From Swedish Pain Center, Seattle, WA, USA (Dr. Aurora); Sunnybrook Women's College Health Sciences Centre, Toronto, ON, Canada (Dr. Gawel); Nashville Neuroscience Group, PC, Nashville, TN, USA (Dr. Brandes); and Allergan, Inc., Irvine, CA, USA (Drs. Pokta and VanDenburgh).
Botulinum Toxin Type A Prophylactic Treatment of Episodic Migraine: A Randomized, Double-Blind, Placebo-Controlled Exploratory Study
Version of Record online: 29 NOV 2006
Headache: The Journal of Head and Face Pain
Volume 47, Issue 4, pages 486–499, April 2007
How to Cite
Aurora, S. K., Gawel, M., Brandes, J. L., Pokta, S., VanDenburgh, A. M. and for the BOTOX North American Episodic Migraine Study Group (2007), Botulinum Toxin Type A Prophylactic Treatment of Episodic Migraine: A Randomized, Double-Blind, Placebo-Controlled Exploratory Study. Headache: The Journal of Head and Face Pain, 47: 486–499. doi: 10.1111/j.1526-4610.2006.00624.x
- Issue online: 29 NOV 2006
- Version of Record online: 29 NOV 2006
- Accepted for publication August 11, 2006.
- botulinum toxin type A;
- episodic migraine headache;
Objective.—This exploratory trial evaluated the safety and efficacy of multiple treatments of botulinum toxin type A (BoNTA; BOTOX®, Allergan, Inc., Irvine, CA, USA) as prophylactic treatment of episodic migraine headaches.
Design and Methods.—This was an 11-month randomized, double-blind, placebo-controlled, exploratory study. Patients were screened during a 30-day baseline period, and eligible patients with 4 or more migraine episodes and ≤15 headache days entered a single-blind 30-day placebo run-in period. Patients were classified as placebo nonresponders (PNR) if they had at least 4 moderate-to-severe migraine episodes and did not experience at least a 50% decrease from baseline in the frequency of migraine episodes following their placebo treatment. All other subjects were classified as placebo responders (PR). Patients were randomized within each stratum (PNR, PR) to 3 treatments with BoNTA (110 to 260 U of BoNTA per treatment cycle) or placebo at 90-day intervals using a modified follow-the-pain treatment paradigm. The primary efficacy outcome measure was the mean change from baseline in the frequency of migraine episodes for the 30 days prior to day 180 in the PNR group. Secondary efficacy measures included the proportion of patients with a decrease from baseline of 50% or more migraine episodes per 30-day period. Patients were allowed to take concomitant acute and prophylactic headache medications. Adverse events were reported.
Results.—A total of 809 patients were screened and 369 patients (89.2% female; mean age, 45 years; range, 20 to 65 years) entered the placebo run-in period and were subsequently randomized to BoNTA or placebo. The mean total dose of BoNTA was 190.5 units (U) (range, 110 U to 260 U). The predetermined primary efficacy endpoint was not met. Substantial mean improvements of 2.4 and 2.2 fewer migraine episodes per month at day 180 in the PNR stratum treated with BoNTA and placebo, respectively, were observed (P > .999). From day 180 through the end of the study (day 270) at least 50% of all patients in each treatment group had a decrease from baseline of 50% or more migraine episodes per 30-day period. However, in the group of patients with ≥12 headache days at baseline (and ≤15 headache days), BoNTA patients experienced a mean change from baseline of −4.0 headache episodes at day 180 compared with −1.9 headache episodes in the placebo group (P= .048). The majority of treatment-related adverse events were transient and mild to moderate in severity. Only 7 patients (1.9%) discontinued the study due to adverse events (6 BoNTA, 1 placebo).
Conclusion.—There were no statistically significant between-group differences in the mean change from baseline in the frequency of migraine episodes per 30-day period. There were substantial, sustained improvements during the course of the study in all groups. Multiple treatments with BoNTA were shown to be safe and well tolerated over an active treatment period lasting 9 months.
botulinum toxin type A
chronic daily headache
International Headache Society
International Classification of Headache Disorders
Migraine Disability Assessment
analysis of covariance
Migraine headaches affect about 28 million Americans (approximately 18% of women and 7% of men),1,2 cause significant disability,3,4 and require substantial utilization of healthcare resources.1,2,5,6 Migraine headaches have been found to be highly comorbid with depression7 and worsen patients' perceptions of their quality of life.5 Billions of dollars annually in direct and indirect costs are attributed to migraine primarily due to oral pharmacotherapy, physician visits, emergency department visits, and loss of productivity.6,8–10
Migraineurs may be at risk of developing a clinically progressive disorder.11–16 A recent epidemiologic study reported that approximately 14% of episodic migraineurs developed chronic (≥15 days/month) migraine headaches over a 1-year period.16
Prophylactic treatment of migraine may reduce the frequency of migraine episodes and healthcare resource utilization.17,18 However, the approved and/or available treatments for migraine are associated with limited efficacy and frequently intolerable side effects. As a result, gaps exist for current pharmaceutical prophylactic treatment options for migraine.19–22
Botulinum toxin type A (BoNTA; BOTOX®, Allergan, Inc., Irvine, CA, USA) is a purified protein that blocks the release of acetylcholine from presynaptic neurons, resulting in the inhibition of muscle contractions.23,24 Focal administration of BoNTA has been used therapeutically as treatment of disorders characterized by muscle overactivity such as cervical dystonia (CD)25 and spasticity.26,27 BoNTA treatment has resulted in improvement in pain in some patients with CD25 and post-stroke spasticity,26,27 an effect initially attributed to relaxation of excessive muscle activity. However, reports of pain reduction prior to muscular improvement have been noted, suggesting BoNTA has antinociceptive effects.28
Preclinical in vitro and in vivo data have shown that BoNTA also blocks the release of nociceptive mediators such as substance P, glutamate, and calcitonin gene-related peptide, suggesting that BoNTA reduces nociceptive input into the central nervous system from the periphery.28–30 The results of several randomized double-blind, placebo-controlled clinical trials demonstrated that BoNTA significantly reduced the frequency of headache episodes in the prophylactic treatment of migraine and chronic daily headache (CDH).31–33 BoNTA treatment of headache has been demonstrated to be safe with minimal side effects and low discontinuation rates due to adverse events.31–33
The current study was a phase 2 clinical trial that was part of a series of exploratory trials designed to investigate the safety and therapeutic effect of BoNTA as prophylactic treatment for various headache disorders. Specific goals were to identify a responsive patient population, a safe and efficacious dose, and a well-defined treatment regimen.
