Efficacy and safety of erenumab in Japanese migraine patients with prior preventive treatment failure or concomitant preventive treatment: subgroup analyses of a phase 3, randomized trial

These were subgroup analyses of data from a Phase 3, randomized, double-blind, placebo-controlled study of 261 Japanese patients with EM or CM (ClinicalTrials.​gov identifier: NCT03812224) [18]. The study comprised an initial screening phase (up to 3 weeks), a 4-week baseline phase, and a 24-week double-blind treatment phase (DBTP), followed by a 28-week open-label extension period and an 8-week follow-up period (Supplemental Fig. 1). In the DBTP, patients were randomized (1:1) to receive erenumab 70 mg or placebo once monthly subcutaneously for 24 weeks. This study reports efficacy and safety data from the 24-week DBTP only.

The study enrolled Japanese patients with EM or CM aged 20–65 years, with a history of migraine (International Classification of Headache Disorders, 3rd edition criteria) with or without aura for ≥12 months before screening. CM was defined as ≥15 headache days per month, of which ≥8 headache days on average across the 3 months prior to screening met the criteria for migraine days. EM was defined as <15 headache days per month, of which ≥4 headache days on average across the 3 months prior to screening met the criteria for migraine days. During the baseline phase, patients were required to have the same migraine type as prior to screening and demonstrate ≥80% compliance with a daily electronic diary (eDiary) used to record migraine symptoms.

The main exclusion criteria included age >50 years at migraine onset, body mass index >40 kg/m², history of cluster/hemiplegic migraine, migraine with continuous pain, no therapeutic response to ≥3 prior migraine preventive classes (defined as no reduction in headache frequency, duration, or severity after 6 weeks of treatment at the generally accepted therapeutic dose, as judged by the investigator), or any significant medical condition that precluded study entry.

Patient subgroups were defined on the basis of prior and concomitant migraine preventive treatment status. The “failed-yes” subgroup included patients who had previously failed ≥1 prior preventive treatment category due to lack of efficacy and/or unacceptable tolerability, as recorded by the investigator. The “failed-no” subgroup included patients who had never failed prior preventive treatment (composed of patients who had not previously received preventive treatment or received preventive treatment and had not failed). The “concomitant preventive-yes” subgroup included patients receiving a concomitant standard of care migraine preventive treatment and the “concomitant preventive-no” subgroup included patients not receiving a concomitant migraine preventive treatment.

Migraine preventive treatments were defined according to the American Academy of Neurology/American Headache Society [19], and Japanese Headache Society guidelines [6]. The following were the categories of prior migraine preventive treatments: topiramate; beta-blockers (e.g. propranolol or metoprolol); tricyclic antidepressants (e.g. amitriptyline or nortriptyline); divalproex sodium or sodium valproate; calcium channel blockers (e.g. flunarizine, verapamil, lomerizine); serotonin-norepinephrine reuptake inhibitors (e.g. venlafaxine, desvenlafaxine, duloxetine, and milnacipran); botulinum toxin; antihypertensives (lisinopril or candesartan); or other medications. Except for the category of “other medications,” any of the aforementioned medications were considered a concomitant migraine preventive medication, along with clonidine, guanfacine, cyproheptadine, pizotifen, methysergide, butterbur, feverfew, magnesium, and riboflavin.

Headache data were captured daily via eDiary measurement during the baseline period and the DBTP. The primary efficacy endpoint was the change from baseline in mean MMD. A migraine day was defined as any calendar day on which the patient had onset, continuation, or recurrence of a qualified migraine (i.e. migraine with or without aura lasting for ≥4 h with either ≥2 pain features or ≥1 associated nonpain symptom, or both) as recorded in the eDiary. Any calendar day on which acute MSM was used was also counted as a migraine day regardless of the headache duration, pain features, and associated symptoms. Secondary efficacy endpoints were the proportion of patients achieving a ≥50% reduction from baseline in mean MMD and the change from baseline in mean monthly acute MSM treatment days.

All efficacy endpoints were assessed over the last 3 months (Months 4–6) of the DBTP. Treatment-emergent adverse events (TEAEs) were collected throughout the DBTP and graded according to Common Terminology Criteria for Adverse Events (CTCAE) version 4.03.

All statistical analyses were conducted using SAS version 9.2 (SAS Institute Inc., Cary, NC, USA). All efficacy analyses in the prior treatment failure subgroups were prespecified, and efficacy analyses in the concomitant preventive treatment subgroups were post hoc. The full analysis set included all randomized patients. The efficacy analysis set included all patients who received ≥1 dose of erenumab 70 mg or placebo and had ≥1 change from baseline measurement in MMD during the DBTP. The safety analysis set included all randomized patients who received ≥1 dose of erenumab 70 mg or placebo, unless a patient received the incorrect dose during the DBTP. Changes from baseline in mean MMD and mean monthly acute MSM treatment days over Months 4–6 were analyzed using generalized linear mixed-effect models, which included treatment, visit, treatment-by-visit interaction, stratification factor of migraine type (EM or CM), and baseline value as covariates and assumed a first-order autoregressive covariance structure. An additional stratification factor of migraine preventive treatment status (ever used or never used) was not included in the adjusted models. Adjusted analysis results were obtained using contrasts. The least squares mean (LSM) changes from baseline with 95% confidence intervals (95% CIs) for each treatment group, the treatment difference (erenumab 70 mg − placebo) with the 95% CI, and nominal p-values without multiplicity adjustment were calculated. Achievement of a ≥50% reduction from baseline in mean MMD was analyzed using a stratified Cochran-Mantel-Haenszel test (CMH), with missing data imputed as nonresponse. Adjusted odds ratios (ORs) and associated p-values were obtained from the CMH test, stratified by migraine type (EM or CM). Treatment-by-subgroup interaction p-values were obtained from generalized linear mixed models adjusting for treatment group, stratification factor of migraine type (EM or CM), subgroup and treatment-by-subgroup interaction, with data imputed using the nonresponder imputation method.

