Randomized Trial of Sumatriptan and Naproxen Sodium Combination in Adolescent Migraine
BACKGROUND: Treatment of adolescent migraine remains a significant unmet medical need. We compared the efficacy and safety of 3 doses of sumatriptan and naproxen sodium (suma/nap) combination tablets to placebo in the acute treatment of adolescent migraine.
METHODS: This randomized, parallel group study in 12 to 17 year olds required 2 to 8 migraines per month (typically lasting >3 hours untreated) for ≥6 months. Subjects entered a 12-week run-in phase, treating 1 moderate-to-severe migraine (attack 1) with single-blind placebo. Subjects reporting headache pain 2 hours after dosing were randomly assigned into a 12-week double-blind phase, treating 1 moderate-to-severe migraine (attack 2) with placebo (n = 145), suma/nap 10/60 mg (n = 96), 30/180 mg (n = 97), or 85/500 mg (n = 152). The primary end point was the percentage of subjects pain-free at 2 hours.
RESULTS: The attack 2 adjusted (age; baseline pain severity) 2-hour pain-free rates were higher with suma/nap 10/60 mg (29%; adjusted P = .003), 30/180 mg (27%; adjusted P = .003), and 85/500 mg (24%; adjusted P = .003) versus placebo (10%). Posthoc primary end-point analyses did not demonstrate differences among the 3 doses or an age-by-treatment interaction. Statistically significant differences were found for 85/500 mg versus placebo for sustained pain-free 2 to 24 hours (23% vs 9%; adjusted P = .008), 2-hour photophobia-free (59% vs 41%; adjusted P = .008), and 2-hour phonophobia-free (60% vs 42%; adjusted P = .008). Analyses of other pain, associated symptoms, rescue medication use, and health outcome end points supported higher efficacy for active doses versus placebo. All active doses were well tolerated.
CONCLUSIONS: All doses of suma/nap were well tolerated, providing similarly effective acute treatment of adolescent migraine pain and associated symptoms, as compared with placebo.
- AE —
- adverse event
- CMH —
- ITT —
- NS —
- nasal spray
- suma/nap —
- sumatriptan and naproxen sodium
What’s Known on This Subject:
Treatment of adolescent migraine remains a significant unmet medical need. In adults, the combination of sumatriptan and naproxen sodium has demonstrated superior efficacy, with similar tolerability, to its components in the acute treatment of migraine.
What This Study Adds:
This study constitutes the first large-scale, placebo-controlled evidence for the acute relief of adolescent migraine pain and associated symptoms with an oral medication.
Migraine has an estimated prevalence of 8% to 23% in children ≥11 years of age.1 Treatments, both acute and prophylactic, are similar to those used in adults.2,3 Only a few well-controlled pediatric trials have been conducted,4,5 the most rigorously studied agents being ibuprofen, acetaminophen, and 6 of the 7 available triptans.4,6
The authors of various trials have examined triptan safety and efficacy in adolescents.7–12 Ibuprofen and acetaminophen have revealed efficacy and safety in children <12 years of age.13,14 There are no controlled trials of naproxen for acute treatment of adolescent migraine. In Europe, only nasal sumatriptan is approved for acute adolescent migraine treatment.4 In the United States, almotriptan is indicated for acute treatment of adolescent migraine pain lasting >4 hours, but in the pivotal trial, it failed to achieve the primary end point of “migraine freedom” because of lack of efficacy for migraine-associated symptoms (nausea/photophobia/phonophobia).12,15 Thus, most triptans remain an “off-label” treatment option.
