Abstract
Objective: To investigate the clinical profile of rofecoxib, a long-acting (≈17-hour half-life) selective cyclo-oxygenase-2 inhibitor, for the acute treatment of migraine.
Methods: A randomized, double-blind, placebo-controlled, parallel-group study was conducted. Patients age ≥18 treated a moderate or severe migraine headache with placebo (n = 182), rofecoxib 25 mg (n = 183), or rofecoxib 50 mg (n = 192). The primary efficacy measure was headache relief (mild or no pain) 2 hours after dose.
Results: The proportions of patients with migraine headache relief at 2 hours after dose were 34.3% for placebo, 54.0% for rofecoxib 25 mg (p < 0.001 vs placebo), and 56.7% for rofecoxib 50 mg (p < 0.001 vs placebo). Rofecoxib 25 and 50 mg were superior to placebo in providing pain freedom at 2 hours, 24-hour sustained headache relief, and 24-hour sustained pain freedom; in reducing photophobia, phonophobia, nausea (50 mg only), and functional disability at 2 hours after dose; and in improving some quality-of-life scores over 24 hours. More patients on rofecoxib 50 mg reported adverse events (39.6%) than patients on rofecoxib 25 mg (26.8%) or placebo (23.6%) regardless of drug relatedness; however, the incidences of drug-related adverse events were similar between treatment groups. These adverse events were generally mild or moderate in severity. The most commonly reported adverse events were dry mouth, dizziness, somnolence, nausea, dyspepsia, paresthesia, and asthenia, with similar incidences between treatment groups.
Conclusion: Rofecoxib 25 and 50 mg were effective and generally well tolerated for the acute treatment of migraine attacks.
A variety of medications are used for acute migraine treatment1; the nonsteroidal anti-inflammatory drugs and the triptans are among the most frequently prescribed. Nonsteroidal anti-inflammatory drugs, which are inhibitors of cyclo-oxygenase, are usually relied on as first-line treatment.2,3⇓ However, nonselective nonsteroidal anti-inflammatory drugs, which inhibit both cyclo-oxygenase-1 and cyclo-oxygenase-2, are associated with gastrointestinal side effects that may reduce acceptance and long-term compliance in the acute treatment of migraine.
Rofecoxib is a long-acting (≈17-hour half-life) selective cyclo-oxygenase-2 inhibitor.4 Rofecoxib 50 mg once daily has analgesic effects similar to nonselective nonsteroidal anti-inflammatory drugs in the treatment of acute pain conditions such as dental pain, primary dysmenorrhea, and postorthopedic surgical pain.4-7⇓⇓⇓ In addition, for the treatment of dental pain, rofecoxib 50 mg is superior to oxycodone/acetaminophen 5/325 mg and codeine/acetaminophen 60/600 mg and is similar to oxycodone/acetaminophen 10/650 mg.8-10⇓⇓ At lower doses, rofecoxib is used to treat chronic conditions such as osteoarthritis and rheumatoid arthritis.4 We studied the clinical profile of rofecoxib 25 and 50 mg vs placebo for the acute treatment of migraine. The primary hypothesis tested was that rofecoxib would be more effective than placebo in providing headache relief at 2 hours after dose.
Methods.
Study population.
Men and women age 18 or over in good physical health were eligible for entry into the study as determined by the site investigator. Patients must have had on average per month one to eight migraine attacks with or without aura as defined by International Headache Society criteria11 for 6 months prior to study entry. Patients in the following categories or with the following features were excluded: patients having difficulty distinguishing migraine attacks from other types of headaches; women who were pregnant or nursing or, if of childbearing potential and sexually active, who were not willing to use effective barrier or appropriate oral contraception during the study; patients who had demonstrated hypersensitivity to or experienced a serious adverse event in response to rofecoxib or had any contraindication for the use of rofecoxib; patients who had clinical or laboratory evidence of uncontrolled hypertension or significant pulmonary, renal, hepatic, endocrine, neurologic (apart from migraine), psychiatric, or other systemic disease; and patients with a history of drug or alcohol abuse within 1 year.
Study design.