Objective.— The objective of this study was to evaluate the safety and efficacy of multiple treatments of BoNTA compared with placebo for the prophylactic treatment of migraine headaches in a population of patients suffering from episodic migraine.
Study Design.— This was a randomized, double-blind, placebo-controlled, parallel-group, multicenter, clinical study of repetitive treatments of BoNTA compared with placebo in the management of patients with International Headache Society (IHS)-defined migraine headache, with or without aura. The study was conducted at 20 North American study centers. The overall maximal duration of the study for each patient was 11 months, including a 30-day baseline period, followed by a 30-day placebo run-in period, and then a 9-month active-treatment period encompassing up to 3 treatment cycles (Fig. 1). Prior to study initiation, investigators obtained Institutional Review Board approval. This study was conducted in compliance with the ethical principles that have their origins in the Declaration of Helsinki regarding biomedical research on human subjects and with informed consent regulations.
Inclusion Criteria.— Eligible patients were females and males aged 18 to 65 years who suffered from an average of at least 4 moderate-to-severe migraine episodes but less than or equal to 15 headache days per month (as confirmed by a headache diary used during baseline and defined by IHS 1988 International Classification of Headache Disorders-I).34 Patients' migraine episodes had to have occurred at least 1 year prior to enrollment and first diagnosed before age 50 years. Patients had to have a stable medical condition and be willing and able to give written informed consent. Patients' chronic medication regimens, if any, had to be stable—including migraine prophylactic medications—for at least 3 months immediately prior to the initiation of the baseline period. Patients had to be willing and able to stay on current medications during the course of the study, complete the entire course of the study, and comply with study instructions.
Exclusion Criteria.— Patients were excluded if they had any medical condition or used any agent that may have put them at risk with exposure to this formulation of BoNTA (eg, neuromuscular disorders or agents that might interfere with neuromuscular function) or if they had an infection or skin problem at any of the injection sites or a known allergy or sensitivity to the study medication or its components.
Patients were also excluded if they had a history of “complicated” migraine (eg, migrainous infarction, hemiplegic migraine, ophthalmoplegic migraine, or basilar migraine), or an inadequate response (in the investigator's opinion) to 2 or more prophylactic treatments after an adequate trial. Patients with psychiatric problems that were severe enough to interfere with study implementation were excluded from participation, as were those with previous therapy with botulinum toxin of any serotype, an injection of anesthetics or steroids into the study-targeted muscles during the 30 days immediately prior to initiation of the baseline period, or those patients overusing or abusing symptomatic medication, alcohol, or drugs.
Concurrent chronic use or chronic use in the 3 months prior to the screening period of muscle relaxants (eg, Flexeril®[McNeil Consumer and Specialty Pharmaceuticals, Fort Washington, PA, USA], benzodiazepines) was prohibited. Patients who were concurrently participating in another investigational study or who had participated in such a study in the 30 days immediately prior to baseline or who had a condition that in the investigator's opinion might have put the patient at significant risk, might have confounded the study, or might have interfered with the patient's participation in the study were also excluded. Females who were pregnant, nursing, or planning a pregnancy during the study, or who were unable or unwilling to use a reliable form of contraception during the study were excluded.
Treatment Schedule.— The overall maximal duration of the study for each patient was 11 months including a 30-day baseline period, followed by a 30-day single-blind placebo run-in period, and then a 9-month double-blind period encompassing up to 3 treatment cycles of BoNTA or placebo. During the baseline period and throughout the course of the study, patients recorded specified characteristics of their headache episodes and headache pain medication use on a daily basis using an electronic telephone diary (ClinPhone, Nottingham, UK).
Following the baseline period, patients meeting the inclusion/exclusion criteria were treated (single-blind) with placebo in a minimum of 6 head and neck muscle areas (Table 1) with 23 to 58 injection sites within these areas and assessed after 30 days. Patients were classified as placebo nonresponders (PNR) if they had at least 4 moderate-to-severe migraine episodes and did not experience at least a 50% decrease from baseline in the frequency of migraine episodes following their placebo treatment. All other subjects were classified as placebo responders (PR). The stratification between PNR and PR was included in the study protocol as an attempt to address the well-known typical high placebo response that can be an important confounder in evaluating treatment of painful conditions such as headache. The PNR were considered as the group of primary interest, although all patients were kept in the trial and included in a priori defined analyses.