Sumatriptan is the most widely studied triptan in adolescents, being examined in various doses in oral16,17 and nasal spray formulations.7,18 The authors of these studies treated moderate-to-severe migraine, evaluated pain-relief, and noted relatively high placebo response rates. All failed to demonstrate efficacy versus placebo for the primary end points, chiefly due to high placebo response, a primary rationale for the failure of triptans to garner a Food and Drug Administration adolescent migraine indication.7,10,18–23
In adults, the combination of sumatriptan and naproxen sodium (suma/nap) has demonstrated superior efficacy, similar tolerability, and improved quality of life and medication satisfaction to its components.24–27 As suma/nap has similarly demonstrated benefits in the early intervention paradigm28–30 and in migraine subpopulations,31–33 it was thought that suma/nap might be efficacious in adolescents. This study, along with pharmacokinetics and long-term safety studies, evaluates suma/nap for the acute treatment of adolescent migraine.
This adolescent study was conducted at 77 US sites (primary care, specialists, and research centers; and emphasizing pediatric experience). Enrollment occurred between December 1, 2008, and January 13, 2010, and subject activities ended June 10, 2010. The design was an outpatient, double-blind, randomized, placebo-controlled, parallel group. The study (protocol TXA107979; www.clinicaltrials.gov identifier NCT00843024) included up to 3 visits over ∼25 weeks, consisting of a 12-week single-blind run-in phase, a 12-week double-blind treatment phase, and a final/early withdrawal visit.
The study was approved by sites’ institutional review boards. Written informed consent from a subject’s parent/guardian, and the subject’s assent, was obtained before any study procedures. Subjects treating 1 moderate-to-severe migraine with single-blind placebo during the run-in phase and reporting pain 2 hours postdose (placebo nonresponders) were randomly assigned into the double-blind phase. In the double-blind phase, subjects treated 1 moderate-to-severe migraine with either matching placebo or suma/nap 10/60 mg, 30/180 mg, or 85/500 mg. Placebo and suma/nap tablets were identical in appearance (size/markings/color/weight).
Subjects were administered study medication at the onset of moderate-to-severe migraine headache pain and were allowed to rescue >2 hours postdose with a single oral dose of naproxen sodium (not >15 mg/kg in 24 hours), over-the-counter pain reliever (not >daily recommended dose), or an antiemetic. Subjects completed paper headache/medication/health outcome diaries.
Sample size estimates were based on previous data from the following: adolescent sumatriptan trials, 18 to 21 year olds in suma/nap adult trials, and a published zolmitriptan adolescent trial.10 To detect a 14% difference between suma/nap and placebo at α = .05 and power = 80% given the 3:2:2:3 randomization ratio, 510 subjects were needed. An anticipated 15% double-blind phase nontreatment rate required randomization of 600 subjects.
Key Required Inclusion and Exclusion Criteria
Eligible participants were 12 to 17 years old at screening and had ≥6 months history of 2 to 8 migraines per month (with or without aura: International Classification of Headache Disorders-II 1.2.1 or 1.1),34 typically lasting ≥3 hours and associated with moderate-to-severe headache pain. Triptan-naïve subjects were eligible.
Exclusionary criteria included the following: ≥15 headache days per month; uncontrolled hypertension; ≥3 cardiovascular or any cerebrovascular risk factors; weight <74 lb (33.3 kg); epilepsy or structural brain lesions history; methysergide or dihydroergotamine use in the past 3 months; daily medications with dose changes in the past 2 months; positive pregnancy test; and positive toxicology screen not attributed to treatment of a medical condition.
Blinding and Randomization
Parent/guardian and subject were informed of the 2 blinded study phases and that the subject might receive placebo or 1 of 3 doses of active drug. Parent/guardian and subject were informed that they would not know which study drug was provided for the first or second migraine treated, and that the investigator would not know which study drug was provided for the second migraine treated. To maintain blinding, sites received education on describing the single-blind phase.
Approximately equal numbers were randomized in each age group (12–14 or 15–17 years), so neither group was >60% nor <40% of the total. Upon determination of subject eligibility and completion of the randomization visit, site staff telephoned an interactive voice response system and dispensed randomly assigned, age-group stratified treatment.