This randomized, double-blind, double-dummy, placebo-controlled, parallel-group, outpatient study was conducted at 34 sites in the United States from August 24, 2001, until May 29, 2002. A total of 627 outpatients were randomized according to a computer-generated allocation schedule to treat a single migraine attack with either placebo tablets (visually matching the rofecoxib 25-mg and 50-mg tablets), rofecoxib 25-mg tablet plus placebo tablet (visually matching the rofecoxib 50-mg tablet), or rofecoxib 50-mg tablet plus placebo tablet (visually matching the rofecoxib 25-mg tablet), upon experiencing a moderate or severe migraine headache that was not resolving spontaneously and provided they had not taken any prohibited medications in the prior 24 hours. These prohibited medications were also not allowed within 2 hours after taking study tablets and included nonsteroidal anti-inflammatory drugs (including selective cyclo-oxygenase-2 inhibitors), ergot derivatives, triptans, any analgesic, opiates, barbiturates, antiemetics, or sedating medications. Nonsteroidal anti-inflammatory drugs (including selective cyclo-oxygenase-2 inhibitors) were also prohibited within 24 hours of dosing. Patients who did not obtain headache relief (mild or no pain) at 2 hours after dose were permitted to take their usual rescue medication at that time and any time thereafter. The allocation schedule was generated by a statistician at Merck Research Laboratories according to in-house blinding conditions. There was a one-to-one-to-one ratio of allocations between the three treatment groups, and a blocking size of six was used with stratification by study center. Study drug was dispensed in sequentially numbered pairs of bottles. To preserve the blind, patients’ treatment assignments were kept in a sealed envelope until the database was locked. Approvals were obtained from the necessary regulatory bodies and ethical review committees, and informed consent was signed by each patient who participated in the study. The trial was conducted in accordance with the Declaration of Helsinki and conformed to Good Clinical Practices.
Assessment of efficacy and tolerability.
Patients recorded the following measurements on a diary card immediately prior to taking test medication (0 hour) and at 0.5, 1, 1.5, 2, 3, and 4 hours after the initial dose: 1) headache severity on a 4-point scale (no pain, mild pain, moderate pain, severe pain); 2) presence or absence of associated migraine symptoms (photophobia, phonophobia, nausea, vomiting); and 3) functional disability on a 4-point scale. Patients with headache relief or pain freedom at 2 hours after dose documented, respectively, any moderate or severe headache recurrence or return to mild, moderate, or severe headache from 2 to 24 hours after dose. The use of rescue medication taken between 2 and 24 hours after study dose intake were also recorded. At the end of the 24-hour treatment period or when the migraine had resolved, patients completed a 24-Hour Migraine Quality-of-Life Questionnaire, which assessed quality of life over five domains including work function, social function, energy/vitality, migraine symptoms, and feelings/concerns.12 Each domain included three questions, and each question was answered on a 7-point scale, with 1 indicating maximum impairment of quality of life and 7 indicating no impairment. At the pre- and posttreatment visits, a history and physical examination, vital signs, and laboratory screening were performed. Adverse events were recorded by the patient on the diary card and assessed by the investigator.
Statistical analysis.
The primary efficacy endpoint was headache relief (improvement of moderate or severe pain at baseline to mild or no pain) at 2 hours after dose. The primary efficacy analysis was based on a modified intention-to-treat approach that included all randomized patients who were treated with study drug and who had at least one assessment of headache severity within 2 hours after dose (n = 538). The last-observation-carried-forward (LOCF) method was used for missing data as long as a postbaseline assessment prior to 2 hours was available; however, the baseline data were not used for imputation and were not carried forward. Treatment comparisons for the primary endpoint were performed using logistic regression models with terms for treatment, baseline headache severity, and geographic region within the United States (region I: California and Nevada; region II: Midwest and West; region III: Southeast; region IV: Northeast). With 168 patients in each treatment group, the study had 95% power to detect a 20-percentage-point difference for migraine headache relief at 2 hours after dose between rofecoxib 50 mg and placebo (e.g., 55% for rofecoxib 50 mg vs 35% for placebo). A closed step-down procedure was employed to avoid multiplicity, whereby rofecoxib 50 mg was tested first against placebo, followed by rofecoxib 25 mg. Pain freedom, sustained headache relief, sustained pain freedom, and associated symptoms were analyzed using the same methods as for the primary endpoint. Functional disability was analyzed using a cumulative logistic regression, time to rescue medication was assessed using Cox regression, and the 24-Hour Migraine Quality-of-Life domains were assessed using an analysis-of-variance model; all models included terms for treatment, baseline headache severity, and geographic region. The 24-Hour Migraine Quality-of-Life data were also globally assessed using the O’Brien method for multiple endpoints.13 All analyses presented were prespecified, and p values were two sided and considered statistically significant at the ≤0.05 level, except for analyses looking at covariate interactions, where significance was set at the ≤0.1 level. If a treatment-by-covariate interaction was significant, it was further explored by pairwise comparisons of treatments to assess if the interaction was qualitative using the Gail and Simon method.14 All 557 patients who took study medication were included in the safety analyses. The treatment groups were compared in a pairwise fashion using the Fisher Exact Test, without adjustment for multiplicity, for the percentage of patients with adverse experiences, drug-related adverse experiences, and serious adverse experiences.