|Muscle Injected (Allowable Dose Range)||Treatment Cycle 2 Day 0||Treatment Cycle 3 Day 90||Treatment Cycle 4 Day 180|
|Frontal/glabellar (25 to 40 U)||37.5 U (40 U)||37.2 U (40 U)||37.2 U (40 U)|
|Occipitalis (20 U)||19.7 U (20 U)||19.8 U (20 U)||19.8 U (20 U)|
|Temporalis (20 to 50 U)||40.2 U (40 U)||41.0 U (40 U)||41.3 U (40 U)|
|Masseter (optional; 0 to 50 U)||6.0 U (0 U)||7.0 U (0 U)||6.8 U (0 U)|
|Trapezius (20 to 60 U)||48.9 U (60 U)||51.3 U (60 U)||51.8 U (60 U)|
|Semispinalis (10 to 20 U)||17.8 U (20 U)||18.0 U (20 U)||17.9 U (20 U)|
|Splenius capitis (10 to 20 U)||18.2 U (20 U)||17.6 U (20 U)||17.5 U (20 U)|
|Total||188.4 U (200 U)||192.0 U (200 U)||192.3 (200 U)|
Following the placebo run-in period, patients within each stratum (PNR, PR) returned to be randomized to the first of 3 masked BoNTA (110 units [U] to 260 U) or placebo treatments at the initiation of the treatment period (day 0). The effects of treatment were subsequently analyzed separately for each stratum and also for the pooled population. Patients were injected with BoNTA or placebo using a consistent injection protocol (same volumes into the same muscle sites) throughout the study, including the placebo run-in period. Patients received treatments at day 0, 90, and 180 and returned for follow-up visits at 30-day intervals following each treatment through day 270. If a patient exited the study at any visit prior to day 270, all exit procedures and evaluations were to be completed at that visit.
Intervention.— Each vial of BOTOX® contained 100 U of Clostridium botulinum toxin type A, 0.5 mg albumin (human), and 0.9 mg sodium chloride in a sterile, vacuum-dried form without a preservative. Each vial of placebo contained 0.9 mg sodium chloride in a sterile, vacuum-dried form without a preservative. Each identical-looking vial of BOTOX® or placebo was reconstituted with 2.0 mL of unpreserved 0.9% saline for a final concentration of 50 U/mL.
Blinding and Randomization.— The placebo run-in (day −30 to day 0) was a single-blind treatment whereby the investigator, but not the patient, knew that the treatment administered was placebo. Starting at day 0, this was a double-blind study and neither the investigator nor the patient knew which treatment was to be administered.
Patients were randomized within each stratum (PNR, PR) to treatment (BoNTA or placebo) in a ratio of 1:1 in blocks of 4 based on the electronic headache diary data. Neither the investigator nor the patient knew the treatment stratum or random block size. An individual with no other study involvement reconstituted 3 vials of study medication and drew the study drug into the syringes for administration. The syringes were then given to the investigator for injection.
Treatment Protocol.— The injection protocol used was similar to the “follow-the-pain” approach described by Blumenfeld and colleagues35 to allow a more individualized or patient-tailored treatment approach. The number of injection sites and units to be injected into each muscle area was defined by the physician based on the pain distribution pattern and the severity of pain in the particular muscle area as specified in Table 1, except for the occipitalis muscle, where the dosage was fixed. However, for each muscle area, the dosage and number of sites had to be within a predetermined range per muscle, as specified in the study protocol. Patients were to be injected with the same dose and in the same muscle areas and sites for all treatments and, whenever possible, treatments for each patient were to be performed by the same physician throughout the study.
Prior and Concomitant Therapy.— Throughout the study, patients were required to continue stable headache prophylactic treatments. Additionally, patients were allowed to continue using their usual acute headache pain medications as needed. The use of any concurrent medication was recorded along with the reason the medication was taken during the 7 days prior to the baseline and during the study.
Efficacy Measures.— The primary efficacy measure was the mean change from baseline in the frequency of migraine episodes in a 30-day period for the PNR group, determined from data recorded by patients in the telephone diary. The day 180 visit was the primary visit for determining efficacy, with the evaluation reflecting the prior 30-day period. Baseline for the efficacy measures was defined as the frequency of migraine episodes during the first 30 days of the study. A difference of 1.5 headaches between BoNTA and placebo in the mean change from baseline in the frequency of migraine episodes of any severity in a 30-day period at day 180 was considered clinically significant.
Secondary efficacy measures included the proportion of patients with a decrease from baseline of 50% or more migraine episodes in a 30-day period and the proportion of patients with a decrease from baseline of 2 or more migraine episodes in a 30-day period. Other exploratory efficacy measures included the number of days with any headache episode (in a 30-day period); the number of days with migraine episodes (in a 30-day period); the frequency of moderate-to-severe migraine episodes (per 30-day period); the number of headache episodes with migraine features; the number of days that acute migraine headache pain medication was used during the study (in a 30-day period); the Migraine Disability Assessment (MIDAS) scale administered at days −60, 90, 180, and 270; the Headache Pain-Specific Quality of Life Questionnaire administered at days −60, 0, 60, 90, 180, and 270, and the patient's global assessment of response to treatment from baseline.
Safety Measures.— Safety was assessed by reports of adverse events, physical and neurological examinations, and clinical laboratory tests. At each visit following treatment at day 0, adverse events were recorded and documented with information regarding the date of onset, resolution date (if applicable), severity (mild, moderate, or severe), duration, frequency, relationship to study treatment, action taken regarding study treatment, treatment required (if any), and outcome.
