Abstract
Objective: The study was carried out to assess the efficacy and tolerability of rizatriptan orally disintegrating tablet (ODT) for treating acute migraine in patients who are non-responders to sumatriptan.
Background: Many migraineurs report dissatisfaction with sumatriptan efficacy. It is unclear whether sumatriptan 100 mg non-responders will respond to other triptans.
Methods: This was a randomized, placebo-controlled, double-blind study in adults with >1-year history of ICHD-II (International Classification of Headache Disorders, second edition) migraine who reported that they generally do not respond to sumatriptan (≥50% unsatisfactory response). In the baseline phase, participants treated a single moderate/severe migraine attack with open-label generic sumatriptan 100 mg. Those who continued to experience moderate/severe pain at two hours post-dose were eligible to enter the double-blind treatment phase, during which participants treated three migraine attacks in crossover fashion (two with rizatriptan 10-mg ODT, one with placebo) after being randomly assigned to one of three treatment sequences (1 : 1 : 1 ratio). The primary endpoint was two-hour pain relief.
Results: A total of 102 (94%) participants treated at least one study migraine. Pain relief at two hours was significantly greater with rizatriptan compared with placebo (51% vs. 20%, p < .001). Response rates also favored rizatriptan on two-hour pain freedom (22% vs. 12%, p = .013) as well as 24-hour sustained pain relief (38% vs. 14%, p < .001) and sustained pain freedom (20% vs. 11%, p = .036). Treatment was generally well tolerated.
Conclusion: Rizatriptan 10-mg ODT was superior to placebo at providing two-hour pain relief and two-hour pain freedom in the treatment of acute migraine in those who do not respond to sumatriptan 100 mg. Rizatriptan was generally well tolerated in this population.
Introduction
The triptans, selective serotonin receptor (5-HT1B/1D) agonists, have established efficacy and tolerability for the treatment of acute migraine and are considered first-line therapy for moderate-to-severe migraine headache (1). Sumatriptan was the first member of this class to be developed and has recently become available in a generic oral formulation. A review of randomized clinical trials conducted with sumatriptan (IMITREX, GlaxoSmithKline, Research Triangle Park, NC) indicated that approximately 40% of patients in those studies did not respond to either 50 mg or 100 mg of oral sumatriptan (2,3), with response defined as migraine pain relief at two hours post-treatment.
Non-response to sumatriptan does not predict non-response to other triptans. Several randomized, placebo-controlled, clinical trials have demonstrated efficacy for almotriptan (4), naratriptan (5) and eletriptan (6) in sumatriptan (50 mg) non-responders. Additionally, an open-label, crossover trial also demonstrated efficacy for zolmitriptan and rizatriptan in sumatriptan (50 mg) non-responders (7).
Rizatriptan has not been rigorously tested in a randomized, double-blind, placebo-controlled clinical trial in a sumatriptan non-responder population. No triptan has been tested in a population of non-responders to sumatriptan 100 mg. Rizatriptan could reasonably be expected to be effective in this population, as it demonstrates some favorable pharmacokinetic characteristics compared with sumatriptan, such as a shorter Tmax and higher oral bioavailability (8,9). In the context of an unsatisfactory patient response, it is common practice to try a different triptan. Further study of rizatriptan could provide clinicians with an additional treatment option for migraineurs who do not respond to sumatriptan. Consequently, the purpose of the current study was to assess the efficacy and tolerability of rizatriptan 10 mg ODT (orally disintegrating tablet/oral lyophilisate) when administered to patients who have had an unsatisfactory response to sumatriptan 100 mg due to a lack of efficacy.
Methods
Participants
The target population was men and women ≥18 years of age with a history of migraine longer than one year, with or without aura, as defined by International Headache Society (International Classification of Headache Disorders, second edition [ICHD-II]) criteria (10). Eligible participants had, by self-report, ≥2 to ≤8 moderate or severe migraine attacks per month during the two months before screening and had to be able to clearly distinguish migraine attacks from other types of headache. Subjects with >15 headache-days/month or who had taken medication for acute headache on more than 10 days/month in any of the previous three months were excluded.