Results.
Characteristics of patients and treated migraine attacks.
A total of 627 patients were randomized, of whom 557 took study medication and 538 completed the study per protocol. The 87 patients who discontinued the study included 70 patients who did not take study medication (either because they did not experience a moderate or severe migraine headache or chose not to treat a moderate or severe migraine headache with study drug during the 3 months of the study) and 17 patients who were lost to follow-up and may have taken study drug but did not return to the clinic. A study flow chart is presented in figure 1. The baseline demographic characteristics of the 557 treated patients, along with characteristics of the treated migraine attacks at baseline, are displayed in table 1 and were similar between treatment groups, except that somewhat more patients in the placebo group had severe headaches than in the rofecoxib groups.
Figure 1. Study flow chart.
Table 1 Baseline patient and migraine attack characteristics
Efficacy.
Of the 557 patients who were treated, 19 had no diary data. A total of 538 patients were therefore included in the 2-hour analysis of headache relief, and only 12 (2.2%) had missing data at 2 hours. For these 12 patients, data from earlier time points were used (LOCF method).
Both doses of rofecoxib were effective and statistically superior to placebo in the acute treatment of a migraine attack (table 2). The time course of headache relief up to 2 hours in each treatment group is shown in figure 2.
Table 2 Efficacy endpoints
Figure 2. Percentage of patients reporting migraine headache relief within 2 hours of dosing. *p < 0.05; **p < 0.01; ***p < 0.001, vs placebo.
Rofecoxib 25 mg and 50 mg were also statistically superior to placebo on a variety of secondary endpoints over multiple time points, including pain freedom, photophobia, phonophobia, nausea, functional disability, and use of rescue medication (see table 2). The percentages of patients with moderate or severe headache recurrence 2 to 24 hours after dose after having achieved headache relief at 2 hours after dose were 31.7% in the placebo group, 29.5% in the rofecoxib 25-mg group, and 15.1% in the rofecoxib 50-mg group. Because recurrence data are difficult to interpret in isolation from data on initial response and cannot be formally analyzed (owing to the fact that only those patients who respond at 2 hours can have a recurrence), alternative measures have been proposed to evaluate maintenance of an initial response.15 These include 24-hour sustained headache relief (headache relief at 2 hours, no recurrence from 2 to 24 hours, and no use of rescue medication from 2 to 24 hours) and 24-hour sustained pain freedom (pain freedom at 2 hours, no recurrence from 2 to 24 hours, and no use of rescue medication from 2 to 24 hours). Both doses of rofecoxib were superior to placebo on these latter measures. Rofecoxib 25 mg was superior to placebo on one of the five quality-of-life domains over 24 hours, whereas rofecoxib 50 mg was superior to placebo on all five domains (see table 2). Both doses of rofecoxib were superior to placebo for the comparison over all five quality-of-life domains (see table 2).
The earliest time points at which each dose of rofecoxib provided a statistically significant (p ≤ 0.05) benefit over placebo, based on measurements at prespecified fixed time points of 0.5, 1, 1.5, 2, 3, and 4 hours after dose, were as follows: 1) headache relief: rofecoxib 25 mg = 1 hour (33.5% for rofecoxib vs 14.9% for placebo), rofecoxib 50 mg = 0.5 hour (13.4 vs 5.7%); 2) pain freedom: rofecoxib 25 mg = 2 hours (19.9 vs 8.0%), rofecoxib 50 mg = 1.5 hours (12.8 vs 4.0%); 3) photophobia: rofecoxib 25 mg = 1.5 hours (68.8 vs 79.9%), rofecoxib 50 mg = 0.5 hour (82.3 vs 90.8%); 4) phonophobia: rofecoxib 25 mg = 2 hours (52.8 vs 64.0%), rofecoxib 50 mg = 1.5 hours (52.8 vs 68.4%); 5) nausea: rofecoxib 25 mg = 3 hours (21.0 vs 33.7%), rofecoxib 50 mg = 1 hour (38.9 vs 55.4%); 6) functional disability: rofecoxib 25 mg = 1 hour (10.8 vs 5.7% with normal function), rofecoxib 50 mg = 0.5 hour (6.4 vs 2.3% with normal function). Differences after 2 hours are difficult to interpret because rescue medication was allowed from that point onward.