Statistical Analyses.— As planned prior to unmasking the treatment code, the “as-treated” population for both the safety and efficacy analyses included all patients treated with the second injection session at day 0 (randomization), regardless of subsequent injections. They were analyzed in the responder/nonresponder stratum indicated by baseline and placebo run−in data, regardless of which stratum they were assigned to during randomization. Multiple comparison adjustments were made by using the O'Brien-Fleming group sequential method to set the significance level at 0.048 for the primary analysis at day 180. All other hypothesis tests used a type I error of α= 0.05 to determine statistical significance, except treatment-by-subgroup interactions, which were examined at the 0.10 level.
For the frequency of migraine episodes at baseline and its change from baseline in the nonresponder stratum, comparisons between treatment groups were performed using the Wilcoxon rank-sum test.36 If there were significant baseline differences among treatments in the primary variable, a baseline covariate was to be included in an analysis of covariance (ANCOVA) of the ranks of the variable. Two secondary efficacy variables were summarized for each 30-day interval: the incidence of patients with a decrease from baseline of 50% or more migraine episodes and the incidence of patients with a decrease from baseline of 2 or more migraine episodes. The secondary efficacy variables using the observed data were analyzed with Pearson's χ2 or Fisher's exact tests. The statistical methods used for the other efficacy variables summarized for each 30-day period were the same as for the primary comparison.
The frequency of migraine episodes in each 30-day period was determined by the data recorded in the telephone electronic diary. A migraine episode was determined by its start and stop times, characteristics, and medications taken. If any 2 consecutive migraine episodes had less than 24 hours between the stop time of the first headache and the start time of the second headache, they were counted as 1 continuous migraine headache. A headache day was defined as a day where any headache episode occurred for any period of time in the 24-hour period from midnight at the start of the day to midnight at the end of the day. A single headache day was counted regardless of the number of episodes during that day. A headache-free day was defined as a day with no headache.
Patient Demographics and Baseline Characteris-tics.— Of the 809 patients screened and assessed over the baseline period, 369 (mean age, 45 years; range, 20 to 65 years; 329/369 [89.2%] female) were considered eligible for participation and were enrolled. A total of 440 (54%) patients were not enrolled for the following reasons: fewer than 4 headaches per month (51%), >15 headache days per month (11%), withdrawal of consent (13%), other (8%), not enough diary calls (6%), lost to follow-up (4%), lack of efficacy (3%), medication/condition unstable (2%), and taking exclusionary medications (2%).
BoNTA-treated patients were significantly older than placebo-treated patients (mean age, 46.0 years vs 43.9 years, respectively, P= .041). There were no other statistically significant differences between treatment groups in their demographic or baseline characteristics (Table 2).
|Baseline Characteristic||BoNTA 105 to 260 U (N = 187)||Placebo (N = 182)||Total (N = 369)||P Value|
|Years since onset||24.1 (13.2)||21.2 (11.7)||22.7 (12.6)||.054*|
|Age at onset (years)||21.3 (11.1)||22.0 (12.0)||21.6 (11.6)||.653*|
|Use of prophylactic treatment, n (%)||80 (42.8%)||61 (33.5%)||141 (38.2%)||.067†|
|Days of acute medication use||7.8 (2.9)||7.9 (2.7)||7.85 (2.8)||.878*|
|Migraine headache episodes per month||6.4 (2.4)||6.6 (2.4)||6.5 (2.4)||.249*|
|Duration of migraine episodes (hours)||25.5 (20.6)||26.8 (20.9)||20.7||.539*|
|Baseline MIDAS score||34.7 (22.8)||35.6 (23.0)||35.1 (22.9)||.742*|
At the end of the placebo run-in period, 203 of 369 (55%) patients were classified as PNR and 166 of 369 (45%) patients as PR (Fig. 1). Subsequently patients were randomized within each group (PNR, PR) to receive either BoNTA or placebo. Within the PNR stratum, 103 patients received BoNTA and 100 patients received placebo. Within the PR stratum, 84 patients received BoNTA and 82 patients received placebo.
Patient Disposition.— A total of 77.2% of patients (285/369) completed the study. The percentage of patients discontinuing the study were similar for both BoNTA and placebo groups. Of the 369 enrolled patients, 84 patients discontinued early: 18 (4.9%) discontinued for lack of efficacy; 10 (2.7%) for personal reasons; 7 (1.9%) for adverse events; 4 (1.1%) for inability to follow study instructions; 4 (1.1%) were lost to follow-up; and 41 (11.1%) for other reasons.
History, Frequency, and Severity of Migraine Episodes During Baseline.— Patients reported a mean of 6.5 (± 2.4) migraine episodes per month for a mean duration of 26.2 (± 20.7) hours (Table 2). In both treatment groups, a history of migraine with or without aura was reported for all patients. The mean time since onset of migraine episodes was 22.7 years, and the mean age at onset was 21.6 years (Table 2). The mean baseline MIDAS score (35.1) indicated a severe disability.
BoNTA Treatment.— The mean (median) total dose of BoNTA for the 3 treatment cycles was 188.3 U (200 U) at day 0, 192.0 U (200 U) at day 90, and 192.3 U (200 U) at day 180 (Table 1).
Medication Use at Baseline.— Prophylactic headache medication use at the time of entry to the study was reported at the baseline visit for 42.8% (80/187) of patients treated with BoNTA and for 33.5% (61/182) of patients treated with placebo (P= .067) (Table 2). The prophylactic headache medications patients were using at baseline were antidepressants (n = 66), anticonvulsants (n = 55), β-blockers (n = 34), and/or calcium channel blockers (n = 25). Patients could use more than one prophylactic treatment.