Eligible participants also reported that they generally did not respond to treatment with sumatriptan and, at a minimum, that they have demonstrated a ≥50% unsatisfactory response (e.g. no pain relief at two hours post-dose) to sumatriptan across their total migraine history (including failure to respond to at least two administrations of sumatriptan). Subjects who reported failing to respond to three or more triptans were excluded.
Participants were permitted to take medications for migraine prophylaxis if they had been on a stable dose for at least three months prior to screening. Participants agreed to discontinue using monoamine oxidase inhibitors and propranolol one month before receiving study medication. Subjects who used opioids as primary migraine therapy or who used daily opioids for migraine or other chronic pain syndromes were excluded. In addition, daily analgesics taken for any reason were not permitted (except for aspirin ≤325 mg/day). Participants were allowed to continue selective serotonin reuptake inhibitors (SSRI) or serotonin norepinephrine reuptake inhibitors (SNRI) as long as the daily dose had been constant for the previous three months and would not change during the study.
Female participants of childbearing potential agreed to use adequate contraception during the study. Subjects with ischemic heart disease, uncontrolled hypertension, coronary artery vasospasm (including Prinzmetal’s variant angina) or other significant underlying cardiovascular disease were excluded.
Study design
This was a multicenter, randomized, double-blind, placebo-controlled, crossover, multiple attack, acute treatment trial conducted in the United States (25 sites) from June 2009 through January 2010 (protocol 087). The study was funded by Merck & Co., Inc. A Scientific Advisory Committee comprised of headache experts and Merck Research Laboratories scientists developed the protocol, formulated the statistical analysis plan, analyzed and interpreted the data, and authored this report. The study was conducted in accordance with principles of Good Clinical Practice and was approved by the appropriate institutional review boards and regulatory agencies. All participants gave written informed consent before undergoing any study procedures.
During the baseline phase of the study, participants treated a single qualifying moderate (Grade 2) or severe (Grade 3) migraine attack with a single dose of open-label, generic sumatriptan 100 mg. Participants who failed to respond to sumatriptan (i.e. continued to experience moderate or severe pain at two hours post-dose) were classified as non-responders and were eligible to enter the double-blind, randomized, treatment phase of the study. Participants could take their own headache medication as rescue, including analgesics (e.g. non-steroidal anti-inflammatory drugs [NSAIDs]), anti-emetics, opiates or other medications not explicitly excluded. The only triptan allowed as rescue during the baseline phase was sumatriptan from the participant's own supply.
As part of the baseline assessment, patients completed three questionnaires. The Generalized Anxiety Disorder-7 (GAD-7) questionnaire is designed to assess patients for signs and symptoms of generalized anxiety disorder (11). The Patient Health Questionnaire-9 (PHQ-9) is designed to assess patients for signs and symptoms of depression (12). The Migraine Assessment of Current Therapy (Migraine-ACT) is designed to assess how well an acute migraine medication is working and whether a change in therapy might be indicated (13). Subjects who had evidence of suicidality (PHQ-9 item 9 score ≥2) or depression of moderately severe intensity (PHQ-9 total score ≥15) were excluded from the study and referred for appropriate care.
The treatment phase of the study used a three-period, double-blind, crossover design. Eligible participants received blinded study medication and were instructed to treat three qualifying moderate or severe migraine attacks. Two attacks were to be treated with rizatriptan 10 mg ODT and one attack with placebo; the sequence of these treatments was randomly assigned in a 1 : 1 : 1 ratio, using an allocation schedule generated by the study sponsor. Participants were randomly assigned to a treatment sequence using an electronic interactive voice response system (IVRS). All study personnel, participants, and the sponsor remained blinded to treatment allocation throughout the study. Participants administered a single dose of study medication when they experienced a moderate or severe (Grade 2 or 3) migraine headache, recording the date, time and severity of the attack in a paper diary before taking study medication. For each qualifying migraine attack, if the participant continued to experience a Grade 2 or 3 migraine at two hours post-dose, or if the migraine recurred two hours after administering study medication, the participant could take his/her own headache medication as rescue; the only triptan allowed as rescue during the treatment phase was rizatriptan from the participant's own supply.