A total of 261 patients were randomized: 130 patients to the erenumab 70 mg group and 131 patients to the placebo group. Of the 261 patients randomized, 117 (44.8%) patients were in the failed-yes subgroup. The 144 patients in the failed-no subgroup included 59 (41.0%) treatment-naïve patients, while 85 (59.0%) patients had prior preventive treatment use but had not failed. There were 92 (35.2%) patients in the concomitant preventive-yes subgroup (Table 1). There was a higher proportion of patients with ≥1 prior preventive treatment failure(s) in the concomitant preventive-yes subgroup compared with the concomitant preventive-no subgroup: 50 (54.3%) versus 67 (39.6%) patients, respectively. Calcium channel blockers were the most common concomitant preventive treatment received by patients in the concomitant preventive-yes subgroup (Supplemental Table 1).

Across the subgroups, the mean age was 44–45 years and most patients were female. The mean (standard deviation [SD]) number of MMD at baseline ranged from 10.2 (4.9) to 13.9 (6.0), and the mean (SD) number of monthly acute MSM treatment days ranged from 8.3 (4.8) to 10.6 (6.4) (Table 1). The failed-yes and concomitant preventive-yes subgroups generally had more severe disease than the failed-no and concomitant preventive-no subgroups, based on the greater proportion of patients with CM and the higher number of mean MMD. The concomitant preventive-yes subgroup had more patients who had previously failed prior preventive treatment than the concomitant preventive-no subgroup. Patients in the failed-yes subgroup had a higher number of mean monthly acute MSM treatment days than those in the failed-no subgroup (Table 1).

From baseline to Months 4–6, significant reductions in mean MMD occurred with erenumab 70 mg versus placebo in all patient subgroups. The differences in LSM change from baseline in mean MMD with erenumab 70 mg versus placebo were − 1.9 (95% CI: − 3.3, − 0.4; P = 0.013) and − 1.4 (95% CI: − 2.5, − 0.3; P = 0.012) in the failed-yes and failed-no subgroups, respectively. The differences in LSM change from baseline in mean MMD with erenumab 70 mg versus placebo were − 1.7 (95% CI: − 3.3, 0.0; P = 0.053) and − 1.6 (95% CI: − 2.6, − 0.5; P = 0.003) in the concomitant preventive-yes and concomitant preventive-no subgroups, respectively (Fig. 1).

At Months 4–6, a higher proportion of patients treated with erenumab 70 mg achieved a ≥50% reduction in mean MMD versus placebo in all patient subgroups (Fig. 2). In patients treated with erenumab 70 mg, a ≥50% reduction in mean MMD was achieved by 27.1%, 35.2%, 22.5%, and 35.6% of patients in the failed-yes, failed-no, concomitant preventive-yes, and concomitant preventive-no subgroups, respectively.

The common ORs versus placebo of achieving a ≥50% reduction in mean MMD were 4.8 (95% CI: 1.5, 15.3; P = 0.004) and 1.7 (95% CI: 0.8, 3.4; P = 0.15) in the failed-yes and failed-no subgroups, respectively, and 1.9 (95% CI: 0.6, 5.5; P = 0.26) and 2.4 (95% CI: 1.2, 5.0; P = 0.013) in the concomitant preventive-yes and concomitant preventive-no subgroups, respectively (Fig. 2).

From baseline to Months 4–6, clinically relevant reductions in the mean number of monthly acute MSM treatment days were observed in patients receiving erenumab 70 mg versus placebo in all patient subgroups. The differences in LSM change from baseline in mean monthly acute MSM treatment days with erenumab 70 mg versus placebo were − 1.7 (95% CI: − 3.0, − 0.5; P = 0.007) and − 1.3 (95% CI: − 2.2, − 0.3; P = 0.009) in the failed-yes and failed-no subgroups, respectively. The differences in LSM change from baseline in mean monthly acute MSM treatment days with erenumab 70 mg versus placebo were − 1.2 (95% CI: − 2.5, 0.1; P = 0.076) and − 1.6 (95% CI: − 2.6, − 0.7; P < 0.001) in the concomitant preventive-yes and concomitant preventive-no subgroups, respectively (Fig. 3).

Overall, 60.3%–69.5% and 53.8%–65.0% of patients in the failed-yes and concomitant preventive-yes subgroups, respectively, had ≥1 TEAE. Across all subgroups, most TEAEs were CTCAE grade 2. The incidence of TEAEs for placebo compared with erenumab 70 mg was broadly comparable within each subgroup; however, a lower proportion of patients experienced ≥1 TEAE with placebo versus erenumab 70 mg in the concomitant preventive-yes subgroup (53.8% versus 65.0% of patients). The incidence of TEAEs in the concomitant preventive-yes subgroup treated with erenumab 70 mg was similar to the incidence in the concomitant preventive-no subgroup receiving erenumab 70 mg or placebo. The number of serious adverse events was low, and there were no adverse events leading to discontinuation of investigational product or fatal adverse events (Table 2).