phase 2b/3 trial in Japanese and Korean patients with chronic migraine (ClinicalTrials.gov identifier, NCT03303079) has been previously described [18]. Briefly, patients 18–70 years of age with chronic migraine (as assessed by the International Classification of Headache Disorders [ICHD], 3rd edition [beta version] [21] or clinical judgment), who were aged ≤ 50 years when their migraines began, were enrolled. The trial from which data for this analysis were taken was conducted in compliance with the International Conference on Harmonisation Good Clinical Practice Guideline, the Declaration of Helsinki and local regulatory requirements and was approved by an institutional or independent review board or ethics committee. All patients provided written informed consent at enrollment.

Treatment

Patients were randomized (1:1:1 ratio) at baseline to receive subcutaneous injections of monthly fremanezumab, quarterly fremanezumab or placebo as three doses of either fremanezumab or placebo at 4-week (‘monthly’) intervals. Monthly fremanezumab group patients received fremanezumab 675 mg as three 225 mg/1.5 ml injections at baseline and then fremanezumab 225 mg as a single 225 mg/1.5 ml injection at month 1 and 2. Quarterly fremanezumab group patients received fremanezumab 675 mg as three 225 mg/1.5 ml injections at baseline and single 1.5 ml injections of placebo at month 1 and 2. Placebo group patients received three 1.5-ml placebo injections at baseline and a single 1.5-ml placebo injection at month 1 and 2.

Regarding the use of a higher initial dose of fremanezumab in the monthly group, CM is considered a more severe disease than EM requiring early appropriate preventive treatment. Hence, a loading dose of 675 mg was implemented to facilitate a rapid onset of effect.

Post Hoc Analysis

For this post hoc analysis, data from Japanese patients enrolled in the original trial only were analyzed [18]. We stratified patients based on whether they met MO criteria or not. MO was defined as acute headache medication use for ≥ 15 days, migraine-specific acute medication use for ≥ 10 days or use of combination headache medications for ≥ 10 days during the 28 days before baseline [21].

Outcomes included: (1) the mean change from baseline in the average number of headache days of moderate or greater severity per month (28 days) (the primary endpoint of the original trial) during the 12-week period following the first study medication dose; (2) the mean change from baseline in the average number of migraine days per month during the 12-week period after the initial study medication dose; (3) the proportion of patients reaching ≥ 50% reduction in the average number of headache days of moderate or greater severity per month during the 12-week period after the initial study medication dose (50% responder); (4) the mean change from baseline in the average number of days with use of any acute headache medications per month during the 12-week period following the initial dose of study medication; (5) the mean change from baseline in disability score of the six-item Headache Impact Test (HIT-6) at 4 weeks after the third (final) injection, and during the 12-week period following the initial dose of study medication [22], at 4 weeks following the final (third) injection; (6) the proportion of patients changing status from MO to non-MO during the 12-week period after the first initial study medication dose at 4 weeks after the third (final) injection and during the 12-week period following the initial study medication dose.

Statistics

Descriptive analysis was performed in which the qualitative variables are presented as number (n) and proportion (percentage) of patients, and the quantitative variables are described by means, standard deviation (SD), standard error (SE) and 95% confidence intervals (95% CI). For the quantitative variables, the least-squares mean (LSM) ± SE change from baseline was analyzed via an analysis of covariance (ANCOVA) model, in which treatment, sex, country and baseline preventive medication use were fixed effects and covariates were the baseline number of headache days of moderate or greater severity and years since onset of migraines. The LSM difference between fremanezumab and placebo in each patient group and its two-sided 95% CI and p values were calculated. For the proportion of 50% responders, the difference between placebo and fremanezumab was assessed via the Cochran-Mantel-Haenszel method adjusted by preventive medication use at baseline. For all analyses, a p value < 0.05 for the difference was defined as significant. In this post hoc analysis, all statistical analyses were calculated using SAS version 9.4 (SAS Institute, Cary, NC).

Results

Study Population

In total, 571 chronic migraine patients were randomized in the original study; of these, 479 Japanese patients received treatment with fremanezumab monthly (n = 159), fremanezumab quarterly (n = 159) or placebo (n = 161). Of these 479 patients, 320 patients (66.8%) had MO at baseline (fremanezumab monthly, n = 109 [68.6%]; fremanezumab quarterly, n = 99 [62.3%]; placebo, n = 112 [69.6%]).

