A migraine day was defined as when at least 1 of the following situations occurred: - a calendar day (0:00 to 23:59) demonstrating at least 4 consecutive hours of a headache endorsing criteria for migraine with or without aura - a calendar day (0:00 to 23:59) demonstrating at least 4 consecutive hours of a headache endorsing criteria for probable migraine, a migraine subtype where only 1 migraine criterion is missing - a calendar day (0:00 to 23:59) demonstrating a headache of any duration that was treated with migraine-specific medications (triptans and ergot compounds) Monthly averages are derived and normalized to 28 days equivalent by the following formula: (# days of efficacy variable over relevant period / # days with assessments recorded in the e-diary over the relevant period) * 28. The change is calculated as postbaseline value - baseline value.

Full analysis set

An adverse event was defined as any untoward medical occurrence that develops or worsens in severity during the conduct of a clinical study and does not necessarily have a causal relationship to the study drug. Severity was rated by the investigator on a scale of mild, moderate and severe, with severe= an AE which prevents usual activities. Relationship of AE to treatment was determined by the investigator. Serious AEs include death, a life-threatening adverse event, inpatient hospitalization or prolongation of existing hospitalization, persistent or significant disability or incapacity, a congenital anomaly or birth defect, OR an important medical event that jeopardized the patient and required medical intervention to prevent the previously listed serious outcomes.

Safety population

Responder rates were defined as the percentage of total subjects who reached at least a 50% reduction in the monthly average of headache days (as subjectively reported by participants in the study diary) of at least moderate severity relative to the baseline period. For the overall analysis (Month 1-3), patients who discontinued early were considered nonresponders. Monthly averages are derived and normalized to 28 days equivalent by the following formula: (# days of efficacy variable over relevant period / # days with assessments recorded in the e-diary over the relevant period) * 28. The percentage reduction in monthly average is calculated as: ((baseline value - postbaseline value) / baseline value) * 100

Full analysis set

Patients recorded any migraine medications (name of drug, number of tablets/capsules, and the dose in milligrams per tablet/capsule) taken on each day in their electronic headache diary device. Acute migraine-specific medication included triptans or ergots. Monthly averages are derived and normalized to 28 days equivalent by the following formula: (# days of efficacy variable over relevant period / # days with assessments recorded in the e-diary over the relevant period) * 28. The change is calculated as postbaseline value - baseline value.

Full analysis set

A migraine day was defined as when at least 1 of the following situations occurred: - a calendar day (0:00 to 23:59) demonstrating at least 4 consecutive hours of a headache endorsing criteria for migraine with or without aura - a calendar day (0:00 to 23:59) demonstrating at least 4 consecutive hours of a headache endorsing criteria for probable migraine, a migraine subtype where only 1 migraine criterion is missing - a calendar day (0:00 to 23:59) demonstrating a headache of any duration that was treated with migraine-specific medications (triptans and ergot compounds) Monthly averages are derived and normalized to 28 days equivalent by the following formula: (# days of efficacy variable over relevant period / # days with assessments recorded in the e-diary over the relevant period) * 28. The change is calculated as postbaseline value - baseline value.

Full analysis set; includes participants with observations

A subset of patients (specified in the protocol not to exceed 30%) were allowed to use 1 concomitant migraine preventive medication. This outcome only includes those participants who did not take concomitant preventive migraine medication during this study. A migraine day has been previously defined. Monthly averages are derived and normalized to 28 days equivalent by the following formula: (# days of efficacy variable over relevant period / # days with assessments recorded in the e-diary over the relevant period) * 28. The change is calculated as postbaseline value - baseline value.

Full analysis set of participants who did not receive concomitant migraine prevention medication

The MIDAS questionnaire is a 5-item instrument developed to assess headache-related disability based on lost days of activity in 3 domains (work, household work, and nonwork) over the previous 3 months. The total score, ie, the sum of the # lost days answered for the first 5 questions, is used for grading of disability, with scores of 0-5 lost days = grade 1 (little or no disability), 6-10 lost days =grade 2 (mild disability), 11-20 lost days = grade 3 (moderate disability), and ≥21 lost days interpreted as grade 4 (severe disability). Negative change from baseline scores indicate a reduction (improvement) in headache-related disability.

