Efficacy and safety of eptinezumab in patients with chronic migraine

Standard protocol approvals, registrations, and patient consents

The study was approved by the independent ethics committee or institutional review board for each study site. All clinical work was conducted in compliance with current Good Clinical Practices as referenced in the International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use guidelines, local regulatory requirements, and the principles of the Declaration of Helsinki. All patients enrolled in the study provided written informed consent before their participation. All clinical sites were required to sign a study-specific site blinding plan. This study is registered on ClinicalTrials.gov (NCT02974153).

Study design and patients

This phase 3, double-blind, randomized, placebo-controlled, parallel-group, efficacy and safety study was performed at 128 sites in 13 countries (United States, Spain, Ukraine, Russian Federation, United Kingdom, Republic of Georgia, Hungary, Italy, Slovakia, Germany, Czech Republic, Denmark, and Belgium) during the period of November 30, 2016, to April 20, 2018.

Adults 18 to 65 years of age (inclusive) with a diagnosis of migraine at or before 50 years of age were eligible for participation if they had a history of CM for ≥12 months before screening, completed the headache electronic diary (eDiary) on ≥24 of the 28 days after screening visit and before randomization (the screening period), and experienced ≥15 to ≤26 headache days and ≥8 migraine days during the 28-day screening period.² Patients taking prescription or over-the-counter medication for acute or preventive treatment of migraine were eligible only if the medications had been prescribed or recommended by a health care professional; migraine preventive medication use had to be stable for ≥3 months before screening. Hormonal therapy (e.g., contraceptive, hormone replacement) was also permitted if it was stable and ongoing ≥3 months before screening. Patients using barbiturates or prescription opioids ≤4 d/mo were eligible for participation if use was stable for ≥2 months before screening, and this restriction was maintained through week 24 of the study. Other medications for the treatment of acute migraine such as triptans, nonsteroidal anti-inflammatory drugs, and simple analgesics were not restricted. Patients with CM and medication-overuse headache with the exception of the overuse of barbiturates or opioids were eligible for inclusion. The study steering committee wished to include patients using opioids and barbiturates because more than one-third of people in the United States who have migraine use these medications as acute prescription treatments. The study steering committee chose to limit opioid and barbiturate use because high use often makes the patient refractory to preventive treatments. In pursuit of a balance between generalizability and enrolling the target population, patients using opioids or barbiturates ≥5 d/mo were excluded.

Individuals were excluded from participation if they had a confounding pain disorder or clinically significant pain syndromes; uncontrolled or untreated psychiatric conditions; acute or active temporomandibular disorders; history or diagnosis of a headache or migraine disorders that did not meet the International Classification of Headache Disorders, 3rd edition (ICHD-3) beta version (2013) Section 1.3 criteria for CM; present or previous malignancies (except history of squamous or basal cell carcinoma with excision for cure or breast or cervical cancer ≥10 years since diagnosis/treatment without evidence of recurrence); any active, progressive, or unstable cardiovascular, neurologic, or autoimmune disorder; newly diagnosed or uncontrolled hypertension (mild primary hypertension that was well controlled for ≥6 months before screening was allowed); history or evidence of substance abuse or dependence; clinically significant abnormal ECG findings; a concurrent medical condition or laboratory abnormality during the screening period or before dosing on day 0; body mass index ≥39 kg/m²; or recent or planned surgery requiring general anesthesia within 8 weeks before screening or during the duration of the study. Also excluded were patients who had received any experimental, unregistered therapy within 30 days or 5 plasma half-lives before screening or who had used any prohibited devices, neuromodulation, neurostimulation, or injectable therapy within 2 months before screening or during the screening period; botulinum toxin (any type) for migraine or for any other medical/cosmetic reasons requiring injections within 4 months before screening or during the screening period; monoamine oxidase inhibitors, ketamine, methylergonovine, or nimesulide within 3 months before screening or during the screening period; any monoclonal antibody treatment within 6 months of screening; or eptinezumab or any monoclonal antibody targeting the CGRP pathway. Women who were pregnant, breastfeeding, or planning to become pregnant during the study were excluded from participation, as were patients participating in any other clinical study or who were positive for HIV, hepatitis B surface antigen, or hepatitis C.