In general, patient demographics and baseline clinical characteristics were similar among patients with or without MO at baseline (Table 1), including in relation to age, gender proportions and weight/body mass index.

Table 1 Baseline demographics and characteristics in Japanese patients with chronic migraine with or without MO (n = 479)
Full size table

Outcomes

During the 12-week period following the initial study medication dose, the mean change from baseline in the monthly (28-day) average number of headache days of moderate or greater severity was significantly decreased with fremanezumab compared with placebo, regardless of MO (Fig. 1). The mean ± SE differences versus placebo in this outcome in patients with and without MO receiving fremanezumab monthly were – 2.0 ± 0.6 (p = 0.0012) and – 1.6 ± 0.8 (p = 0.0437) headache days/month, respectively, while the mean differences versus placebo in patients with and without MO receiving fremanezumab quarterly were – 1.8 ± 0.6 (p = 0.0042) and − 1.5 ± 0.8 (p = 0.0441) headache days/month, respectively.

Fig. 1
figure 1

Least-squares mean (LSM) ± standard error (SE) change from baseline in the average number of headache days of moderate or greater severity during the 12-week period after the initial study medication dose in patients with and without MO

Full size image

In patients with baseline MO, the mean change from baseline in the monthly average number of migraine days during the 12-week treatment period was significantly greater with fremanezumab monthly (p = 0.0008) and quarterly (p = 0.0124) compared with placebo (Table 2). Patients without MO at baseline who received fremanezumab monthly (p = 0.0129) but not quarterly (p = 0.1837) showed similar results.

Table 2 Summary of other endpoints in Japanese patients with chronic migraine with or without MO (n = 479)
Full size table

The proportion of 50% responders during the 12-week period following the initial fremanezumab dose was significantly greater in patients with MO who received fremanezumab monthly (p = 0.0040) or quarterly (p = 0.0070) compared with patients who received placebo. Furthermore, significantly more patients without MO who received fremanezumab monthly (p = 0.0132) and quarterly (p = 0.0126) were 50% responders compared with placebo-treated patients (Table 2).

Fremanezumab treatment led to significantly greater reductions in the average number of days of any acute headache medication use per month versus placebo in patients with MO at baseline (monthly p = 0.0138; quarterly p = 0.0251) and patients without MO at baseline (p = 0.0202 and p = 0.0122, respectively; Table 2).

Fremanezumab monthly and quarterly also improved HIT-6 scores after 12 weeks; however, in patients with baseline MO, only fremanezumab monthly led to a significant improvement versus placebo (p = 0.0311). In contrast, in patients without MO at baseline, only fremanezumab quarterly was associated with a significant improvement versus placebo (p = 0.033; Table 2).

In chronic migraine patients with baseline MO, significantly greater proportions of patients treated with fremanezumab monthly or quarterly reverted to no MO during the 12-week treatment period than those who received placebo (p = 0.0115 and p = 0.0272, respectively; Fig. 2).

Fig. 2
figure 2

Proportion of patients changing status from with MO to without MO during the 12-week period after the initial study medication dose

Full size image

Discussion

As far as we can determine, this paper is the first to demonstrate the usefulness of CGRP-related drugs among Asian patients with MO, specifically Japanese patients with chronic migraine and MO. Fremanezumab significantly reduced the number of headache days of moderate or greater severity per month and the average number of days with use of any acute headache medications per month, regardless of whether patients had MO or not at baseline. Fremanezumab was associated with a higher proportion of 50% responders and of patients changing to no MO status over the 12-week treatment period. Some differences were observed between fremanezumab monthly and fremanezumab quarterly dosage regimens in relation to specific endpoints although these do not greatly influence the overall conclusions regarding the benefits of fremanezumab in this setting.

