Original Article Cephalalgia 2021, Vol. 41(10) 1075–1088 Fremanezumab for the Preventive ! International Headache Society 2021 Treatment of Migraine: Subgroup Article reuse guidelines: sagepub.com/journals-permissions Analysis by Number of Prior Preventive DOI: 10.1177/03331024211008401 journals.sagepub.com/home/cep Treatments with Inadequate Response Ladislav Pazdera1 , Joshua M Cohen2, Xiaoping Ning2, Verena Ramirez Campos2, Ronghua Yang2 and Patricia Pozo-Rosich3,4 Abstract Objective: To evaluate the efficacy of monthly or quarterly fremanezumab in patients with chronic migraine or episodic migraine and documented inadequate response to 2, 3, or 4 classes of prior migraine preventive medications. Methods: This is an exploratory analysis of a randomized, double-blind, placebo-controlled, phase 3b trial for patients with chronic migraine or episodic migraine and inadequate response to 2 to 4 prior migraine preventive medication classes randomized (1:1:1) to fremanezumab (quarterly or monthly) or placebo. In this exploratory analysis, changes from baseline in the monthly average number of migraine days during 12 weeks of double-blind treatment and adverse events were evaluated for predefined subgroups of patients by number of prior preventive medication classes with inadequate response. Results: Overall, 414, 265, and 153 patients had inadequate response to 2, 3, and 4 preventive medication classes, respectively. Changes from baseline in monthly average migraine days during 12 weeks were significantly greater with fremanezumab compared with placebo for patients with documented inadequate response to 2 classes (least-squares mean difference vs placebo [95% confidence interval]: quarterly, –2.9 [–3.83, –1.98]; monthly, –3.7 [–4.63, –2.75]), 3 classes (quarterly, –3.3 [–4.65, –1.95]; monthly, –3.0 [–4.25, –1.66]), and 4 classes (quarterly, –5.3 [–7.38, –3.22]; monthly, –5.4 [–7.35, –3.48]) of migraine preventive medications (all p < 0.001). No significant treatment-by-subgroup interactions were observed for any outcome (p interaction > 0.20 for all). Adverse events were comparable for placebo and fremanezumab. Conclusion: Significant improvements in efficacy were observed with fremanezumab compared with placebo, even in patients who had previously experienced inadequate response to 4 different classes of migraine preventive medications. ClinicalTrials.gov identifier: NCT03308968. Keywords Chronic migraine, episodic migraine, treatment failure, CGRP Date received: 8 March 2021; revised: 8 March 2021; accepted: 18 March 2021 1 Vestra Clinics, Rychnov nad Kneznou, Czech Republic 2 Teva Branded Pharmaceutical Products R&D, Inc., West Chester, PA, USA Introduction 3 Headache Unit, Neurology Department, Vall d’Hebron University Hospital, Barcelona, Spain Treatment guidelines recommend offering migraine 4 Headache and Neurological Pain Research Group, Vall d’Hebron preventive treatment to individuals with as few as Institute of Research (VHIR), Universitat Autònoma de Barcelona, 3 monthly headache days and severe disability (1). Barcelona, Spain. However, until the recent approvals of monoclonal Corresponding author: antibodies targeting the calcitonin gene-related Ladislav Pazdera, Vestra Clinics - Dedicated Research Clinics, peptide (CGRP) pathway, recommended preventive Jiráskova 1389, 51601 Rychnov nad Kneznou, Czech Republic. medications were not specifically designed or Email: pazdera@vestraclinics.org 1076 Cephalalgia 41(10) developed for migraine and were limited by slow onset of migraine (onset 50 years of age) according to of action, inadequate efficacy, poor adherence, and International Classification of Headache Disorders 3 suboptimal safety and tolerability (2). Treatment dis- beta version (ICHD-3 beta) criteria (8) and a history continuations are common with oral migraine preven- of migraine for 12 months prior to screening. tive medications (3), potentially resulting in inadequate Eligible patients also had documented inadequate migraine treatment, acute medication overuse, chroni- response within the past 10 years to 2 to 4 of the fol- fication, and higher healthcare costs (4,5). The burden lowing classes of prior migraine preventive medications: of migraine is generally higher for patients who have beta blockers, anticonvulsants, tricyclic antidepressants, failed prior preventive treatments, with negative effects calcium channel blockers, onabotulinumtoxinA, and on patients’ work and personal lives (6). valproic acid. An inadequate response was generally Fremanezumab is a fully humanized monoclonal documented in the patient’s medical record and defined antibody (IgG2Da) that selectively targets CGRP (7). by no clinically meaningful improvement (per the treat- Results from the FOCUS trial demonstrated the ing physician’s judgement) after 3 months of stable- efficacy and tolerability of fremanezumab treatment dosed treatment, discontinuation due to poor tolerabil- in patients with episodic migraine (EM) or chronic ity, or contraindication or unsuitability of treatment for migraine (CM) and documented inadequate the patient (7). At the time of screening, patients could response to 2 to 4 prior migraine preventive medication not use migraine preventive medications; patients con- classes (7). Understanding the influence of inadequate tinuing treatment with migraine preventive medications response to prior migraine preventive treatments on were excluded from the study. fremanezumab treatment outcomes could inform clini- cal decision making and guide optimal usage of frema- nezumab in patients with migraine. Here we report Standard protocol approvals, registrations, and results from analyses of the efficacy and tolerability patient consents of fremanezumab in predefined