Paper diaries captured headache features and concomitant medications before/upon/after study medication. Migraine attacks 1 and 2 diary assessments included pain intensity, associated symptoms (nausea/vomiting/photophobia/phonophobia) at 0, 1, 2, 4, and 4 to 24 hours, and a single functioning question at 0, 1, 2, and 4 hours. Only attack 2 responses were assessed. Medication satisfaction was recorded at the screen visit and at 2 and 24 hours after treatment of attack 2. Medication satisfaction was evaluated by using the 3 global questions from the Revised Patient Perception of Migraine Questionnaire35 and was recorded on a 5-point scale from “very satisfied” to “very dissatisfied.”
Primary Efficacy End Point
The primary efficacy end point was the percentage of subjects pain-free at 2 hours posttreatment attack 2. Pain-free was defined as the absence of headache pain posttreatment from moderate or severe at baseline, without previous use of rescue medication. Diary entries provided the basis for all pain scores.
Secondary Efficacy End Points
Ten secondary efficacy end points were evaluated for attack 2. Sustained pain-free (or sustained symptom-free) response was defined as pain-freedom (or symptom-freedom) maintained from 2 to 24 hours posttreatment without use of rescue medication.
Other Efficacy End Points
Other efficacy end points were evaluated for attack 2 (only results for 4–24 hours freedom from pain, photophobia, and phonophobia are presented).
Health Outcome End Points
Attack 2 health outcome end points evaluated were as follows:
Percentage of subjects satisfied with study medication at 2 and 24 hours posttreatment; and
Percentage of subjects who returned to normal functioning at 1, 2, and 4 hours posttreatment.
Adverse events (AEs) were assessed.
Efficacy and health outcome analyses used the intent-to-treat (ITT) population, defined as subjects who took ≥1 dose of double-blind, randomized treatment and provided any assessment of their attack 2 migraine pain or associated symptoms. A fixed testing methodology was used to control the overall type 1 error rate at 0.05, while comparing the 3 active dose groups versus placebo regarding the primary efficacy end point, and comparing 85/500 mg versus placebo for 10 secondary efficacy end points. First, for the primary end point, 85/500 mg dose group was compared with the placebo group. If this comparison was statistically significant at α = .05, then the primary end point was tested in a step-down fashion for 30/180 mg and 10/60 mg versus placebo at α = .0375, and 10 secondary end points were tested in a step-down fashion for 85/500 mg versus placebo groups at α = .0125.
All binary end points were analyzed by using the Cochran-Mantel-Haenszel (CMH) test. For pain-related end points, CMH tests controlled for age and baseline pain severity. For migraine-associated symptoms, CMH tests adjusted for age only. The time-to-first-use of rescue medication was analyzed by using the log-rank test. Safety data were summarized by using descriptive statistics. AEs were either single-blind placebo-emergent (safety population, defined as subjects who took single-blind placebo) or double-blind treatment-emergent (modified-safety population, defined as subjects who took double-blind treatment). All efficacy and safety data were assessed across all subjects and examined by age group (12–14 and 15–17 years); however, this study was not designed to detect statistical differences between age groups. All statistical analyses were conducted by using SAS software, version 9.1.3 (SAS Institute, Inc, Cary, NC).
Figure 1 summarizes subject disposition. Fewer subjects were enrolled (408:457) and randomized (268:321) in the 12 to 14 years age group compared with the 15 to 17 years age group; this trend continued for subjects taking single-blind or double-blind study medication and completing the study. Within each age group, similar proportions were enrolled, provided data, and completed the study. Demographic and baseline migraine characteristics were generally similar across age and treatment groups (Table 1).
Migraine treatments at entry (ITT) were as follows: nonsteroidal antiinflammatory drugs (59%), over-the-counter analgesics (46%), or triptans (24%). The most commonly used medications during the study (modified-safety) were ibuprofen (40%) and acetaminophen (32%).