These post hoc analyses results confirm those of a similar post hoc subanalysis of HALO CM, an international, double-blind, placebo-controlled, parallel-group randomized phase 3 trial, which demonstrated significant reductions in number of headache days of moderate or greater severity, migraine days and acute medication use with fremanezumab monthly and quarterly in chronic migraine patients with or without MO [15]. Furthermore, regardless of MO, significantly more patients receiving fremanezumab monthly or quarterly in HALO CM achieved a clinically meaningful response rate (≥ 50% reduction in the monthly average number of headache days of moderate or greater severity), and more fremanezumab recipients reverted to no MO during the 12-week treatment period [15]. As described above, a significant improvement in the monthly average number of migraine days with fremanezumab quarterly versus placebo was not observed in patients without MO in this analysis. In contrast, the HALO CM post hoc analysis reported significant reductions in this endpoint with both fremanezumab monthly and quarterly versus placebo in patients with and without MO [15]. In search of a possible reason for this discrepancy, we initially observed that the number of overall headache days was not notably different between patients with and without MO, whereas the numbers of migraine days and moderate or more headache days were fewer in patients without MO. Although patients without MO group met the entry criteria, it is possible that a higher proportion of patients with more tension-type-like headaches were included in this group. To date, there are no specific studies or subgroup analyses showing that fremanezumab or other anti-CGRP antibodies are less effective in such patients. However, there is some evidence to show that patients with both migraine and tension-type headache tend to have higher levels of depression and neuroticism [23], and lower health-related quality of life [24], than patients with migraine or tension-type headache alone. We acknowledge that this evidence is related to migraine and tension-type headache rather than migraine-like and tension-type-like headache as used in the ICHD definition of CM but is cited as a reference. Although speculative and in need of validation, we suggest that headache (migraine-like and tension-type-like headache) may have a more complex pathophysiology and a resulting lower treatment response as seen in our analysis.

From a clinical perspective, it is important to show whether administration of fremanezumab could reduce patient dependence on acute headache medication, so it is encouraging that this post hoc analysis demonstrated 44–46% of fremanezumab recipients with MO at baseline no longer overused medications at 12-week trial end. These findings are promising, as they suggest that patients receiving fremanezumab may be able to revert from MO without experiencing detoxification and the symptoms typically associated with treatment withdrawal [25]. This is in line with reports that CGRP-targeting agents (such as erenumab, galcanezumab, fremanezumab and eptinezumab) are effective in chronic migraine patients with MO headache, suggesting that discontinuation of the causative drug may not be necessary [25,26,27].

Our work has several limitations. First, these subanalyses were not prespecified in the trial protocol, so the original trial was not powered to investigate the impact of MO on the efficacy of fremanezumab, which may explain why significant differences versus placebo were not observed for some endpoints when they were observed in the original trial [18]. While the definition of MO is the same as that used in the HALO CM trial, details of the types and combinations of medications used at baseline were not considered and may affect the results of the analysis. As this study did not collect information to allow subdivision of patients according to the types of overused medication, certain restrictions on the significance of clinical diagnosis and treatment must be acknowledged. However, at least in relation to opioids, overuse is considered very rare in Japan given that opioids and barbiturates are not contained in over-the-counter combination analgesics and opioids cannot be prescribed for headache without terminal cancer pain [28]. In a similar fashion, despite a similar proportion of patients who received preventive medication at baseline, patients were not analyzed according to preventive medication type, which represents another limitation of this analysis. We also acknowledge that the period for collection of data on the extent of medication use (28 days before baseline) was considerably shorter than the period specified in the definition of MO according to ICHD-3 criteria (> 3 months) [21]. Furthermore, the impact of patient comorbidities, such as anxiety as well as other factors (e.g., emotional state), that may contribute to MO was also not considered. Another limitation of the current analysis is that only Japanese patients from the original trial were included; thus, the results may not be extrapolated to other populations. However, the results of this post hoc analysis were broadly similar to those of a post hoc analysis of international trials of fremanezumab [15], indicating that the results may be applicable to other patient populations. It is also important to highlight the substantial migraine burden and barriers to care in Japan [29, 30], and the results of this analysis may be used to improve patient outcomes in Japanese individuals with chronic migraine.

Conclusion

These post hoc analyses show that fremanezumab is effective for preventive treatment of chronic migraine in Japanese patients with or without MO and demonstrate a benefit over placebo in reducing MO.