subgroups of patients The FOCUS study was approved by an independent with migraine from the FOCUS study by number of ethics committee or institutional review board at each classes of prior migraine preventive medications to study site, and each patient provided written informed which they had inadequate response (2, 3, or 4 classes). consent. Based on the clinical benefit with fremanezumab observed in a prespecified subgroup analysis of patients in the FOCUS study who had previously not Study design responded to topiramate and up to 3 other classes of The FOCUS study included a screening visit; a 28-day migraine preventive medications (7), we hypothesized run-in period; a 12-week, double-blind, placebo-con- that quarterly and monthly dosing of fremanezumab trolled treatment period; and a 12-week, open-label would be consistently effective in patients with inade- treatment period. For the 12-week, double-blind quate response to 2, 3, or 4 classes of prior migraine treatment period, patients were randomized (1:1:1) preventive medications. to monthly fremanezumab (Month 1: CM, 675 mg; EM, 225 mg; Months 2 and 3: 225 mg), quarterly fre- Methods manezumab (Month 1: 675 mg; Months 2 and 3: pla- cebo), or matched monthly placebo. Efficacy was This was an exploratory analysis of predefined sub- evaluated using information entered daily by patients groups in the international, multicenter, randomized, in an electronic headache diary throughout the treat- double-blind, placebo-controlled, parallel-group, phase ment period (7). 3b FOCUS trial (ClinicalTrials.gov identifier: NCT03308968) in patients with CM or EM who had documented inadequate response to 2 to 4 prior classes Outcomes assessed in subgroups by number of prior of migraine preventive medications (4). The study migraine preventive medication classes with design, eligibility criteria, methods, and statistical anal- inadequate response ysis methods for that trial have been reported in detail For these subgroup analyses of patients by the number previously (7) and are summarized briefly here. of classes of prior migraine preventive medications to which they had inadequate response (2, 3, or 4 classes), Patients the mean change from baseline in the monthly Patients were recruited for the FOCUS study from average number of migraine days was evaluated November 2017 through July 2018. Eligible patients during the 12-week, double-blind period after the first included adults (18–70 years of age) with a diagnosis dose of study drug (primary endpoint for the study). Pazdera et al. 1077 For these subgroup analyses, the proportion of before the first dose of study drug. For the monthly patients achieving 50% reduction in the monthly average number of migraine days during the 12 weeks average number of migraine days, mean change from after the first dose of study drug, a mean number of baseline in the average number of headache days of at migraine days was calculated for each of the 3 months, least moderate severity, and mean change from baseline and the mean of these values was then calculated. For in average days of any acute headache medication use the change from baseline in the monthly average were also evaluated during the 12-week period after the number of migraine days during the 12-week treatment first dose of fremanezumab. For the 4-week period period (primary endpoint of the study), analyses were after the first dose of fremanezumab, the mean performed using an analysis of covariance (ANCOVA) change from baseline in monthly average migraine method that included treatment, sex, region, special days, proportion of patients achieving 50% reduction group of treatment failure, migraine classification, in the monthly average number of migraine days, and and treatment-by-migraine classification interaction mean change from baseline in the monthly average as fixed effects and baseline number of migraine days number of headache days of at least moderate severity and years since onset of migraine as covariates. were also assessed in these subgroups. Statistical methods for all secondary and exploratory Additional outcomes assessed in these subgroups endpoints are summarized in Supplemental material 1. included mean change from baseline in disability All statistical tests were 2-tailed at the 0.05 level of scores based on the 6-item Headache Impact Test significance. All data listings, summaries, and statisti- (HIT-6) and Migraine Disability Assessment cal analyses were generated using SASVR software (ver- (MIDAS) and mean change from baseline in patient sion 9.4 or later of SAS Systems for Windows, SAS satisfaction, as measured by the Patient Global Institute Inc., Cary, NC, USA). Impression of Change (PGIC) during the 4 weeks The safety analysis set of the FOCUS study included after the third dose of fremanezumab. A PGIC all randomized patients who received 1 dose of study responder was defined as a patient who reported a drug. AEs were coded using the Medical Dictionary for rating of 5 to 7 (moderately better, better, or a great Regulatory Activities (MedDRA) version 18.1. deal better) on the PGIC. Adverse events (AEs), seri- ous AEs (SAEs), and AEs leading to discontinuation Data availability were evaluated in these subgroups by number of prior For the analyses described in this manuscript, anony- migraine preventive medications classes with inade- mized data will be shared upon request from any qual- quate response. ified investigator by the author investigators or Teva Branded Pharmaceutical Products R&D, Inc. Statistical analyses A sample size of 705 evaluable patients (235 per treat- Results ment group) completing the study was needed for the primary analysis for 90% power to show a difference of Of the 837 patients in the modified