The acute headache pain medications patients were using upon entry into the study included triptans, opioids, simple analgesics, and antiemetics. The average number of days per month that patients used acute headache pain medications at baseline was 7.8 (± 2.85) days for the BoNTA-treated patients and 7.9 (± 2.69) days for the placebo-treated patients (P= .878) (Table 2). In the PNR stratum, 102 of 103 (99%) BoNTA patients and 99 of 100 (99%) placebo patients used acute headache pain medications (P> .999). In the PR stratum, 83 of 84 (98.8%) BoNTA patients and 81 of 82 (98.8%) placebo patients used acute headache pain medications (P> .999).
Analyses of Efficacy.—
Mean Change from Baseline in the Frequency of Migraine Episodes per 30-Day Period.— At day 180 in the PNR stratum, the mean change from baseline in the frequency of migraine episodes per 30-day period was −2.4 in the BoNTA group compared with −2.2 in the placebo group (P> .999) (Table 3). There were no statistically significant differences between treatment groups in the mean changes from baseline in the frequency of migraine episodes per 30-day period at any time point.
|Placebo Nonresponder (Predetermined Group of Primary Interest)||Placebo Responder|
|Time Period||BoNTA (n = 103)||Placebo (n = 100)||P Value||BoNTA (n = 84)||Placebo (n = 82)||P Value|
|Baseline||5.6 (1.7)||5.6 (1.5)||.907||4.8 (1.6)||4.9 (1.6)||.802|
|Placebo run-in||0.2 (1.66)||0.3 (1.90)||.574||−2.6 (1.8)||−2.5 (1.9)||.936|
|Day 30||−1.1 (2.4)||−1.0 (2.5)||.563||−2.5 (2.3)||−2.1 (2.3)||.262|
|Day 60||−1.4 (2.6)||−1.6 (2.4)||.951||−2.6 (2.1)||−2.5 (2.1)||.802|
|Day 90||−1.3 (2.5)||−1.6 (2.5)||.545||−2.6 (1.7)||−2.5 (2.5)||.702|
|Day 120||−2.1 (2.6)||−2.2 (2.2)||.797||−2.7 (2.0)||−2.5 (2.1)||.743|
|Day 150||−2.0 (2.6)||−2.0 (2.4)||.833||−2.3 (2.0)||−2.6 (2.4)||.366|
|Day 180||−2.4 (2.6)||−2.2 (2.7)||>.999||−2.5 (2.2)||−2.2 (2.2)||.594|
|Day 210||−2.2 (2.9)||−2.8 (2.2)||.235||−3.1 (1.9)||−2.7 (2.2)||.406|
|Day 240||−2.1 (2.8)||−2.8 (2.5)||.101||−2.9 (2.1)||−3.0 (2.1)||.907|
|Day 270||−1.7 (3.2)||−2.6 (2.5)||.082||−3.2 (2.4)||−3.2 (2.0)||.947|
To compare the treatment response for BoNTA and placebo in the entire randomized patient population, the PR and PNR strata were pooled and a preplanned analysis was completed for this and other efficacy measures.
Pooled data of both strata (PNR and PR) showed that both treatment groups had a sustained decrease from baseline over time in the mean frequency of migraine episodes per 30-day period. From day 120 through day 270, the mean decreases from baseline in the frequency of migraine episodes per 30-day period ranged from 2.1 to 2.5 in the BoNTA group and from 2.2 to 2.9 in the placebo group.
Decrease from Baseline of 50% or More Migraine Episodes per 30-Day Period.— There were no statistically significant differences between treatment groups in the proportion of patients with a decrease from baseline of 50% or more migraine episodes per 30-day period. Pooled data of both strata (PNR and PR) showed that from day 180 onward at least 50% of patients in both treatment groups had a decrease from baseline of 50% or more migraine episodes per 30-day period (Fig. 2).
Proportion of Patients with a Decrease from Baseline of 2 or More Migraine Episodes per 30-Day Period.— There were no statistically significant differences between treatment groups in the proportion of patients with a decrease from baseline of 2 or more migraine episodes per 30-day period. Pooled data of both strata (PNR and PR) showed that from day 60 onward at least 60% of patients in all treatment groups had a decrease from baseline of 2 or more migraine episodes per 30-day period (Fig. 3).
Other Efficacy Measures.— In the analyses of other efficacy measures, there were only occasional statistically significant differences between treatment groups on various measures at certain time points favoring either BoNTA (patient's global assessment of response to treatment from baseline at day 30) or placebo (number of days with headache; the average [usual] migraine headache severity; the percentage of patients using acute headache pain medications; and the number of days with any acute headache pain medication use).
Post hoc Analysis of Efficacy by Baseline Head- ache Day Count.— In light of recent trials in migraineurs with CDH showing significant improvement for BoNTA versus placebo in the frequency of headache episodes,32,33,37 a post hoc exploratory analysis of the patients with the highest frequency of headache days was performed. The analysis included 88 (41 BoNTA, 47 placebo) patients with ≥12 baseline headache days (and ≤15 headache days). Occasional statistically significant differences between groups on some efficacy measures were found. For instance, BoNTA patients experienced a mean reduction from baseline of 4.0 headache episodes at day 180 compared with a reduction of 1.9 headache episodes in the placebo group (P= .048) (Fig. 4). A statistically significant difference favoring BoNTA was also observed in the incidence of patients using acute headache pain medications at day 60 (78.9% of BoNTA patients vs 100% of placebo patients; P= .031).
A post hoc analysis of the population not using concurrent headache prophylaxis was also performed (BoNTA 57.2%[107/187]; placebo 66.4%[121/182]) (P= .067). There were no significant differences favoring BoNTA over placebo.