Participants recorded subjective assessments in a paper migraine diary during the 48 hours after taking study medication. Assessments recorded were pain severity, presence or absence of associated symptoms, degree of functional disability at specified time intervals, use of rescue medication and migraine headache recurrence. Treatment of a new migraine attack with study medication required a pain-free period of at least 48 hours since the last attack. Participants returned to the study site within two to seven days after treatment of the third study migraine attack for review of the diary, assessment of medication compliance and safety monitoring. Study personnel contacted participants by telephone 14 days after their last dose of study medication to assess ongoing and new adverse events (AEs).
Efficacy and safety endpoints
The primary efficacy endpoint was the proportion of treated attacks resulting in pain relief at two hours post-dose. Participants rated headache severity (0 = no pain; 1 = mild pain; 2 = moderate pain; 3 = severe pain) at baseline, 0.5, 1, 1.5, 2, 4, 24 and 48 hours post-dose and recorded it in the study diary. Participants recorded migraine-associated symptoms (photophobia, phonophobia, nausea and vomiting) and rating of functional disability (0 = normal, 1 = mildly impaired, 2 = severely impaired, 3 = unable to do activities, requires bed rest) at the same time points. Participants also noted use of rescue medication. Participants completed the Overall 24-Hour Assessment of Study Medication at 24 hours post-dose to rate satisfaction; this measure uses a seven-point scale ranging from 1 (completely satisfied) to 7 (completely dissatisfied). Safety was monitored by collection of AEs recorded in the migraine diary and assessment of vital signs.
The primary efficacy hypothesis for the study was that rizatriptan is superior to placebo, as measured by the proportion of treated attacks resulting in pain relief (reduction in headache severity from Grade 3/2 at baseline to Grade 1/0) at two hours post-dose. The primary safety hypothesis was that rizatriptan is generally well tolerated when administered for acute migraine in patients who are sumatriptan non-responders. The secondary hypothesis was that rizatriptan is superior to placebo as measured by the proportion of treated attacks resulting in pain freedom (reduction in headache severity to Grade 0) at two hours post-dose. Other pain endpoints included sustained pain relief from 2–24 hours (pain relief at two hours, with no administration of rescue medication and no occurrence thereafter of a moderate/severe headache within 24 hours after dosing with study medication) and 2–48 hours post-dose, sustained pain freedom from 2–24 hours and 2–48 hours post-dose and total migraine freedom (pain freedom and absence of photophobia, phonophobia, nausea and vomiting) at two hours post-dose. Additional endpoints were “normal” ratings of functional disability at two hours post-dose, satisfaction with treatment at 24 hours post-dose, absence and elimination of migraine-associated symptoms at two hours post-dose, and use of rescue medication up to 24 hours post-dose. Treatment effects across subgroups defined by baseline scores on the GAD-7, PHQ-9 and Migraine-ACT questionnaires were also assessed.
Statistical methods
The study sponsor performed the statistical analyses of the study data. After medical/scientific review was performed, protocol violators identified and data declared final and complete, the sponsor unblinded the official, final database. The primary efficacy analysis used the Full Analysis Set (FAS) population. For each two-hour endpoint (pain relief, pain freedom, ratings of functional disability, absence of symptoms), inclusion in the FAS population required that a participant had at least one evaluable attack. An individual attack was evaluable if the participant administered study treatment and had both a baseline headache severity measurement and at least one post-dose efficacy measurement before or including the two-hour time point. The safety analyses used the All-Patients-as-Treated population, defined as all randomized participants who received at least one dose of study medication.