intention-to-treat 1.8 in migraine days (assuming a common standard analysis set, 832 patients were included in these sub- deviation of 6 days) at an alpha level of 0.05. group analyses. Of these patients, 414 patients (50%) Assuming a 12% discontinuation rate for the double- had inadequate response to 2 classes of migraine pre- blind period, 268 patients per treatment group were ventive medications, 265 (32%) to 3 classes, and 153 (18%) to 4 classes. A total of 5 patients who had inad- planned for randomization. For the FOCUS study, equate response to either 1 or >4 medication classes efficacy analyses were performed on the modified were excluded from these subgroup analyses. Patient intention-to-treat analysis set, which included all ran- baseline characteristics are presented in Table 1. domized patients who received 1 dose of study drug and had 10 days of post-baseline efficacy assessments for the primary endpoint. In these predefined sub- Monthly migraine days groups of patients by the number of classes of prior For the primary endpoint, the reduction from baseline migraine preventive medications to which they had in the monthly average number of migraine days inadequate response (2, 3, or 4 classes), the analysis during the 12-week treatment period was significantly of the change in the monthly average number of greater with both quarterly and monthly fremanezu- migraine days was part of the original statistical anal- mab compared with placebo in patients who had ysis plan; the analyses of all other endpoints within prior inadequate response to 2, 3, and 4 migraine pre- these subgroups were conducted post hoc. ventive medication classes (all p < 0.001; Figure 1a). Baseline data for the monthly average number of Among patients with inadequate response to 2 medica- migraine days were collected during the 28-day period tion classes, the least-squares mean (LSM) change from 1078 Cephalalgia 41(10) baseline (standard error [SE]) was 1.0 (0.43) with pla- fremanezumab cebo, –3.9 (0.44) with quarterly fremanezumab (LSM (5.71) 85.7 (69.0) (5.3) (2.4) (6.0) 12.5 (6.2) 65.0 (4.3) Monthly difference [LSMD] vs placebo, –2.9 [95% confidence (n ¼ 50) interval (CI): –3.83, –1.98]), and –4.7 (0.44) with 15.3 17.6 10.1 14.2 monthly fremanezumab (–3.7 [–4.63, –2.75]). For patients with inadequate response to 3 medication clas- fremanezumab ses, the LSM (SE) change from baseline was –0.7 (0.56) (4.61) 74.6 (50.2) Quarterly (3.6) (2.5) (4.5) 14.3 (5.1) 64.8 (3.8) with placebo, –4.0 (0.58) with quarterly fremanezumab (n ¼ 49) (LSMD vs placebo, –3.3 [95% CI: –4.65, –1.95]), and 9.9 16.0 17.6 14.7 –3.6 (0.58) with monthly fremanezumab (–3.0 [–4.25, –1.66]). For patients with inadequate response to 4 (5.63) 72.7 (58.9) (4.6) (2.6) (5.9) 14.0 (7.0) 65.1 (4.5) medication classes, the LSM (SE) change from baseline 4 classes (n ¼ 54) Placebo was 1.4 (0.97) with placebo, –3.9 (0.97) with quarterly 9.8 16.8 18.7 15.3 fremanezumab (LSMD vs placebo, –5.3 [95% CI: –7.38, –3.22]), and –4.0 (0.85) with monthly fremane- fremanezumab zumab (–5.4 [–7.35, –3.48]). (5.82) 66.3 (48.2) Regardless of migraine classification (CM or EM), (5.5) (2.6) (6.1) 12.8 (6.0) 63.9 (4.7) Monthly (n ¼ 98) significantly greater reductions from baseline in the 15.2 17.4 10.3 13.7 monthly average number of migraine days were observed with both fremanezumab dosing regimens fremanezumab compared with placebo over 12 weeks of treatment in (5.85) 62.2 (50.0) patients with inadequate response to 2, 3, or 4 migraine Quarterly (5.1) (2.9) (6.1) 12.6 (6.7) 64.0 (4.7) (n ¼ 85) preventive medication classes (Figures 1b and 1c). In 8.7 14.4 16.8 12.6 patients with CM, the reduction from baseline in the monthly average number of migraine days during the (6.05) 61.6 (58.1) 12 weeks after the first dose of study drug was signif- (5.0) (2.8) (5.7) 11.4 (6.0) 64.0 (4.1) 3 classes (n ¼ 82) Placebo icantly greater with fremanezumab compared with 14.2 17.8 12.4 9.0 placebo in patients with prior inadequate response SD, standard deviation; HIT-6, 6-item Headache Impact Test; MIDAS, Migraine Disability Assessment. to 2 medication classes (LSMD vs placebo: quarterly fremanezumab, –3.5 [95% CI: –5.00, –1.93]; monthly fremanezumab fremanezumab, –4.9 [–6.45, –3.34]; both p < 0.001), (5.09) 51.2 (42.2) (4.2) (2.8) (5.3) 11.5 (5.8) 63.7 (4.2) (n ¼ 133) Monthly 3 medication classes (quarterly fremanezumab, –3.3 [–5.02, –1.62]; monthly fremanezumab, –3.1 [–4.74, 12.8 16.2 11.4 9.0 –1.45]; both p < 0.001), and 4 medication classes (quarterly fremanezumab, –2.8 [–4.78, –0.76]; monthly fremanezumab fremanezumab, –3.2 [–5.24, –1.10]; both p  0.007; (5.66) 55.7 (46.7) Quarterly (5.3) (2.6) (5.9) 12.4 (6.3) 64.2 (4.2) (n ¼ 140) Figure 1b). In patients with EM, the reduction from baseline in monthly average migraine days during the 9.5 13.3 17.1 11.5 12 weeks after the first dose of study drug was signif- icantly greater with fremanezumab versus placebo in (6.17) 57.0 (55.6) (5.6) (2.5) (5.8) 12.2 (6.0) 63.8 (5.6) (n ¼ 141) patients who had prior inadequate response to 2 med- 2 classes Placebo ication classes (LSMD vs placebo [95% CI]: quarterly 13.6 17.6 12.2 9.0 fremanezumab, –2.3 [–3.29, –1.37]; monthly fremane- zumab, –2.5 [–3.50, –1.53]; both p < 0.001), 3 medica- Monthly days of any acute headache tion classes (quarterly fremanezumab, –3.4 [–5.26, Monthly migraine days, mean (SD) Monthly headache days of at least Table 1. Baseline characteristics. moderate severity, mean (SD) –1.44]; monthly fremanezumab, –2.4 [–4.28, –0.57]; medication use, mean (SD) both p  0.011), and 4 medication classes (quarterly MIDAS score, mean (SD) HIT-6 score, mean (SD) fremanezumab, –7.4 [–10.36, –4.37]; monthly fremane- zumab, –7.8 [–10.41, –5.11]; both p < 0.001; Figure 1c). Episodic