Safety and Tolerability.— The adverse event data for the PNR and PR strata were pooled and comparisons between treatment groups were made. During the study a total of 70.7% (261/369) of patients experienced 1 or more adverse events, regardless of causality. The overall incidence was significantly higher (P< .001) for BoNTA patients (81.3%, 152/187) compared with placebo patients (59.9%, 109/182). Seven patients (1.9%; 6 BoNTA, 1 placebo) discontinued the study due to adverse events. The majority of adverse events was transient and mild to moderate in severity.
Treatment-related adverse events were reported for a significantly higher percentage of BoNTA (60.4%, 113/187) compared with placebo patients (21.4%, 39/182; P< .001) (Table 4). The most frequently reported treatment-related adverse events in the BoNTA group were muscular weakness (26.2%, 49/187), neck pain (17.1%, 32/187), blepharoptosis (15.5%, 29/187), and neck rigidity (10.2%, 19/187). The most frequently reported treatment-related adverse events in the placebo group were headache (4.9%, 9/182) and neck pain (4.4%, 8/182). BoNTA patients had significantly higher incidences of treatment-related neck pain, neck rigidity, arm pain, muscular weakness, hypertonia, skin tightness, blepharoptosis, and eyelid edema (P≤ .009). No treatment-related adverse event had a significantly higher incidence in the placebo group than the BoNTA group. The majority of treatment-related adverse events was transient and mild to moderate in severity.
|Preferred Term||BoNTA 105 to 260 U (N = 187)||Placebo (N = 182)||P Value*|
|Any adverse event||113 (60.4%)||39 (21.4%)||<.001†|
|Muscular weakness‡||49 (26.2%)||2 (1.1%)||<.001†|
|Neck pain||32 (17.1%)||8 (4.4%)||<.001†|
|Blepharoptosis||29 (15.5%)||3 (1.6%)||<.001†|
|Neck rigidity||19 (10.2%)||6 (3.3%)||.009†|
|Arm pain||14 (7.5%)||2 (1.1%)||.003†|
|Skin tightness||14 (7.5%)||1 (0.5%)||<.001†|
|Hypertonia||13 (7.0%)||2 (1.1%)||.004†|
|Eyelid edema||12 (6.4%)||0 (0.0%)||<.001†|
|Headache||11 (5.9%)||9 (4.9%)||.691|
|Hypesthesia||7 (3.7%)||3 (1.6%)||.337§|
|Migraine||6 (3.2%)||1 (0.5%)||.122§|
|Injection site pain||4 (2.1%)||1 (0.5%)||.372§|
|Dizziness||4 (2.1%)||0 (0.0%)||.123§|
|Paresthesia||4 (2.1%)||0 (0.0%)||.123§|
|Back pain||3 (1.6%)||1 (0.5%)||.623§|
|Pain||3 (1.6%)||0 (0.0%)||.248§|
|Facial paralysis||3 (1.6%)||0 (0.0%)||.248§|
|Ecchymosis||2 (1.1%)||3 (1.6%)||.682§|
|Injection site hemorrhage||0 (0.0%)||4 (2.2%)||.058§|
No patients died during the study. Four patients (3 BoNTA, 1 placebo) experienced 4 serious adverse events, of which none was reported by the investigator to be related to study medication (BoNTA). There were no clinically significant differences between the BoNTA and placebo groups in the changes from baseline to follow-up for any laboratory parameter or vital sign.
In this study, both BoNTA and placebo treatment considerably reduced the frequency of migraine episodes. However, BoNTA treatment did not demonstrate superiority over placebo. Several factors may have contributed to the outcome of this trial, including a strong placebo response and the allowance of concomitant acute and prophylactic medications.
In this trial, the magnitude of response (mean reduction from baseline in the frequency of migraine episodes per month) to BoNTA was consistent with, or larger than, other trials evaluating migraine prophylactic treatments (BoNTA: mean, 2.4; topiramate: mean, 2.3, 2.1; extended-release divalproex sodium: mean, 1.2).21,22,38 However, the mean reduction from baseline in this trial was similar for both treatment groups (BoNTA: 2.4; placebo: 2.2), resulting in a mean between-group difference of only 0.2. The placebo response profile in the current study was notably more robust than that observed in these other trials of migraine prophylaxis treatments. Thus, even with similar or greater mean reductions from baseline of migraine episodes per month with BoNTA treatment, other trials have reported higher mean between-group differences of 0.6 to 1.3 migraine episodes, potentially as the result of less placebo response.21,22,38
In 2 recent trials,21,22 a mean reduction from baseline of migraine episodes per month of 2.3 and 2.1, respectively, was observed after treatment with topiramate at the recommended dosage, compared with mean reductions from baseline of 1.1 and 1.0 after placebo treatment. This resulted in mean between-group differences of 1.2 and 1.1 monthly migraine episodes. In a trial of extended-release divalproex sodium treatment of episodic migraine,38 a mean reduction from baseline of 1.2 migraine episodes per month compared with 0.6 was reported for divalproex sodium and placebo, respectively, resulting in a significant mean between-group difference of 0.6 migraine episodes (P= .006). Similarly, in a trial of gabapentin treatment of episodic migraine,39 gabapentin reduced the median migraine headache rate per 4 weeks by a value of 2.0 compared to 0.8 for placebo; a significant median between-group difference of 1.2 (P= .013).
It should be noted, however, that interpretations of results from different trials are limited due to variations in study design including patient selection, choice of efficacy outcome measure, the size of the patient population, the duration of the trial, and potentially confounding factors such as the use of concomitant headache prophylaxis and acute headache treatments.