Missing data in a given treatment period for headache severity, ability to perform daily activities, and associated symptoms were imputed by carrying forward the preceding values in the same phase within each attack up to four hours post-dose. Baseline values were not carried forward to impute missing post-treatment data. Missing values were not imputed for the 24-hour or 48-hour time points. No data were imputed across treatment periods.
A generalized linear mixed model, with categorical terms for treatment (rizatriptan and placebo), period (1, 2 and 3) and baseline headache severity within each period (moderate or severe), with age as a continuous covariate, was used to compare the proportion of treated attacks resulting in pain relief at two hours post-dose (primary efficacy analysis). An unstructured covariance matrix was used to model the correlation among repeated measurements within a patient. The treatment comparison was based on the model-derived odds ratio and the associated 95% confidence interval (CI). The study was considered positive provided the primary hypothesis was satisfied (i.e. p ≤ .05 in favor of rizatriptan 10 mg). Similar methodology was used for the secondary and most of the exploratory endpoints; summary statistics were provided for the functional disability endpoint.
Safety and tolerability were assessed by clinical review of all relevant parameters including adverse AEs and vital signs measurements. The primary approach was a summary of AEs associated with treatment of the first migraine attack, that is, AEs that occurred after treatment for the first attack (regardless of whether or not the participant took rescue medication) up to 14 days after the initial dose or until treatment with study medication of a subsequent attack, whichever was earlier. A supportive approach was a summary of the number and percentage of participants with AEs tabulated by treatment groups (rizatriptan and placebo) across the entire study. Among them, all AEs recorded within 14 days of any administration of rizatriptan (via initial migraine treatment, or due to treatment of a previous attack) were attributed to the rizatriptan group. AEs that occurred within 14 days of treatment with placebo (but not within 14 days of treatment with rizatriptan) were attributed to placebo.
The safety results were assessed via point estimates with 95% CIs provided for between-group comparisons using the Miettinen and Nurminen method (14). The 95% CIs provided are for descriptive purposes only and should not be considered hypothesis testing for AEs and predefined limits of change.
The study needed to randomize ∼108 participants to achieve 80% power (two-sided, α = .05) to demonstrate the primary hypothesis that rizatriptan is superior to placebo as measured by the proportion of treated attacks resulting in pain relief at two hours post-dose. This is based on the following assumptions: attacks are independent within each participant, the true response rate of pain relief at two hours post-dose is 17% for placebo and at least 35% for rizatriptan, and the rate of non-evaluable attacks is no more than 30%.
Results
Participant accounting
Of 194 patients screened, 159 patients met eligibility criteria and entered the baseline phase, during which they were to treat a migraine attack with generic sumatriptan 100 mg (Figure 1). Of these 159 patients, six were screen failures, 44 (27.7%) responded to sumatriptan treatment and were excluded and 109 non-responders were randomized into the treatment phase of the study. A total of 102 (93.6%) randomized participants treated at least one migraine attack with double-blind study medication: 98 treated all three attacks, three treated two attacks and one treated only one attack. Of the seven participants who did not treat a migraine, one did not have a qualifying event, and six were lost to follow-up. Overall, 102 participants treated 202 migraine attacks with rizatriptan 10-mg ODT, and 99 participants treated 99 migraine attacks with placebo.
Patient accounting. Suma = sumatriptan. Riza = rizatriptan.
Participant demographics and baseline characteristics
Baseline patient demographics and migraine history by dosing strategy (all treated patients)
Riza = rizatriptan 10 mg. Pbo = placebo. SD = standard deviation. NSAID = non-steroidal anti-inflammatory drug.
Participant response to the following question: “On average, how often do your moderate/severe migraine headaches respond to triptan medication?” However, all participants responded “yes” to “Have you ever treated a migraine with a triptan?”
n = number of treated patients.
Baseline migraine characteristics (all treated attacks)
n = number of treated attacks.