Treatment effects across several prespecified covariates were tested by evaluating treatment-by-covariate interactions for the primary efficacy endpoint of headache relief. These analyses were exploratory, and no adjustments for the multiplicity of comparisons were made. The covariates included age, gender, race, baseline headache severity, presence or absence of migraine aura, concomitant migraine prophylactic medication, concomitant oral contraceptives, menstruation (±3 days), dysmenorrhea, and prior response to nonsteroidal anti-inflammatory drugs. No statistically significant (p ≤ 0.1) treatment-by-covariate interactions were observed except for prior response to nonsteroidal anti-inflammatory drugs (p = 0.029) and geographic region (p = 0.058). Further analysis showed that none of the interactions was qualitative (p > 0.1) when comparing each pair of treatments.
Tolerability.
The proportions of patients with one or more clinical adverse events (whether related or not to study drug) were similar in the rofecoxib 25-mg (26.8%) and placebo (23.6%) groups (p = 0.547) (table 3). However, more patients had a clinical adverse event in the rofecoxib 50-mg group (39.6%) than in the placebo (p = 0.001) and rofecoxib 25-mg (p = 0.009) groups. In each group, >90% of these adverse events were mild or moderate in intensity. The most common clinical adverse events (≥2% incidence in at least one treatment group) were dry mouth, dyspepsia, nausea, asthenia, dizziness, paresthesia, and somnolence and were similar between treatment groups (all with incidences <7%). The incidences of drug-related adverse events were similar between treatment groups (see table 3). There were two serious adverse experiences and no deaths. One patient in the placebo group had intervertebral disc herniation and sciatica. Another patient in the rofecoxib 50-mg group had asymptomatic deep vein thrombosis, discovered in the same leg where she had had a prior deep vein thrombosis, and concurrent phlebitis during a routine physical examination 8 days after dosing with rofecoxib; this patient was not on concomitant anticoagulation or antiplatelet therapy despite this history and a prior carotid endarterectomy. Both serious adverse experiences were considered definitely not related to study drug by the investigator at each site.
Table 3 Summary of clinical adverse events
As this was a single-dose study, there were no patients who discontinued treatment because of an adverse experience. There were no clinically relevant changes in blood pressure, pulse, or laboratory measures, although these measurements were typically obtained 3 to 7 days after administration of study drug.
Discussion.
This multicenter study of 557 migraine patients demonstrated that rofecoxib 25 mg and 50 mg were effective and generally well-tolerated acute therapies in patients experiencing migraine attacks with or without aura. Following treatment of a moderate or severe migraine headache, both doses of rofecoxib demonstrated superiority over placebo on most clinically important endpoints at 2 hours after dose. These outcome measures reflect various characteristics of acute migraine therapy that are important in gauging overall treatment success. One such characteristic is the efficacy in reducing pain at 2 hours after dose, measured by migraine headache relief and pain freedom, in which both doses of rofecoxib were effective.
Rapid efficacy is an important predictor of patient satisfaction with acute migraine therapy.16,17⇓ In this respect, significant differences from placebo were observed as soon as 1 hour after dosing with rofecoxib 25 mg and 30 minutes after dosing with rofecoxib 50 mg, although the absolute percentages were smaller than at the 2-hour time point and only a minority of patients had relief at the earlier time points. Sustained efficacy is another desirable characteristic of acute migraine treatment. Twenty-four-hour sustained headache relief and 24-hour sustained pain freedom have been proposed as stringent outcome measures that combine aspects of initial response at 2 hours after dose, worsening of headache after initial improvement, and use of rescue medication.15 Both doses of rofecoxib were significantly more effective than placebo on these outcome measures, although only a minority of patients had sustained headache relief or sustained pain freedom.
Migraine is a syndrome that includes symptoms such as photophobia, phonophobia, nausea, and vomiting in addition to headache pain. Treatment of headache without improvement in associated migraine symptoms would represent an incomplete response. Rofecoxib 25 mg and 50 mg significantly relieved most of these associated migraine symptoms compared with placebo within 2 hours of dosing. Even where statistical significance was not observed (e.g., for vomiting, where the small number of events makes the data difficult to interpret), the trend was clearly in the same direction.