Migraine Characteristics and Treatment Details for Treated Attacks (Attack 2)
Migraine pain, associated symptoms, and attack characteristics were generally similar across age and treatment groups. Pain severity at treatment was severe (48%), moderate (52%), and mild (<1%). The mean time between migraine onset and study medication use was 2.25 hours.
Efficacy Measures (Attack 2)
Primary Efficacy Measure: 2-Hour Pain-Free (Each Active Dose Versus Placebo)
The suma/nap active doses were each significantly more effective than placebo 2 hours postdosing for incidences of pain-freedom (adjusted P = .003 for each active dose versus placebo; Fig 2; Table 2). The percentage of subjects pain-free at 2 hours was 29%, 27%, 24%, and 10%, for 10/60 mg, 30/180 mg, 85/500 mg, and placebo, respectively. There was no evidence that time from onset to treatment (≤30 minutes, >30 minutes) was a predictor for 2-hour pain-free response. Posthoc analyses of dose response and of triptan naïve status did not suggest any differences between active doses for the primary end point. The ITT population comprised 490 subjects compared with the planned 510. This could have reduced the prestudy power; however, because the treatment effect was statistically significant, the smaller sample size is not of concern and the study conclusions are valid.
Secondary Efficacy Measures (85/500 mg Versus Placebo)
Fixed sequence testing methodology compared 85/500 mg versus placebo for 10 secondary end points in the order listed in Table 2. Statistically significant postdose differences for 85/500 mg versus placebo occurred for sustained pain-free 2 to 24 hours and 2-hour photophobia-free and phonophobia-free (adjusted P = .008 for each comparison). Other than 1-hour pain-free and 2-hour nausea-free, all other secondary end points trended numerically better for 85/500 mg versus placebo (unadjusted P < .01).
Secondary Efficacy Measures (Each Active Dose Versus Placebo)
Generally for each secondary end point, analyses of each of the active doses versus placebo had similar results (Table 2). An exception was 2-hour nausea-free, for which the response varied by dose and appeared better than placebo at the 10/60 mg dose only.
Efficacy Results by Age Group (Each Active Dose Versus Placebo)
Nausea-free at 2 hours in 12 to 14 year olds was 84% with 10/60 mg and 83% with 30/180 mg, and greater than placebo (77%); the 85/500 mg rate was 72%. A greater percentage of 15 to 17 year olds were nausea-free at 2 hours for all active doses versus placebo but exhibited a trend for more nausea-freedom with decreasing dose (81%, 72%, 69%, and 63%, for 10/60 mg, 30/180 mg, 85/500 mg, and placebo, respectively).
Trends favored 10/60 mg over 85/500 mg in 12 to 14 year olds regarding 1-hour pain-free (16% vs 6%), 2-hour pain-free (46% vs 27%), 2-hour photophobia-free (67% vs 63%), 2-hour phonophobia-free (74% vs 63%), and 2-hour nausea-free (84% vs 72%). For 15 to 17 year olds, 85/500 mg provided comparable or numerically higher response versus 10/60 mg.
Aside from these trends, the primary and secondary efficacy results were generally similar across both age groups.
Other Efficacy Measures (Each Active Dose Versus Placebo)
At 24 hours postdose, subjects reported the following dose responses for the 4- to 24-hour period: pain-free 54%, 64%, 68%, and 41%; photophobia-free 68%, 72%, 78%, and 55%; and phonophobia-free 70%, 76%, 80%, and 54% for 10/60 mg, 30/180 mg, 85/500 mg, and placebo, respectively.
Health Outcomes (Attack 2)
Satisfaction With Medication Efficacy, Side Effects, and Overall Satisfaction
A higher percentage of subjects treated with any suma/nap dose versus placebo reported being satisfied/very satisfied for “how effective the medication is overall” and “overall satisfaction with medication” at 2 and 24 hours postdose (unadjusted P ≤ .014). More subjects treated with 10/60 mg or 30/180 mg, versus placebo, reported being satisfied/very satisfied for “side effects with medication” at 2 and 24 hours postdose (unadjusted P ≤ .005). However, satisfaction with side effects was comparable between 85/500 mg and placebo (unadjusted P ≥ .412).