Efficacy
Rizatriptan 10 mg ODT was significantly better than placebo at providing pain relief at two hours in this population of sumatriptan non-responders (51% of attacks vs. 20%, p < .001; Figure 2), and a significantly greater proportion of attacks resulted in sustained pain relief for 2–24 hours and 2–48 hours with rizatriptan than with placebo. Rizatriptan was also significantly better than placebo at providing pain freedom at two hours (22% of attacks vs. 12%, p = .013), as well as sustained pain freedom from 2–24 hours and total migraine freedom at two hours (Figure 2). All other efficacy endpoints, with the exception of vomiting and sustained pain freedom from 2–48 hours, significantly favored rizatriptan 10 mg ODT (Figures 3 and 4).
Pain endpoints. * Primary endpoint. Bars represent the 95% confidence interval for the percentage (model-based). Additional efficacy endpoints. Bars represent the 95% confidence interval for the percentage (model-based). Migraine-associated symptoms. Bars represent the 95% confidence interval for the percentage (exact).
Response to rizatriptan and placebo (pain relief at two hours post-treatment) was also analyzed according to subgroups defined by baseline scores on the GAD-7 (normal or mild, moderate, or severe anxiety), PHQ-9 (normal or mild, moderate, moderate-severe or severe depression), and the Migraine-ACT (“no” response to 0, 1, 2, 3 or 4 questions). No appreciable differences were observed in response rates across these subgroups (data not shown).
Safety
Adverse event summary (all attacks, within 14 days of treatment, ≥2% in one or more treatment groups)
Patients took at least one dose of study medication. It is possible for one patient to be counted twice (once in each treatment group). Although a patient may have had two or more adverse events of the same type, the patient is counted only once for that type of adverse event.
Adverse events occurring within 14 days of administration of rizatriptan are attributed to rizatriptan group even if placebo was administered more recently. Each patient was to treat two attacks with rizatriptan and one with placebo in the study.
Discussion
Non-responsiveness to oral triptans is an important issue in migraine treatment and may be associated with various triptan-specific and patient-specific factors (7,9,15,16). Drug-related factors affecting response may be associated with pharmacokinetic characteristics such as Tmax and oral bioavailability, which show some variability across the class. Patient responsiveness may also be influenced by variability in drug absorption, whether drug is used early or late in an attack, or by complicating tolerability issues, among other possible factors. Following non-response to a particular triptan, it is common practice for clinicians to prescribe an alternative triptan for the patient. This is a useful strategy for managing migraine treatment because patient responsiveness to a specific triptan appears to be largely consistent across attacks (5,9,15,16). Therefore, the short-term search for an efficacious triptan for any given patient likely leads to more sustained treatment success.
The current study represents the first prospective, randomized controlled trial of rizatriptan ODT in a population whose migraine does not respond to treatment with sumatriptan 100 mg. Rizatriptan demonstrated efficacy in this population, as the response to rizatriptan was statistically superior to that of placebo on pain relief and pain freedom at two hours and sustained pain freedom from 2–24 hours, as well as on other endpoints examined in the study. Rizatriptan also appeared to be generally well tolerated in this population, given that few AEs were reported overall, and the majority of these were mild. In consideration of these AE rates, it is important to note that twice as many attacks were treated during the study with rizatriptan as with placebo. Additionally, the reporting was conservative in that AEs were attributed to study medication (rizatriptan or placebo) up to 14 days following administration, or until treatment of a subsequent qualifying migraine, even if rescue medication was used during this interval.
The response rates observed in this study appeared to be somewhat lower than those observed in previous studies of rizatriptan. A meta-analysis of previously conducted trials in the general migraine population reported rizatriptan response rates of 70% on two-hour pain relief, 40% on two-hour pain freedom, and 25% on 24-hour sustained pain freedom (9). Such results for the current study might be expected due to the fact that this study population was comprised of experienced triptan users, and a population reporting non-response to one triptan could be more refractory to treatment overall compared with the general migraine population.