It is evident from the results of this trial that there was a robust response to both BoNTA and placebo treatments. A high placebo response is a known phenomenon of clinical trials examining painful conditions, including headache,40 for reasons that are not entirely understood. Zubieta et al,41 using molecular imaging techniques, observed significant placebo-induced activation of μ-opioid receptor-mediated neurotransmission in both higher-order and subcortical brain regions with regional activations corresponding to lower ratings of pain intensity. In a subsequent study using saline injections as a placebo, Zubieta et al found a strong positive correlation between the expectations of analgesia—including the patient's estimate of a treatment's analgesic efficacy—with activation of the opioid system.42 Expectation of analgesia in addition to the perceived analgesic efficacy of a particular treatment may therefore be important in eliciting a neurochemical antinociceptive response associated with an injected placebo.41–43
An analysis of 8 placebo-controlled trials in migraine found that approximately 28% of placebo patients (78/282) improved by at least 50%.40 An analysis of 11 placebo-controlled trials of analgesic efficacy in migraine patients found an average response rate of 30% in patients given placebo,44 and a recent analysis of 31 placebo-controlled clinical trials evaluating oral triptans as treatment of acute migraine in adults published from January 1991 to April 2002 found that 28% of placebo patients given placebo experienced a response to treatment.45 In 2 randomized, double-blind, placebo-controlled trials of topiramate prophylactic treatment of episodic migraine, the responder rates for the placebo groups were 23% and 22.6%, respectively.21,22 In comparison, in a recently published randomized, double-blind, placebo-controlled trial of BoNTA treatment of CDH, 38% of placebo patients experienced at least a 50% reduction in the frequency of headache episodes at the primary time point (day 180).32 In the current trial more than 50% of placebo patients achieved at least a 50% reduction in migraine episodes from day 180 onward. The higher placebo response in BoNTA clinical trials may be related to the higher expectation of analgesic efficacy related to an injectable therapeutic.
Experts suggest that concomitant prophylaxis may confound the outcome of headache trials and recommend that patients enrolled into clinical trials evaluating headache prophylaxis treatments should not be taking other migraine prophylactic medications.46,47 Furthermore, trials with other prophylactic agents have not only forbidden the use of concomitant prophylaxis but have also excluded patients who have failed several prophylactic treatments.21,22,38,39 In the current trial, 38% of patients were taking a headache prophylaxis treatment at baseline and throughout the study. Furthermore, patients were frequently using acute headache treatments (mean ∼8 days of use per month). For patients enrolled into this study, it is unknown how many different types of headache prophylaxis treatments patients had previously tried and failed. In a recent subanalysis of the Mathew et al CDH trial, Dodick et al evaluated the efficacy of BoNTA treatment in a subset of patients not using concomitant prophylactic headache medications.37 In this subanalysis, a statistically significant difference favoring BoNTA over placebo was observed in the mean change from baseline in the number of headache-free days at day 180 (BoNTA, 10.0 headache-free days; placebo, 6.7 headache-free days; P= .038) for this particular patient subset. In addition, the mean change from baseline in the frequency of headache episodes per 30-day period was greater in the BoNTA group at all time points, with statistically significant between-group differences observed starting at the first month (P= .004) and at most of the time points thereafter (range, 2.2 to 4.2 headache episodes). The mean reduction in the frequency of headache episodes was of greater magnitude in patients not taking concomitant headache prophylaxis than in the intent-to-treat population.37
Pooled data from both the PNR and PR strata in this trial showed that from day 120 through day 270 (the end of the trial), BoNTA and placebo treatment resulted in a mean decrease from baseline of 2 or more migraine episodes per 30-day period. The percentage of patients achieving a mean reduction of 2 or more migraine episodes per month was at least 60% of patients in all treatment groups from day 60 onward. Interestingly, in a post hoc analysis in patients experiencing a higher baseline frequency of migraine headache days (≥12 headache days), BoNTA treatment significantly reduced the frequency of migraine episodes per 30 days from baseline by a mean 4 migraine episodes at day 180 (the predetermined efficacy time point) compared with a reduction of 1.9 migraine episodes for the placebo patients (P < .048). Although the day 180 time point was the only time point at which the between-group difference reached statistical significance, a consistent trend was observed within this subgroup of patients favoring BoNTA at all time points throughout the study.
BoNTA prophylactic treatment of migraine was safe and well tolerated in this trial. The discontinuation rate due to adverse events in this 9-month study was 1.9%. This compares favorably to the discontinuation rates due to adverse events observed in other migraine prophylaxis trials.21,22 For instance, at least 19% of patients given 100 and 200 mg per day in both topiramate trials discontinued due to adverse events within the first 6 months of treatment.21,22
This trial was part of an ongoing series of exploratory trials designed to improve understanding of the potential role of BoNTA as prophylactic treatment for headache conditions including episodic migraine and CDH. There is currently no universal consensus regarding methodology for conducting clinical headache trials and a lack of agreement on which efficacy endpoint is sensitive enough to detect a treatment effect. The IHS guidelines for conducting migraine trials recommend using the frequency of headache episodes, which may be the most sensitive for detecting a clinically relevant treatment effect,46 but the definition of what constitutes a headache episode needs to be clarified in terms of its clinical characteristics, duration, severity, and whether medications are needed.