migraine Chronic migraine Total population Reductions from baseline in the monthly average number of migraine days at 4 weeks after the first dose of study drug were also significantly greater with Disability fremanezumab versus placebo in patients who had prior inadequate response to 2, 3, and 4 migraine pre- ventive medication classes (all p < 0.001, Figure 2). Pazdera et al. 1079 (a) 2 classes 3 classes 4 classes (b) 2 classes 3 classes 4 classes 4 3.7 1.5 1.4 2 0.5 LSM change from baseline, days LSM change from baseline, days –0.5 0 –0.7 –0.4 –1.5 –1.0 –2 –1.5 –2.5 –2.8b –4 –3.7a –3.7a –3.8a –4.0a –4.1a –3.5 a –3.6 –6 –3.9a –4.0a –3.9a –4.0a –4.5 –4.7a –8 –5.5 Placebo Quarterly fremanezumab Monthly fremanezumab Placebo Quarterly fremanezumab Monthly fremanezumab (c) 2 classes 3 classes 4 classes –0.5 –0.7 –0.7 –1.0 –1.5 LSM change from baseline, days –2.5 –3.5 –3.4c –3.8d –4.0a –4.5 –4.2a –4.3a –5.5 –5.6a –6.5 Placebo Quarterly fremanezumab Monthly fremanezumab Figure 1. Change from baseline in monthly average migraine days over 12 weeks of treatment. (a) All patients (b) EM patients (c) CM patients. LSM, least-squares mean; EM, episodic migraine; CM, chronic migraine. a p < 0.001 versus placebo. bp ¼ 0.011 versus placebo. cp ¼ 0.007 versus placebo. dp ¼ 0.003 versus placebo. Responder rates 2 classes 3 classes 4 classes The proportion of patients who achieved 50% reduc- 2 1.5 tion in the monthly average number of migraine days 1 LSM change from baseline, days during the 12 weeks of double-blind treatment was sig- 0 nificantly higher with fremanezumab compared with –1 –0.8 –0.7 placebo in patients who had prior inadequate response –2 to 2, 3, and 4 migraine preventive medication classes –3 (all p  0.002; Figure 3a). With quarterly fremanezu- –4 –3.7a mab, monthly fremanezumab, and placebo, respective- –4.2a –4.5a –4.1a –4.0a –5 –4.7a ly, 50% reduction in the monthly average number of –6 migraine days was achieved in 39%, 41%, and 11% of Placebo Quarterly fremanezumab Monthly fremanezumab patients with inadequate response to 2 medication clas- ses; 32%, 28%, and 7% of patients with inadequate Figure 2. Change from baseline in the monthly average number response to 3 medication classes; and 27%, 32%, and of migraine days over the first 4 weeks of treatment. 4% of patients with inadequate response to 4 medica- LSM, least-squares mean. tion classes. a p < 0.0001 versus placebo. The proportion of patients who achieved 50% reduction in the monthly average number of migraine days during the 4 weeks after the first dose of study medication classes (all p  0.003; Figure 3b). With drug was also significantly higher with fremanezumab quarterly fremanezumab, monthly fremanezumab, compared with placebo in patients who had prior inad- and placebo, respectively, 50% reduction in monthly equate response to 2, 3, and 4 migraine preventive average migraine days was achieved in 44%, 39%, and 1080 Cephalalgia 41(10) (a) p value Failed 2 prior classesa 5.10 Quarterly fremanezumab < 0.001 5.56 Monthly fremanezumab < 0.001 Failed 3 prior classesa 6.67 Quarterly fremanezumab < 0.001 5.18 < 0.001 Monthly fremanezumab Failed 4 prior classesa 12.43 0.002 Quarterly fremanezumab 62.21 17.10 < 0.001 Monthly fremanezumab 84.97 0 2 4 6 8 10 12 14 16 18 20 OR (95% CI) for fremanezumab vs placebo (b) p value Failed 2 prior classesa 5.48 Quarterly fremanezumab < 0.001 4.54 Monthly fremanezumab < 0.001 Failed 3 prior classesa 6.64 Quarterly fremanezumab < 0.001 5.97 < 0.001 Monthly fremanezumab Failed 4 prior classesa 8.05 0.003 Quarterly fremanezumab 31.32 8.98 0.002 Monthly fremanezumab 34.73 0 2 4 6 8 10 12 14 16 18 20 OR (95% CI) for fremanezumab vs placebo Figure 3. ORs (fremanezumab vs placebo) for achieving 50% reduction in monthly migraine days. (a) Over 12 weeks of treatment (b) At 4 weeks of treatment. OR, odds ratio; CI, confidence interval. a Classes of migraine preventive medications. 13% of patients with inadequate response to 2 medica- period with fremanezumab compared with placebo tion classes; 34%, 33%, and 9% of patients with inad- among patients with inadequate response to 2 prior equate response to 3 medication classes; and 29%, migraine preventive medication classes (quarterly fre- 32%, and 6% of patients with inadequate response to manezumab: 33% [odds ratio (OR), 4.27 (95% CI: 4 medication classes. 1.75, 10.42); p ¼ 0.001] and monthly fremanezumab: Significantly higher proportions of patients with 36% [5.20 (2.13, 12.69); p < 0.001] vs placebo: 11%), CM achieved 50% reduction in the monthly average 3 medication classes (23% [4.52 (1.20, 17.09); p ¼ 0.026] number of migraine days during the 12-week treatment and 26% [5.04 (1.37, 18.50); p ¼ 0.015], respectively, vs Pazdera et al. 1081 6%), or 4 medication classes (21% [5.21 (1.02, 26.71); reduction from baseline in the HIT-6 score during p¼0.048] and 20% [6.51 (1.22, 34.80); p ¼ 0.028], the 4 weeks after the third dose of study drug was respectively, vs 5%). Similarly, in patients with EM, significantly greater with fremanezumab compared higher proportions of patients achieved 50% reduc- with placebo (all p  0.014; Table 3). The reduction tion in monthly average migraine days during the 12- from baseline in the MIDAS score during the 4 week treatment period with fremanezumab compared weeks after the third dose of study drug was also sig- with placebo among patients with inadequate response nificantly greater with monthly fremanezumab com- to 2 medication classes (quarterly fremanezumab: 46% pared with placebo in patients who had prior [OR, 6.05 (95% CI: 2.51, 14.56); p < 0.001] and month- inadequate response to 2, 3, or 4 migraine preventive ly fremanezumab: 45% [6.01 (2.46, 14.69); p < 0.001] vs medication classes (all p  0.010) and with quarterly placebo: 12%), 3 medication classes (52% [11.69 (2.71, fremanezumab compared with placebo in patients 50.51); p ¼ 0.001] and 30% [4.02 (0.95, 17.09); who had prior inadequate response to 4 migraine pre- p ¼ 0.059] vs 9%), or 4 medication classes (50% and ventive medication classes (Table 3). 