Several previous studies have demonstrated efficacy for almotriptan (4), naratriptan (5), eletriptan (6), and zolmitriptan and rizatriptan (7) in a sumatriptan 50 mg non-responder population, and these have been compared in review articles (15,16). The results across these studies are difficult to compare directly due to differences in study design and data analysis. The studies of almotriptan and naratriptan were most similar in design to the current study in that all of these were randomized, double-blind, and placebo-controlled, and each used prospective confirmation of sumatriptan non-response as an enrichment strategy. Response rates to almotriptan were 47.5% for two-hour pain relief, 33.3% for two-hour pain freedom, and 20.9% for 24-hour sustained pain freedom, with response significantly greater than that to placebo on each endpoint. Response rates to naratriptan were 25% for two-hour pain relief and 6% for two-hour pain freedom, although significance was not achieved on the pain freedom endpoint. Interestingly, the current study was the first to enrich using sumatriptan 100 mg, but the sumatriptan response rate was similar to that observed with sumatriptan 50 mg used in the almotriptan and naratriptan studies (approximately 30%). The fact that a response to sumatriptan was observed in some percentage of screened patients might not be surprising given that this population was initially identified according to patient recall of a history of general non-responsiveness, thereby subject to potential recall bias and without control over the extent of prior experience with sumatriptan. Ultimately, this is a lower response rate than would be expected in the general migraine population (4,5). Although the almotriptan, naratriptan, and current study appear to all have successfully enriched for sumatriptan non-responders, these studies also shared some of the same design limitations in that administration of sumatriptan was not double-blinded and no sumatriptan administration was included in the randomized phase of the studies.
Migraine patients may demonstrate a variable response to any particular member of the triptan class (15,16). Consequently, clinicians must manage migraine treatment with triptans by offering non-responders alternate triptans that might work well for them. This will be an important part of the cost-benefit consideration as members of the triptan class begin to enter the generic market. The current study has demonstrated that treatment with rizatriptan is an effective and well-tolerated option for migraine patients who do not respond to sumatriptan.
Conclusion
Rizatriptan 10-mg ODT was superior to placebo at providing two-hour pain relief and two-hour pain freedom in the treatment of acute migraine in participants who do not respond to sumatriptan 100 mg. Rizatriptan was generally well tolerated in this population.
Footnotes
Declaration of conflicting interests
Drs Taylor, Newman and Friedman have received research grants and honoraria from Merck. Drs Seeburger, Ge, Hustad, Hewitt, Ho, Zhang, Fan and Connor, and Ms LaSorda are employed by Merck and may own stock in the company.
Funding
This study was funded by Merck & Co., Inc.
Acknowledgements
The authors thank Lyn Harper Mozley for assistance with execution of the study. The authors thank Jennifer Pawlowski for editing and submission of the manuscript.
MAXALT protocol 087 investigators (all USA):
R. Barger, Boston, MA; D. Brune, Peoria, IL; M. Chehrenama, Alexandria, VA; J. Clark, Charlottesville, NC; M. Conway, Albuquerque, NM; D. Ellison, West Chester, OH; L. Ford, Papillion, TE; M. Freeman, Greensboro, SC; M. Kimmel, Melbourne, FL; W. Koltun, San Diego, CA; D. Kudrow, Santa Monica, CA; N. Mathew, Houston, TX; S.D. Miller, N. Dartmouth, MA; M. Mollen, Phoenix, AZ; L. Mueller, Stratford, NJ; A. Puopolo, Milford, MA; L. Robbins, Northbrook, IL; J. Rubino, Raleigh, NC; G. Ruoff, Kalamazoo, MI; J. Saper, Ann Arbor, MI; T. Smith, St. Louis, MO; C. Strout, Mount Pleasant, SC; P.G Martin, Winston-Salem, NC; D. Ross, Lauderdale Lakes, FL; R. Shiwach, DeSoto, TX.