Ability to Function
Numerically greater percentages of subjects reported normal functioning at 1, 2, and 4 hours postdose with any active dose versus placebo. Across all active doses, a higher percentage of subjects reported normal functioning at 2 hours postdose, compared with placebo (P ≤ .038). This trend continued at 4 hours postdose (P ≤ .010).
In the modified-safety population, 11% of subjects reported a double-blind treatment emergent adverse event (AE) (Table 3). The incidence was similar across treatment groups (13%, 9%, 13%, and 8% for 10/60 mg, 30/180 mg, 85/500 mg, and placebo, respectively). However, double-blind treatment-emergent AEs within 72 hours after dosing and drug-related double-blind treatment-emergent AEs revealed a dose response.
The most common double-blind treatment-emergent AEs (incidence ≥2% in any treatment group) occurred in girls and were nasopharyngitis, hot flush, and muscle tightness (Table 3); however, each of these events occurred in ≤3 subjects. AEs occurred more frequently with active treatment in the musculoskeletal (1%, 0%, 5%, 1%) and nervous system (4%, 3%, 4%, 0%) organ classes (10/60 mg, 30/180 mg, 85/500 mg, and placebo, respectively). No double-blind treatment-emergent serious AEs or AEs leading to withdrawal occurred.
This is the first placebo-controlled efficacy assessment of a fixed-dose triptan/nonsteroidal antiinflammatory drug combination for the acute treatment of adolescent migraine. All active doses were statistically significant versus placebo for the primary end point and did not differ statistically between doses. The active dose response rates of 24% to 29% for 2-hour pain-free is within the rates of 19% to 44% previously reported in other migraine trials.21
In this study, at 2 hours postdose, tolerability and efficacy trended lower for 85/500 mg compared with 10/60 mg, aligning with the lower satisfaction with side effects reported for 85/500 mg. Nausea may be a factor: 85/500 mg nausea-free rates at 2 hours were no better than placebo. In contrast, a greater percentage of subjects were nausea-free at 2 hours in the 10/60 and 30/180 mg dose groups, as compared with placebo. At later time points, better efficacy was observed for 85/500 mg (ie, the dose response for 4- to 24-hour pain-free, photophobia-free, and phonophobia-free) along with good tolerability.
Individually, sumatriptan 10 mg and naproxen 60 mg are considered pharmacologically ineffective36–38; combined, the 2 were also deemed unlikely to be effective and considered a “no effect” dose.37,38 Retrospectively, other programs provide a rationale for re-thinking this perspective. Adolescent sumatriptan exposures in pharmacokinetics study TXA10850439 for 10/60 mg, 30/180 mg, and 85/500 mg were higher than those of adults, particularly with 10/60 mg (50%–60% higher area under the curve and Cmax), suggesting that 10/60 mg might demonstrate efficacy. Much of an NS dose is swallowed, in effect becoming oral dosing: sumatriptan 10 mg failed to achieve statistical significance on primary end points in adolescent NS studies but demonstrated efficacy for selected secondary end points.7,18 Lastly, all active doses in the current study may have demonstrated efficacy due to unique properties of the combination, perhaps more effectively addressing pathophysiologic mechanisms than either alone.24
It is interesting to note that the primary, and some secondary, efficacy end points are numerically better for 10/60 mg versus 85/500 mg, and that efficacy trend differences by age group are suggested; speculation around these trends, however, must be done cautiously in the absence of replicate data.