60%, respectively, vs 0%; sample size was too small The proportion of responders on the PGIC scale for logistic regression model). was significantly higher with both quarterly and monthly fremanezumab compared with placebo in Headache days of at least moderate severity patients with prior inadequate response to 2, 3, or 4 In subgroups of patients with prior inadequate classes of migraine preventive medications (p < 0.001; response to 2, 3, and 4 migraine preventive medication Figure 4). classes, reductions from baseline in the monthly aver- age number of headache days of at least moderate Tolerability severity during the 12 weeks of double-blind treatment Incidences of AEs, AEs leading to discontinuation, and were significantly greater with fremanezumab SAEs were similar with both quarterly and monthly compared with placebo (all p < 0.001; Table 2). fremanezumab and placebo in patients with prior inad- Reductions from baseline in the monthly average equate response to 2, 3, and 4 classes of migraine number of headache days of at least moderate severity preventive medications (Table 4). The most common at 4 weeks after the first dose of study drug were also AEs across subgroups included injection-site erythema significantly greater with monthly fremanezumab com- (range, 4%–10%) and injection-site induration (range, pared with placebo in patients who had prior inade- 2%–10%). The proportion of patients discontinuing quate response to 2, 3, and 4 migraine preventive the study due to AEs was low (range, <1%–3%). medication classes (all p < 0.001; Table 2). No significant treatment-by-subgroup interactions were observed for any outcome, as shown in the Acute headache medication use table in Supplemental material 2 (p interaction > 0.20 The reduction in the monthly average number of days for all). of any acute headache medication use was significantly higher with fremanezumab compared with placebo Discussion during the 12 weeks after the first dose of study In these preplanned subgroup analyses, fremanezumab drug in patients who had prior inadequate response offered significant efficacy over placebo in both CM to 2, 3, and 4 migraine preventive medication classes and EM patients regardless of the number of migraine (all p < 0.001; Table 2). The reduction in the monthly preventive medication classes with inadequate response average number of days of any acute headache medi- prior to this study. Reductions in migraine days and cation use was also significantly higher with fremane- clinically meaningful response rates were achieved as zumab compared with placebo at 4 weeks after the first early as 4 weeks after the first dose of study drug. dose of study drug in patients who had prior inade- quate response to 2, 3, and 4 migraine preventive med- Approximately 30% to 40% of patients who had a ication classes (all p < 0.001; Table 2). documented inadequate response to 2, 3 or 4 migraine preventive treatment classes achieved 50% reduction in monthly migraine days with 4 and 12 weeks of treat- Disability (HIT-6 and MIDAS) and Patient ment. A greater effect size was observed with increasing satisfaction (PGIC) number of prior treatment failures and was generally Regardless of the number of prior migraine preventive demonstrated across all endpoints, with the most nota- medication classes with inadequate response, the ble increase in effect size for patients with inadequate 1082 Table 2. Change from baseline in monthly headache days of at least moderate severity and days of any acute headache medication use over 4 and 12 weeks. 2 classes 3 classes 4 classes Quarterly Monthly Quarterly Monthly Quarterly Monthly Placebo fremanezumab fremanezumab Placebo fremanezumab fremanezumab Placebo fremanezumab fremanezumab (n ¼ 141) (n ¼ 140) (n ¼ 133) (n ¼ 82) (n ¼ 85) (n ¼ 98) (n ¼ 54) (n ¼ 49) (n ¼ 50) 12 weeks Monthly headache days of at least moderate severity LSM change from –1.2 (0.42) 3.9 (0.42) –4.8 (0.42) –0.3 (0.57) –3.9 (0.59) –3.5 (0.59) 0.6 (1.02) –4.7 (1.01) –4.9 (0.88) baseline (SE) LSMD versus placebo –2.7 (–3.64, –1.86) –3.6 (–4.47, –2.65) –3.6 (–4.96, –2.21) –3.2 (–4.56, –1.93) –5.2 (–7.42, –3.07) –5.4 (–7.47, –3.42) (95% CI) p value <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 Monthly days of any acute headache medication use LSM change from –1.2 (0.43) –4.0 (0.44) –4.3 (0.44) –0.4 (0.49) –3.7 (0.51) –3.5 (0.51) 1.2 (0.94) –3.6 (0.93) –4.0 (0.82) baseline (SE) LSMD versus placebo –2.9 (–3.79, –1.94) –3.2 (–4.12, –2.23) –3.2 (–4.44, –2.06) –3.0 (–4.18, –1.89) –4.8 (–6.80, –2.81) –5.2 (–7.05, –3.33) (95% CI) p value <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 4 weeks Monthly headache days of at least moderate severity LSM change from –1.0 (0.43) –4.1 (0.43) –4.7 (0.43) –0.2 (0.56) –4.1 (0.58) –4.0 (0.58) 0.7 (1.03) –5.3 (1.03) –5.2 (0.90) baseline (SE) LSMD versus placebo –3.2 (–4.09, –2.21) –3.8 (–4.71, –2.80) –4.0 (–5.34, –2.60) –3.8 (–5.11, –2.49) –6.0 (–8.30, –3.78) –5.9 (–8.02, –3.81) (95% CI) p value <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 Monthly days of any acute headache medication use LSM change from –0.9 (0.43) –4.2 (0.44) –4.4 (0.44) –0.5 (0.49) –4.0 (0.51) –3.9 (0.51) 1.1 (0.91) –4.3 (0.91) –4.2 (0.80) baseline (SE) LSMD versus placebo –3.2 (–4.18, –2.27) –3.5 (–4.42, –2.48) –3.5 (–4.66, –2.24) –3.3 (–4.51, –2.19) –5.4 (–7.41, –3.41) –5.3 (–7.19, –3.45) (95% CI) p value <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 