Women of childbearing potential and men with partners of childbearing potential agreed to use adequate contraception throughout study participation and for ≥6 months after the last dose of study drug.

Study procedures

Patients used an eDiary to document headaches and migraines for 4 weeks after the screening visit to confirm eligibility criteria and to establish baseline values. Eligible patients were then randomly assigned to receive eptinezumab 100 mg, eptinezumab 300 mg, or placebo in a 1:1:1 ratio. Randomization was stratified by the number of migraine days recorded during the screening period (≤17 vs >17 days) and preventive medication use during the 3 months before screening (use vs no use). The total duration of the study was 32 weeks, with 10 scheduled visits (screening, day 0, and weeks 2, 4, 8, 12, 16, 20, 24, and 32). After the last patient completed the week 12 visit, the analysis of the primary endpoint was performed.

Patients received up to 2 treatments of eptinezumab or placebo (administered IV on day 0 and at week 12). Assignment was concealed. Investigational product (either eptinezumab 100 mg or placebo) was shipped frozen on dry ice as a single-use, preservative-free solution in a tamper-proof 2-mL type I glass vial. Each individual vial was labeled with protocol-specific information to meet International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use and country-specific requirements. On randomization, study sites were provided with enough appropriate investigational product for 1 patient to be dosed, with subsequent treatment packs shipped automatically before the next scheduled treatment visit. Investigational treatments were reconstituted in a total volume of 100 mL 0.9% saline and administered over a period of 30 minutes (plus an additional 15 minutes, per protocol, if required) by the blinded investigator. Patients were monitored for ≥2 hours after treatment. All research participants, clinicians, and research personnel were blinded and remained blinded throughout the duration of the clinical trial.

Outcome measures

Patients completed a daily eDiary from the time of screening through week 24; this included a daily evening report (completed regardless of whether the patient had a headache) and a headache report, which was event based (i.e., per headache). In addition to capturing headache episodes and migraine attacks, the daily eDiary captured acute medication use (including type). A migraine day was defined using information from the eDiary and followed the definition of a CM outlined in ICHD-3. Specifically, a migraine day had to meet 3 criteria: (1): lasted ≥4 hours or lasted 30 minutes to 4 hours and believed by the patient to be a migraine that was relieved by acute medication; (2) had ≥2 of the following: unilateral location, pulsating quality, moderate or severe pain intensity, and aggravation by or causing avoidance of routine physical activity; and (3) had ≥1 of the following: nausea and/or vomiting or photophobia and phonophobia.

Monthly results (e.g., monthly migraine days [MMDs]) were based on the results of the 4-week intervals. If the headache diary was completed for ≥21 days of a 4-week interval, the observed frequency was normalized to 28 days. If the diary was completed for <21 days of a 4-week interval, the results were a weighted function of the observed data for the current interval and the results from the previous interval, with the weight proportional to how many days the diary had been completed.

The primary efficacy endpoint was the change from baseline in MMDs over weeks 1 to 12, assessed with eDiary data. eDiary data also were used for evaluating several of the key secondary efficacy endpoints: ≥75% migraine responder rate over weeks 1 to 4, ≥75% migraine responder rate over weeks 1 to 12, ≥50% migraine responder rate over weeks 1 to 12, percentage of patients with a migraine on the day after dosing, change from baseline in daily migraine prevalence from baseline to week 4, and acute migraine medication use during weeks 1 to 12.

During the scheduled visits, patients completed several patient-reported outcome measures, including the 6-item Headache Impact Test (HIT-6), the 36-item Short-Form Health Survey (version 2.0), the EuroQol 5-Dimensions 5-Levels, and the Patient Global Impression of Change. The Most Bothersome Symptom was evaluated as a tertiary endpoint and is not reported here.

The HIT-6^17,18 version 1.0, an α control endpoint in this study for the 300 mg treatment group, was administered at screening, day 0, and at week 4, 12, 16, 24, and 32 visits to measure the impact of headaches on the ability to function normally in daily life. Patients rated the HIT-6 items using a Likert-type scale anchored by “never” and “always” and including the following response scores: never = 6, rarely = 8, sometimes = 10, very often = 11, and always = 13. Response scores were summed to produce a total score. Scores ≥60 were indicative of severe life impact, 56 to 59 of substantial life impact, 50 to 55 of some life impact, and ≤49 of little to no life impact.