One possible explanation is that adolescents in this study may not be distinctly separating headache from nausea when determining efficacy: the percentage of subjects nausea-free at 2 hours postdose trends similarly (10/60 mg > 30/180 mg > 85/500 mg) to pain-free rates. Nausea in migraine has been thought to reflect gastric stasis. Sumatriptan has been shown to reduce gastric motility, so reduction in gastric stasis is a likely consequence of reduced sumatriptan dose. Further, increasing doses of naproxen may also result in increased rates of nausea, particularly with smaller adolescents, reinforcing this trend.
Alternately, reduced gastric motility from higher sumatriptan doses (85/500 mg) may enable better sustained efficacy and lower headache recurrence via prolongation of naproxen Tmax. The 10/60 mg combination may better balance pain and associated-symptom relief with reduced nausea, improving overall perception of efficacy at 2 hours. Consequently, 10/60 mg might be more suitable for younger adolescents with lower body weight and short-duration migraines, whereas 85/500 mg may better suit longer-duration migraines, older adolescents, or adolescents who commonly experience recurrence.
The placebo response rate of 10% for pain-free at 2 hours postdose is low compared with most adolescent migraine trials: previous trials reveal 2-hour pain-free rates of 15% to 34%.21 This study employed strategies to reduce placebo response rates and improve treatment difference detection including the following: definitive primary end point (pain-free) that did not require assessing degrees of pain; treatment at moderate/severe (versus mild) pain for greater posttreatment pain differentiation; placebo run-in phase to identify and exclude placebo responders; selection of subjects with migraines >3 hours to avoid the impact of natural resolution; site selection emphasizing pediatric experience for improved communication with adolescents and greater generalizability of results; and site staff training, including recognizing the impact of staff/subject interaction and “no response” as a valid response. Although the single-blind phase (9% placebo response rate) may have only minimally contributed to reducing placebo response in the double-blind phase, the combination of strategies may have improved study outcomes and may aid future adolescent studies in migraine or other areas.
There was little difference in overall incidence of AEs, or incidence of moderate or severe AEs, among the active doses. The only drug-related AE with an incidence of ≥2% was muscle tightness in 3 (2%) subjects with 85/500 mg, likely representing “triptan sensations.” Among events that were drug-related, and in events emerging within 72 hours after dosing, a dose response was seen. Despite this, the AE profile is acceptable. Acute, episodic treatment of adolescent migraine for up to 12 months with 85/500 mg (>12 000 exposures) is well tolerated.40
Demographic and migraine characteristics were overall consistent with those reported in a recently completed adolescent migraine trial,40 as well as epidemiologic studies41; therefore, the results of this study may be extrapolated to the general population.
All 3 suma/nap doses were well tolerated and similarly effective. There were no serious AEs after any dose of suma/nap. Comparable efficacy and better tolerability at earlier time points with 10/60 mg may make it a better choice for shorter-duration migraines or younger adolescents. Good tolerability with better efficacy at later time points may make 85/500 mg a better choice for longer-duration migraines, older adolescents, or adolescents who experience recurrence. More definitive dosing recommendations cannot be provided by the authors; the 85/500 mg dose has been well tolerated for up to 12 months.
Treatment of adolescent migraine remains a significant unmet medical need. This study constitutes the first large-scale, placebo-controlled evidence for acute relief of adolescent migraine pain and associated symptoms with an oral medication. This efficacy evaluation, along with the pharmacokinetics and long-term safety studies, and in conjunction with regulatory review, may serve as the basis for an additional treatment option for adolescent with migraines.
We thank Janette Daves for assistance in managing the study; Susan VanMeter, MD for oversight of safety monitoring; and the adolescent subjects who participated.
- Accepted January 18, 2012.
- Address correspondence to Frederick J. Derosier, DO, GlaxoSmithKline, PO Box 13398, Five Moore Dr, Research Triangle Park, NC 27709-3398. E-mail:
All authors participated in the concept and/or design of the study, and/or acquisition of data, and/or analysis and interpretation of data; drafted or revised the article; and approved the article as submitted and the final version to be published.