LSM, least-squares mean; SE, standard error; LSMD, least-squares mean difference; CI, confidence interval. Cephalalgia 41(10) Pazdera et al. 1083 response to 4 classes of migraine preventive –31.7 (–54.07, –9.37) –29.9 (–51.12, –8.70) –6.8 (–9.25, –4.43) medications. fremanezumab Along with the clinically meaningful reductions in –23.2 (9.12) –6.2 (1.04) migraine days, both fremanezumab dosing regimens Monthly (n ¼ 50) <0.001 were shown to reduce migraine- and headache-related 0.006 HIT-6, 6-item Headache Impact Test; LSM, least-squares mean; SE, standard error; LSMD, least-squares mean difference; CI, confidence interval; MIDAS, Migraine Disability Assessment. disability versus placebo in patients who had docu- mented inadequate response to 2, 3, or 4 migraine preventive medication classes. Regardless of the –5.6 (–8.16, –3.03) number of prior migraine preventive medication classes fremanezumab 6.7 (10.59) –25.0 (10.43) with inadequate response, the majority of patients –5.0 (1.18) Quarterly reported improvements with fremanezumab treatment (n ¼ 49) <0.001 0.006 on the PGIC in this study. Many of the currently available medications used for migraine prevention were not intended or 0.6 (1.19) 4 classes (n ¼ 54) Placebo designed for the treatment of migraine and have limited or moderate efficacy and poor tolerability (1,2). Few options exist for patients who do not respond to these –9.8 (–22.68, 3.08) –16.2 (–28.51, –3.90) preventive therapies. As shown in a retrospective US –2.8 (–4.95, –0.57) –3.2 (–5.28, –1.11) claims analysis, 86% of patients discontinued initial fremanezumab –25.3 (5.56) preventive medication by 1 year, and in patients who –5.8 (0.94) switched migraine preventive medications, this number Monthly (n ¼ 98) 0.003 0.010 increased to 90% by the third medication (3). However, with better clinical outcomes for patients taking migraine preventive medications, persistence may Table 3. Change from baseline to 4 weeks after the third dose of study drug in disability outcomes. improve. fremanezumab –9.1 (5.34) –18.9 (5.69) Limited data are available on the efficacy and toler- –2.6 (0.90) –5.4 (0.96) Quarterly (n ¼ 85) ability of migraine preventive treatments by number of 0.014 0.14 prior migraine preventive treatments to which patients have had inadequate response, in particular in patients who have experienced inadequate response to multiple 3 classes (n ¼ 82) Placebo prior preventive treatment classes. Previous subgroup analyses of randomized, double-blind, placebo- controlled studies of erenumab in CM (9) and EM –15.5 (–24.47, –6.46) –3.6 (–5.32, –1.93) (10) showed consistent efficacy for erenumab based on reductions in monthly migraine days, responder fremanezumab –21.5 (4.11) rates, and days of acute medication use in patients –6.4 (0.78) (n ¼ 133) with prior failure of 1 and 2 migraine preventive Monthly <0.001 <0.001 medication categories. However, both studies excluded patients with no therapeutic response to 3 prior –8.7 (–17.47, 0.15) –2.5 (–4.21, –0.88) migraine preventive medications (9,10). The current fremanezumab study demonstrated consistent efficacy and tolerability –6.1 (4.10) –14.7 (4.15) –2.7 (0.77) –5.3 (0.78) for fremanezumab, even in potentially more difficult- Quarterly (n ¼ 140) to-treat patients with inadequate response to 3 or even 0.003 0.054 4 prior migraine preventive medications. There are concerns in the literature that, with an (n ¼ 141) increasing number of prior preventive treatment fail- 2 classes Placebo ures, patients with migraine may not respond to mono- therapy (11). The results of this study suggest that, even for patients with inadequate response to 4 prior LSMD versus placebo LSMD versus placebo migraine preventive medication classes, fremanezumab LSM change from LSM change from baseline (SE) baseline (SE) as monotherapy is still effective. Furthermore, the MIDAS score (95% CI) (95% CI) HIT-6 score patients in this population had additional challenges, including severe disability due to migraine (MIDAS) p value p value (12) and severe headache impact on daily functioning (HIT-6) (13). This severe disability, along with the 1084 Cephalalgia 41(10) 2 classes 3 classes 4 classes 100 90 80 71a 70 64a 58a 58a Patients, % 60 52a 51 a 50 40 35 30 27 19 20 10 0 Placebo Quarterly fremanezumab Monthly fremanezumab Figure 4. Proportion of PGIC responders. PGIC, Patient Global Impression of Change. PGIC responder was defined as a patient who reported a rating of 5 to 7 (moderately better, better, or a great deal better) on the PGIC. a p < 0.001 versus placebo. inadequate response to multiple prior migraine preven- number of patients in the subgroup with inadequate tive medications, highlights the disease severity in this response to 4 prior migraine preventive medications. population. This likely contributed to high point estimates and In the current study, placebo responses were low and wide CIs seen in the responder analyses and some decreased with increasing number of classes of prior other endpoints. However, results were generally simi- preventive medications with inadequate response. In lar for patients treated with fremanezumab across all other studies that evaluated monoclonal antibodies subgroups, suggesting a consistent effect of fremanezu- that target the CGRP pathway in CM, mean reduc- mab regardless of prior treatment experience. tions from baseline in monthly migraine days ranged Additionally, while the subgroups were defined from –2.7 to –4.2 days for patients receiving placebo, a priori and the analysis of the change in migraine whereas the placebo response for the CM population in days in each subgroup was part of the original this study was –0.7 to –1.0 days for a similar endpoint statistical analysis plan, the analyses