Safety was assessed throughout the study via adverse event (AE) monitoring, clinical laboratory tests, vital sign measurements, physical examinations, 12-lead ECGs, the Columbia-Suicide Severity Rating Scale,¹⁹ and concomitant medication use. AEs of special interest also were monitored and included AEs associated with the Columbia-Suicide Severity Rating Scale score, cardiovascular AEs, hepatic AEs, and those associated with study drug administration, including hypersensitivity and anaphylaxis AEs.

Blood samples for immunogenicity analyses were collected on day 0 and at week 2, 4, 8, 12, 24, and 32, including monitoring development of anti-eptinezumab antibodies and assaying for neutralizing potential.

Statistical methods

A total of 350 patients per group provided at least 90% power for the primary endpoint for each comparison, assuming a treatment effect of ≥1 day and a common SD of ≤4 days. These sample size calculations were performed with PASS 2008²⁰ and were based on a t test that approximated the analysis of covariance (ANCOVA) used for the primary endpoint.²⁰

All patients who received study medication were included in the safety and efficacy populations. For the safety analyses, patient results were summarized within the group representing the treatment they received; if they received 2 different doses, they were summarized in the treatment arm of the highest dose received. For the efficacy population, patients' results were summarized within the treatment group to which they were randomly assigned.

A serial procedure was used to account for multiplicity associated with >1 dose level and for multiple endpoints (figure 1). Statistical testing was conducted to maintain a study-wide 2-sided 5% α level. The procedure first evaluated the 300 mg vs placebo comparison for the primary endpoint. If this was significant, testing continued to the first subset of key secondary endpoints for 300 mg. If a significant difference was detected for all tests within this series, the procedure then moved on to the second subset of key secondary endpoints for 300 mg. The procedure then moved on to the 100 mg group for the primary endpoint and subsequently the key secondary endpoint groups for 100 mg. If these endpoints were significant, the procedure moved to the final key secondary endpoints for 300 mg. Within each endpoint group, the Holm procedure²¹ was used.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1 Decision rule for dose levels (primary and key secondary endpoints)

HIT-6 = Headache Impact Test-6; MMD = monthly migraine day. ^aStatistical significance must have been met to proceed to the next step with each series. ^bTo proceed to the next series, all tests in the previous series must have shown a statistically significant difference from placebo.

The primary efficacy endpoint was the change from baseline in MMDs (weeks 1–12) in an ANCOVA model. Key secondary efficacy endpoints were ≥75% migraine responder rates during weeks 1 to 4 and 1 to 12, ≥50% migraine responder rates during weeks 1 to 12, percentage of patients with a migraine on the day after dosing, reduction in average daily percentage of patients with migraine from baseline to week 4, change in HIT-6 total score from baseline to week 12 (300 mg dose only), and acute migraine medication use from baseline to week 12 (300 mg dose only). A ≥50% or ≥75% migraine responder was defined as a patient who achieved ≥50% reduction or ≥75% reduction in MMDs from baseline, respectively. These reductions were evaluated by comparing the baseline frequency of migraine days to the migraine frequency in 4-week intervals. Results from these 4-week intervals will be combined to produce 12-week responder endpoints. The week 1 to 4, 5 to 8, and 9 to 12 changes from baseline measures were averaged, and the average value was compared to baseline. A percent change from baseline was determined, which was used to determine the responder status.

For the key secondary endpoints (migraine responder rates and percentage of patients with a migraine on the day after dosing), testing was based on Cochran-Mantel-Haenszel/extended Cochran-Mantel-Haenszel tests. The tests were stratified by the randomization stratification factor. For the HIT-6 and Acute Migraine Medication endpoints, an ANCOVA model similar to the one used for the primary endpoint was used for testing. Exploratory and safety endpoints were summarized with descriptive statistics. All analyses were conducted with SAS software (SAS Institute, Inc, Cary, NC) version 9.2 or higher. Appendix 2 available from Dryad (doi.org/10.5061/dryad.ck84q23) provides additional information on the statistical analysis.

Data availability

The data reported are part of an ongoing, global sponsor-led clinical development and registration program. Deidentified participant data are not available for legal and ethical reasons.