FINANCIAL DISCLOSURE: Dr Derosier, Mr Goodman, Ms Jimenez, and Drs Granberry and Runken receive salary and stock as employees of GlaxoSmithKline, the maker of sumatriptan and naproxen sodium combination tablet (sumatriptan/naproxen sodium); Dr Lewis received a research grant from Bristol-Meyers Squibb, served as a consultant for study design only (Merck), and served as a consultant for study design and received a research grant from Astra Zeneca and GlaxoSmithKline; Dr Hershey participated on advisory boards for GlaxoSmithKline, Allergan, and MAP Pharmaceuticals, and received grant support and contracts from the National Institutes of Health, National Institute of Neurological Disorders and Stroke, Endo Pharmaceutical, and GlaxoSmithKline; Dr Winner served as a consultant for Allergan, GlaxoSmithKline, Merck, MAP Pharmaceuticals, Zogenix, participated in Speaker’s Bureau (Allergan, GlaxoSmithKline, Merck, Zogenix), and received support for clinical trials (Allergan, GlaxoSmithKline, Merck, MAP Pharmaceuticals, Novartis, Pfizer); Dr Pearlman served as a consultant for GlaxoSmithKline, Merck, and Allergan, participated in Speaker’s Bureau for Merck, Allergan, and GlaxoSmithKline, and received support for clinical trials from GlaxoSmithKline, Merck, Astra Zeneca, Union Chimique Belge, and Neuralieve; Dr Rothner served as a consultant to GlaxoSmithKline, AstraZeneca, Merck, and MAP Pharmaceuticals and received research funding from GlaxoSmithKline and AstraZeneca; and Dr Linder participated on advisory boards and received funding from GlaxoSmithKline, Merck, Lilly, and AstraZeneca.
FUNDING: At the initiation of the study, POZEN was the Investigational New Drug sponsor of sumatriptan and naproxen sodium, which was transferred to GlaxoSmithKline after marketing approval in the United States (April 2008). GlaxoSmithKline financed, developed, and managed this study under both sponsors.
- Lewis D,
- Ashwal S,
- Hershey A,
- Hirtz D,
- Yonker M,
- Silberstein S,
- American Academy of Neurology Quality Standards Subcommittee,
- Practice Committee of the Child Neurology Society
- Farkas V
- Winner P,
- Rothner AD,
- Saper J,
- et al
- Ahonen K,
- Hämäläinen ML,
- Rantala H,
- Hoppu K
- Ueberall M
- Lewis DW,
- Winner P,
- Hershey AD,
- Wasiewski WW,
- Adolescent Migraine Steering Committee
- Ahonen K,
- Hämäläinen ML,
- Eerola M,
- Hoppu K
- Hämäläinen ML,
- Hoppu K,
- Valkeila E,
- Santavuori P
- ↵AXERT (almotriptan malate) Product Information. April 2009
- Hämäläinen ML,
- Hoppu K,
- Santavuori P
- Winner P,
- Prensky A,
- Linder S,
- DeBussy S,
- Asgharnejad M
- Evers S,
- Marziniak M,
- Frese A,
- Gralow I
- Silberstein SD,
- Mannix LK,
- Goldstein J,
- et al
- Taylor FR,
- Heiring J,
- Messina E,
- et al
- Lipton RB,
- Dodick DW,
- Adelman JU,
- et al
- Silberstein S,
- Lipton RB,
- Goldstein J,
- et al
- Headache Classification Committee of the International Headache Society
- Kimel M,
- Hsieh R,
- McCormack J,
- Burch SP,
- Revicki DA
- Walls CM,
- Berges AC,
- Dalton ZX,
- Lener SE,
- Derosier FJ
- Bigal ME,
- Lipton RB,
- Winner P,
- Reed ML,
- Diamond S,
- Stewart WF
- Copyright © 2012 by the American Academy of Pediatrics