of the other end- (14,15). This trend toward lower placebo response in points within these subgroups were conducted patients with prior inadequate response to preventive post hoc. As the results in each subgroup for all end- treatment was also observed in the subgroup analyses points were similar to those in the overall of patients with 1 and 2 prior preventive treatment category failures in the erenumab studies in CM and population and other subgroups and all assessments EM. The low placebo response in this study, even in were performed a priori, the impact of this limitation patients with CM, further supports the selection of this is likely small. population of patients with difficult-to-treat migraine. Quarterly and monthly fremanezumab were well tol- In the current study, fremanezumab reduced monthly erated, and the results of these subgroup analyses indi- migraine days by 3 to 5 days in all patients and by cate that fremanezumab is consistently effective as a 3 to 4 days in with patients with CM compared with preventive treatment in patients with documented inad- placebo across 2, 3, or 4 classes of prior treatment equate response to 2, 3, and 4 classes of migraine pre- failures. ventive medications. These results may be relevant for The results reported here may be subject to certain the clinical management of patients with migraine who limitations. For this analysis, the number of patients have a history of inadequate response to multiple prior included in each subgroup differed, with a smaller classes of migraine preventive medications. Table 4. AEs during double-blind treatment (occurring in 2% of patients in any treatment groupa). 2 classes 3 classes 4 classes Quarterly Monthly Quarterly Monthly Quarterly Monthly Pazdera et al. Placebo fremanezumab fremanezumab Placebo fremanezumab fremanezumab Placebo fremanezumab fremanezumab AEs, n (%) (n ¼ 141) (n ¼ 140) (n ¼ 134) (n ¼ 81) (n ¼ 85) (n ¼ 99) (n ¼ 54) (n ¼ 49) (n ¼ 50) 1 AE 61 (43%) 67 (48%) 58 (43%) 39 (48%) 51 (60%) 47 (47%) 34 (63%) 31 (63%) 23 (46%) 1 SAE 4 (3%) 1 (<1%) 2 (1%) 0 0 2 (2%) 0 1 (2%) 0 AEs leading to discontinuation 3 (2%) 1 (<1%) 1 (<1%) 0 0 3 (3%) 0 0 0 AEs Injection-site erythema 8 (6%) 9 (6%) 6 (4%) 5 (6%) 5 (6%) 5 (5%) 2 (4%) 5 (10%) 5 (10%) Injection-site induration 5 (4%) 4 (3%) 7 (5%) 3 (4%) 3 (4%) 2 (2%) 4 (7%) 5 (10%) 4 (8%) Upper respiratory tract infection 1 (<1%) 1 (<1%) 7 (5%) 2 (2%) 3 (4%) 2 (2%) 0 0 0 Nasopharyngitis 6 (4%) 4 (3%) 3 (2%) 4 (5%) 5 (6%) 3 (3%) 1 (2%) 4 (8%) 1 (2%) Influenza 1 (<1%) 1 (<1%) 4 (3%) 0 0 1 (1%) 1 (2%) 1 (2%) 1 (2%) Fatigue 0 5 (4%) 4 (3%) 2 (2%) 2 (2%) 3 (3%) 1 (2%) 1 (2%) 2 (4%) Injection-site pain 4 (3%) 5 (4%) 3 (2%) 2 (2%) 2 (2%) 3 (3%) 2 (4%) 4 (8%) 3 (6%) Injection-site paresthesia 1 (<1%) 1 (<1%) 2 (1%) 2 (2%) 1 (1%) 0 0 2 (4%) 1 (2%) Injection-site pruritis 2 (1%) 0 2 (1%) 0 2 (2%) 2 (2%) 1 (2%) 1 (2%) 1 (2%) Injection-site bruising 1 (<1%) 0 1 (<1%) 1 (1%) 0 4 (4%) 0 2 (4%) 0 Injection-site rash 0 0 0 2 (2%) 1 (1%) 2 (2%) 0 2 (4%) 1 (2%) Injection-site warmth 0 0 0 0 0 3 (3%) 0 1 (2%) 0 Injection-site discoloration 0 0 0 0 0 0 0 2 (4%) 0 Gastroenteritis 3 (2%) 0 1 (<1%) 2 (2%) 2 (2%) 1 (1%) 2 (4%) 1 (2%) 1 (2%) Cystitis 3 (2%) 0 0 0 0 0 2 (4%) 0 0 Urinary tract infection 2 (1%) 3 (2%) 2 (1%) 0 0 0 3 (6%) 0 1 (2%) Asthenia 2 (1%) 0 1 (<1%) 0 0 1 (1%) 1 (2%) 1 (2%) 1 (2%) Chest discomfort 1 (<1%) 0 0 0 0 1 (1%) 0 0 0 Vertigo 4 (3%) 1 (<1%) 0 0 1 (1%) 1 (1%) 0 0 0 Arthralgia 3 (2%) 0 0 0 1 (1%) 2 (2%) 0 1 (2%) 0 Headache 2 (1%) 0 1 (<1%) 0 2 (2%) 0 0 0 0 Paresthesia 4 (3%) 1 (<1%) 0 1 (1%) 0 0 0 0 0 Sciatica 3 (2%) 0 0 0 0 0 1 (2%) 0 0 Insomnia 1 (<1%) 1 (<1%) 4 (3%) 1 (1%) 2 (2%) 1 (1%) 0 3 (6%) 2 (4%) Palpitations 0 0 1 (1%) 1 (1%) 0 1 (1%) 2 (4%) 1 (2%) 0 Diarrhea 0 2 (1%) 0 2 (2%) 4 (5%) 2 (2%) 2 (4%) 1 (2%) 0 Nausea 1 (<1%) 1 (<1%) 1 (<1%) 2 (2%) 2 (2%) 1 (1%) 3 (6%) 1 (2%) 0 Abdominal pain 2 (1%) 0 0 3 (4%) 2 (2%) 0 0 1 (2%) 0 Back pain 3 (2%) 1 (<1%) 0 2 (2%) 2 (2%) 2 (2%) 0 2 (4%) 0 Neck pain 0 0 0 0 1 (1%) 2 (2%) 0 1 (2%) 1 (2%) Pain in extremity 2 (1%) 1 (<1%) 1 (<1%) 1 (1%) 1 (1%) 2 (2%) 0 0 0 (continued) 1085 1086 Table 4. Continued. 2 classes 3 classes 4 classes Quarterly Monthly Quarterly Monthly Quarterly Monthly Placebo fremanezumab fremanezumab Placebo fremanezumab fremanezumab Placebo fremanezumab fremanezumab AEs, n (%) (n ¼ 141) (n ¼ 140) (n ¼ 134) (n ¼ 81) (n ¼ 85) (n ¼ 99) (n ¼ 54) (n ¼ 49) (n ¼ 50) Dizziness 1 (<1%) 2 (1%) 2 (1%) 2 (2%) 2 (2%) 1 (1%) 0 1 (2%) 1 (2%) Anxiety 0 1 (<1%) 2 (1%) 0 2 (2%) 0 0 0 0 Constipation 1 (<1%) 3 (2%) 0 1 (1%) 2 (2%) 0 0 2 (4%) 1 (2%) Vomiting 1 (<1%) 0 0 0 2 (2%) 0 0 0 0 Hematuria 0 0 0 0 1 (1%) 2 (2%) 1 (2%) 0 0 Pyrexia 2 (2%) 1 (<1%) 0 2 (2%) 0 1 (1%) 0 0 1 (2%) Sinusitis 1 (<1%) 1 (<1%) 0 2 (2%) 0 1 (1%) 1 (2%) 0 1 (2%) Abdominal pain upper 0 1 (<1%) 1 (<1%) 0 1 (1%) 1 (1%) 0 2 (4%) 0 Hypertension 0 1 (<1%) 1 (<1%) 1 (1%) 2 (2%) 0 1 (2%) 0 0 Influenza-like illness 1 (<1%) 0 0 2 (2%) 0 1 (1%) 1 (2%) 1 (2%) 0 Gastritis 0 1 (<1%) 1 (<1%) 0 0 0 2 (4%) 0 0 Migraine 3 (2%) 1 (<1%) 0 1 (1%) 0 2 (2%) 5 (9%) 1 (2%) 1 (2%) Weight increased 0 1 (<1%) 2 (1%) 0 0 0 1 (2%) 2 (4%) 0 Rash 2 (1%) 0 0 0 0 0 0 0 3 (6%) Hematoma 0 0 0 0 0 0 0 2 (4%) 0 Blood glucose increased 0 0 0 0 2 (2%) 1 (1%) 0 0 0 AE, adverse event; SAE, serious adverse event. a Additional AEs reported by 2% (n  1 per treatment group) of patients in the 4 classes subgroup: anemia, neutropenia, increased tendency to bruise, Ehlers-Danlos syndrome, ear pain, allergic conjunctivitis, ocular hyperanemia, breath odor, gingival pain, irritable bowel, abdominal discomfort, eructation, gastroesophageal reflux disease, esophageal ulcer, injection-site discomfort, injection-site hematoma, injection-site hemorrhage, injection-site swelling, injection-site