Full analysis set

12-lead ECGs were performed before other assessments (eg, blood draws and administration of questionnaires) and performed in triplicate. The worst post-baseline finding for the patient is summarized. Only patients with both baseline and post-baseline ECGs are included. The ECG was evaluated by the investigator at the time of recording (signed and dated), and the printout was kept in the source documentation file. When potentially clinically significant findings were detected by the investigator, a cardiologist at a central diagnostic center was consulted for a definitive interpretation. Any ECG finding that was judged by the investigator as a potentially clinically significant change (worsening) compared with a baseline value was considered an adverse event. - NCS = abnormal, not clinically significant - CS = abnormal, clinically significant Shift format is: baseline finding / worst post-baseline finding

Safety population of participants with both baseline and post-treatment ECGs

Vital signs were performed before other assessments (eg, blood draws and administration of questionnaires). Vital signs with at least one participant showing potentially clinically significant abnormal findings included: - Pulse Rate Low: <=50 and decrease of >=15 beats per minute - Systolic Blood Pressure Low: <=90 mmHg and decrease of >=20 mmHg - Diastolic Blood Pressure High: >=105 mmHg and increase of >=15 mmHg - Diastolic Blood Pressure Low: <=50 mmHg and decrease of >=15 mmHg - Respiratory Rate Low: <10 breaths / minute

Safety population of participants with both baseline and post-treatment values for each vital sign.

Serum chemistry and hematology laboratory tests with potentially clinically significant abnormal findings included: - Blood Urea Nitrogen (BUN) High: >=10.71 mmol/L - Bilirubin High: >=34.2 umol/L - Alanine Aminotransferase (ALT): >=3*upper limit of normal (ULN) - Aspartate Aminotransferase (AST): >=3*upper limit of normal (ULN) - Gamma Glutamyl Transferase (GGT): >=3*upper limit of normal (ULN) - Hemoglobin: Male: <115 g/L or Female: <=95 g/L - Hematocrit: Male: <0.37 L/L or Female: <0.32 L/L - Leukocytes: >=20*10^9/L or <=3*10^9/L - Eosinophils/Leukocytes: >=10% - Platelets: >=700*10^9/L or <=75*10^9/L

Safety population of participants with at least one postbaseline result for the tests.

Urinalysis with potentially clinically significant abnormal findings included: - Blood: >=2 unit increase from baseline - Urine Glucose (mg/dL): >=2 unit increase from baseline - Ketones (mg/dL): >=2 unit increase from baseline - Urine Protein (mg/dL): >=2 unit increase from baseline

Safety population of participants with at least one postbaseline result for the tests

Shifts in prothrombin time from baseline to endpoint were summarized using patient counts grouped into three categories: - Low (below normal range) - Normal (within the normal range of 9.4 to 12.5 seconds) - High (above normal range) Shift format is: baseline finding / endpoint finding

Safety population of participants with both baseline and posttreatment values

Counts of participants who reported treatment-emergent injection site reactions as AEs are summarized. Preferred terms from MedDRA version 18.1 are offered without a threshold applied.

Safety population

The electronic Columbia-Suicide Severity Rating Scale (eC-SSRS) was used to assess the patient's suicidal ideation (severity and intensity) and behavior (Posner et al 2011). The eC-SSRS Baseline/Screening version was completed by the patient at visit 2, and the eC-SSRS Since Last Visit version was completed by the patient at all other time points. Any positive findings on the eC-SSRS Since Last Visit version required evaluation by a physician or doctoral-level psychologist. Findings after the first dose of study drug using the eC-SSRS Since Last Visit version are summarized.

Safety population