Although relief of pain and the associated symptoms of migraine are key to successful migraine therapy, improvement in functional disability and a patient’s quality of life are other important outcome measures that encompass many different aspects of patient well-being. Both doses of rofecoxib improved functional disability, by 30 minutes after dosing with rofecoxib 50 mg and 1 hour after dosing with rofecoxib 25 mg. In addition, rofecoxib 50 mg significantly improved patients’ quality-of-life scores over 24 hours on all five domains of the migraine-specific quality-of-life scale, whereas rofecoxib 25 mg significantly improved the quality-of-life score on one of the five domains.
There was no significant difference on most endpoints when comparing rofecoxib 50 mg vs rofecoxib 25 mg. However, when comparing each rofecoxib dose with placebo, 50 mg was superior to placebo at earlier time points (typically 30 minutes earlier) than rofecoxib 25 mg for multiple endpoints, including migraine headache relief; pain freedom; relief of photophobia, phonophobia, and nausea; and improvement in functional disability.
Both rofecoxib 25 mg and 50 mg were generally well tolerated. The overall incidence of clinical adverse events was greater in the rofecoxib 50-mg treatment group than the 25-mg and placebo groups. Review of the data indicated that this overall difference was due to small differences in various nonserious adverse events of mild to moderate intensity. There was no significant difference between the rofecoxib 25-mg and placebo groups. The most commonly reported adverse events (incidence of ≥2%) included dry mouth, dizziness, somnolence, nausea, dyspepsia, paresthesia, and asthenia. Importantly, >90% of the clinical adverse events in each treatment group were of mild or moderate intensity, and there were no significant differences in drug-related adverse events between any of the treatment groups, supporting the contention that rofecoxib 25 and 50 mg were generally well tolerated.
Acknowledgments
Supported by Merck Research Laboratories.
The authors acknowledge the Protocol 161 Clinical Study Group, which consisted of the following participating investigators: Marshall Block, Phoenix, AZ; Stanley Block, Bardstown, KY; Jan Brandes, Nashville, TN; Scott Carlson, Spokane, WA; Paul Chervinsky, North Darmouth, MA; Larry Cowan, Thornville, OH; Merle Diamond, Chicago, IL; John Doria, San Diego, CA; Mildred Farmer, St. Petersburg, FL; Brian Geary, Moorestown, NJ; Harry Geisberg, Anderson, SC; Gary Gerard, Toledo, OH; Jerome Goldstein, San Francisco, CA; Daniel Grosz, Northridge, CA; Jon Heiser, Newport Beach, CA; Hubert Leonard, Portland, OR; Ivan Login, Charlottesville, VA; Antoinette Mangione, Philadelphia, PA; William McEntee, Sarasota, FL; Francis McGee, Jr., Richmond, VA; Ruth Nurnberg, Richmond, VA; Francis O’Donnell, Westerville, OH; Michael Peveler, Louisville, KY; George Rederich, Redondo Beach, CA; Ralph Richter, Tulsa, OK; Marvin Rorick, Crestview Hills, KY; John Schaeffer, Las Vegas, NV; David Seminer, Fair Oaks, CA; Stephen Silberstein, Philadelphia, PA; Richard Singer, Pembroke Pines, FL; Stuart Stark, Alexandria, VA; J Christopher Stringer, Manlius, NY; Nina Stuccio–White, Moorestown, NJ; Stewart Tepper, Stamford, CT; Paul Winner, West Palm Beach, FL; Barbara Zedler, Richmond, VA.
Footnotes
Drs. Silberstein, Tepper, Brandes, Diamond, Goldstein, and Winner have received grant support and speaking honoraria from Merck & Co., Inc. Drs. Diamond, Goldstein, and Winner have received consulting fees from Merck & Co., Inc. S. Venkatraman and F. Vrijens and Drs. Malbecq, Lines, Visser, Reines, and Yuen are employees of Merck & Co., Inc., and hold stock options and equity positions.
- Received April 18, 2003.
- Accepted in final form December 30, 2003.
References
Disputes & Debates: Rapid online correspondence
REQUIREMENTS
If you are uploading a letter concerning an article:
You must have updated your disclosures within six months: http://submit.neurology.org
Your co-authors must send a completed Publishing Agreement Form to Neurology Staff (not necessary for the lead/corresponding author as the form below will suffice) before you upload your comment.
If you are responding to a comment that was written about an article you originally authored:
You (and co-authors) do not need to fill out forms or check disclosures as author forms are still valid
and apply to letter.
Submission specifications:
- Submissions must be < 200 words with < 5 references. Reference 1 must be the article on which you are commenting.
- Submissions should not have more than 5 authors. (Exception: original author replies can include all original authors of the article)
- Submit only on articles published within 6 months of issue date.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- Submitted comments are subject to editing and editor review prior to posting.