urticarial, peripheral edema, pyrexia, thirst, allergy to arthropod sting, bacterial vaginosis, herpes zoster, viral gastroenteritis, oral herpes, pharyngitis, tracheitis, head injury, heat stroke, arthropod sting, epicondylitis, bordetella test positive, international normalized ratio increased, food intolerance, decreased appetite, limb discomfort, muscle spasms, myalgia, rotator cuff syndrome, muscular weakness, musculoskeletal pain, postural dizziness, somnolence, nervousness, sleep disorder, affect lability, affective disorder, abnormal urine odor, ovarian cyst, menopausal symptoms, cough, oropharyngeal pain, epistaxis, atopic dermatitis, dry skin, ecchymosis, eczema, erythema, pruritic rash, hyperhidrosis, peripheral venous disease, hypotension. Cephalalgia 41(10) Pazdera et al. 1087 Clinical Implications • There are few treatment options available for patients who do not respond to standard migraine preventive medications • The results of these subgroup analyses showed the efficacy of fremanezumab by number of prior migraine preventive treatments in patients who experienced inadequate response to multiple prior preventive treat- ment classes • The effect size for fremanezumab generally increased with number of prior treatment failures, which may be relevant to the clinical management of patients with migraine and a history of inadequate response to multiple prior classes of migraine preventive medications Acknowledgments 3. Hepp Z, Dodick DW, Varon SF, et al. Persistence and Editorial assistance was provided by Alyssa Nguyen, switching patterns of oral migraine prophylactic medica- PharmD, of MedErgy, which was in accordance with Good tions among patients with chronic migraine: a Publication Practice (GPP3) guidelines and funded by Teva retrospective claims analysis. Cephalalgia 2017; 37: Pharmaceuticals. 470–485. 4. Bigal ME and Lipton RB. Overuse of acute migraine medications and migraine chronification. Curr Pain Declaration of conflicting interests Headache Rep 2009; 13: 301–307. The authors declared the following potential conflicts of 5. Messali A, Sanderson JC, Blumenfeld AM, et al. Direct interest with respect to the research, authorship, and/or pub- and indirect costs of chronic and episodic migraine in the lication of this article: LP declares no conflict of interest. United States: a web-based survey. Headache 2016; 56: JMC, XN, and VRC are employees of Teva 306–322. Pharmaceuticals. RY is a former employee of Teva 6. Martelletti P, Schwedt TJ, Lanteri-Minet M, et al. My Pharmaceuticals. PPR has received honoraria as a consultant Migraine Voice survey: a global study of disease burden and speaker for Allergan, Almirall, Biohaven, Chiesi, Eli among individuals with migraine for whom Lilly, Medscape, Neurodiem, Novartis, and Teva preventive treatments have failed. J Headache Pain Pharmaceuticals. Her research group has received research 2018; 19: 115. grants from AGAUR, Allergan, la Caixa Foundation, FP7, 7. Ferrari MD, Diener HC, Ning X, et al. Fremanezumab Instituto de Salud Carlos III, Migraine Research Foundation, versus placebo for migraine prevention in patients with MICINN, and PERIS; and has received funding in the last documented failure to up to four migraine preventive 5 years for clinical trials from Alder, electroCore, Eli Lilly, medication classes (FOCUS): a randomised, double- Novartis, and Teva Pharmaceuticals. She does not own blind, placebo-controlled, phase 3b trial. Lancet 2019; stocks from any pharmaceutical company. 394: 1030–1040. 8. Headache Classification Committee of the International Funding Headache Society (IHS). The International Classification The authors disclosed receipt of the following financial sup- of Headache Disorders, 3rd edition. Cephalalgia 2018; 38: port for the research, authorship, and/or publication of this 1–211. article: This study and these analyses were funded by Teva 9. Ashina M, Tepper S, Brandes JL, et al. Efficacy Pharmaceuticals. and safety of erenumab (AMG334) in chronic migraine patients with prior preventive treatment ORCID iD failure: a subgroup analysis of a randomized, double- Ladislav Pazdera https://orcid.org/0000-0003-0053-7003 blind, placebo-controlled study. Cephalalgia 2018; 38: 1611–1621. 10. Goadsby PJ, Paemeleire K, Broessner G, et al. Efficacy References and safety of erenumab (AMG334) in episodic migraine 1. American Headache Society. The American Headache patients with prior preventive treatment failure: a sub- Society position statement on integrating new migraine group analysis of a randomized, double-blind, placebo- treatments into clinical practice. Headache 2019; 59: controlled study. Cephalalgia 2019; 39: 817–826. 1–18. 11. Krymchantowski AV and Bigal ME. Polytherapy in the 2. Reuter U. A review of monoclonal antibody therapies preventive and acute treatment of migraine: fundamen- and other preventative treatments in migraine. tals for changing the approach. Expert Rev Neurother Headache 2018; 58: 48–59. 2006; 6: 283–289. 1088 Cephalalgia 41(10) 12. Stewart WF, Lipton RB, Dowson AJ, et al. Development 14. Tepper S, Ashina M, Reuter U, et al. Safety and and testing of the Migraine Disability Assessment efficacy of erenumab for preventive treatment of chronic (MIDAS) questionnaire to assess headache-related migraine: a randomised, double-blind, placebo- disability. Neurology 2001; 56: S20–S28. controlled phase 2 trial. Lancet Neurol 2017; 16: 425–434. 13. Yang M, Rendas-Baum R, Varon SF, et al. Validation 15. Detke H, Goadsby PJ and Wang S. Galcanezumab of the Headache Impact Test (HIT-6TM) across in chronic migraine: the randomized, double-blind, episodic and chronic migraine. Cephalalgia 2011; 31: placebo-controlled REGAIN study. Neurology 2018; 91: 357–367. e2211–e2221.