There is emerging evidence that BoNTA may be effective as prophylactic headache treatment. BoNTA has been shown to block the transmission of nociceptive mediators from nociceptive neurons in preclinical in vitro and in vivo trials.28–30 To date, 3 large, well-designed, randomized, double-blind, placebo-controlled exploratory trials evaluating the safety and efficacy of BoNTA as prophylactic treatment of episodic migraine31 and CDH32,33 have shown that BoNTA significantly reduced the mean frequency of headache episodes compared with placebo. In a 3-month study conducted by Silberstein and colleagues in episodic migraine patients (n = 123), BoNTA 25 U (but not 75 U) was significantly effective compared with placebo in the mean change from baseline in frequency of moderate-to-severe migraine headache episodes at months 2 (−1.57 vs −0.37; P= .008) and 3 (−1.88 vs −0.98; P= .042) and the mean change from baseline in the frequency of migraine headache episodes of any severity at month 3 (−2.12 vs −0.90; P= .01).31 The lack of effect in the 75-U group may have been the result of the study design, wherein the randomization resulted in a significantly lower baseline frequency of migraines in these patients (4.4) compared with the BoNTA 25-U group (5.5) and the placebo group (5.2) (P≤ .046).
Recently two 11-month clinical trials were conducted that evaluated the safety and efficacy of BoNTA as prophylactic headache treatment in CDH patients who reported predominantly migraine/probable migraine headaches.32,33 Although the primary endpoint (the mean change from baseline in headache-free days) was not met in either trial, statistically significant differences in the mean change from baseline in the frequency of headache episodes per 30 days favoring BoNTA were observed. In the trial conducted by Mathew et al (N = 355),32 BoNTA treatment significantly reduced the mean frequency of headache episodes compared with placebo at day 180 (the primary time point) by 6.1 episodes compared with 3.1 episodes (P= .013) and at most time points in the PNR group (the group of primary interest). In a similarly designed trial, Silberstein et al evaluated BoNTA 75 U, 150 U, and 225 U compared with placebo in the treatment of CDH (N = 702).33 A pooled analysis of PR and PNR revealed significant between-group differences at day 240 (P= .03) favoring the BoNTA 225 U (−8.4; P= .04, ANCOVA) and 150 U (−8.6; P= .02, ANCOVA) groups over placebo (−6.4) for mean changes from baseline in the frequency of headache episodes per 30-day period.33
In the current trial, post hoc analyses suggested that patients with a greater mean frequency of headache days at baseline might have a more robust response to BoNTA treatment. Taken together, these results suggest that BoNTA may be more effective in patients with a higher mean baseline frequency of headache episodes or headache days. The results also add to existing evidence suggesting that higher doses are more effective. However, further study is needed to improve the understanding of the mechanism of the interaction of BoNTA with the nociceptive pathway, the most effective doses of BoNTA, and the types of headache patients that may optimally respond to BoNTA treatment.
Dosing and results reported in this study are specific to the formulation of BoNTA manufactured by Allergan, Inc. The Allergan, Inc. formulation is not interchangeable with other botulinum toxin products and cannot be converted by using a dose ratio.23
BOTOX 037 Study Group: Richard Kaplan, MD (California); Jan Lewis Brandes, MD (Tennessee); Keith R. Edwards, MD (Vermont); Arthur Elkind, MD (New York); Brian Freund, MD, DDS (Ontario); Marvin Schwartz, DDS (Ontario); Benjamin Frishberg, MD (California); Marek Gawel, MD (Ontario); Jack Klapper, MD (Colorado); Alexander Mauskop, MD (New York); Peter McAllister, MD (Connecticut); B. Todd Troost, MD (North Carolina); Bradley Wrubel, MD (California); Sheena Aurora, MD (Washington); Mahan Chehrenama, DO (Virginia); Suzanne Christie, MD (Ontario); Joseph Kandel, MD (Florida); V. Daniel Kassicieh, DO (Florida); Joel Saper, MD (Michigan); Fred Sheftell, MD (Connecticut); Suriana Pokta, PhD (California); Amanda M. VanDenburgh, PhD (California).
Acknowledgment: This study was sponsored by Allergan, Inc., Irvine, CA, USA.
Conflict of Interest: Suriani Pokta, PhD, and Amanda M. VanDenburgh, PhD, are employed by Allergan, Inc., and own stock in the company.
- 4World Health Organization. Headache disorders and public health: Education and management implications. Available at http://www.migraines.org/new/pdfs/who.pdf. Accessed April 5, 2006.
- 23BOTOX®[package insert]. Irvine , CA : Allergan, Inc.; 2004.
- 25for the BOTOX® Cervical Dystonia Prospective Study Group. A randomized, double-masked, crossover comparison of the efficacy and safety of botulinum toxin type A produced from the original bulk toxin source and current bulk toxin source for the treatment of cervical dystonia. J Neurol. 2002;249:57-63., , ,
- 32and the BOTOX CDH Study Group. Botulinum toxin type A (BOTOX®) for the prophylactic treatment of chronic daily headache: A randomized, double-blind, placebo-controlled trial. Headache. 2005;45:293-307., , , , , ,
- 34Headache Classification Subcommittee of the International Headache Society. The International Classification of Headache Disorders: 2nd edition. Cephalalgia. 2004;24(suppl 1):9-160.
- 36Nonparametric Statistics for the Behavioral Sciences. New York : McGraw-Hill; 1956:116-127..
- 37and the BOTOX CDH Study Group. Botulinum toxin type A for the prophylaxis of chronic daily headache: Subgroup analysis of patients not receiving other prophylactic medications: A randomized double-blind, placebo-controlled study. Headache. 2005;45:315-324., , , , , ,
- 47European Agency for the Evaluation of Medicinal Products (EMEA). Note for Guidance on Clinical Investigation of Medicinal Products for the Treatment of Migraine. London : European Agency for the Evaluation of Medicinal Products (EMEA); December 17, 2003.