Improvements in Patient-Reported Outcome Measures Through 64 Weeks’ Rimegepant Treatment for Migraine: Analysis of a Randomized Controlled Trial with Open-Label Extension

Abstract

Purpose

The current exploratory analysis of a Phase 2/3 randomized controlled trial focusses on patient-reported outcome measures from a 64-week study evaluating rimegepant for the preventive treatment of migraine in the USA.

Methods

Participants with migraine were randomly assigned to rimegepant 75 mg or placebo every other day during a 12-week double-blind treatment (DBT) phase. During a 52-week open-label extension (OLE) phase all participants took rimegepant 75 mg every other day and additionally on nonscheduled dosing days if needed. Migraine-Specific Quality of Life Questionnaire (MSQ) v.2.1 (including Role Function Restrictive [RFR] domain) and Migraine Disability Assessment Scale (MIDAS) were assessed throughout.

Results

741 participants (mean [standard deviation] age 41.2 [13.06] years, 82.7% female) were treated with ≥1 dose of study drug in the DBT phase, and 603 participants were treated with rimegepant in the OLE phase. At Week 12 (end of DBT phase), for the rimegepant and placebo groups, respectively, mean (95% CI) improvements from baseline in MSQ RFR score were 18.6 (16.24, 20.94) and 15.0 (12.60, 17.37), and improvements in MIDAS Total score were −13.0 (−16.57, −9.52) and −12.0 (−15.52, −8.54). At Week 64 (end of OLE phase, all participants taking open-label rimegepant), for participants originally assigned DBT rimegepant and DBT placebo, respectively, mean (95% CI) improvements from baseline in MSQ RFR score were 32.6 (29.68, 35.42) and 33.2 (30.03, 36.29), and improvements from baseline in MIDAS Total Score were −25.4 (−30.23, −20.48) and −24.4 (−28.29, −20.59).

Conclusion

Notable numerical benefit for rimegepant versus placebo was observed at the end of the DBT phase (Week 12) for MSQ RFR domain score. During open-label treatment with rimegepant, improvements from baseline in MSQ domain scores and MIDAS Total Score were clinically meaningful at Week 64. Participants who received placebo in the DBT phase achieved similar benefit when they moved into the OLE phase as those who took rimegepant from the start of the study.

Trial Registration

Clinicaltrials.gov NCT03732638.

Plain Language Summary

Migraine affects every part of daily life, including home, work, education, and relationships. Quality of life is decreased, and more frequent migraine attacks cause greater migraine-related disability. Patient-reported results are important in clinical trials of treatments for migraine. This report looked at the patient-reported results from a 64-week study assessing rimegepant for the preventive treatment of migraine. Participants with migraine took rimegepant or placebo (a dummy pill) every other day for 12 weeks, then everyone could take rimegepant up to once each day for the next 52 weeks. During the study, the participants completed questionnaires including the Migraine-Specific Quality of Life Questionnaire (MSQ) and the Migraine Disability Assessment Scale (MIDAS). When researchers looked at the MSQ results at Week 12, participants generally had better scores compared with before treatment, but the improvement seemed bigger for those taking rimegepant than those taking placebo. The researchers found that treatment with rimegepant for up to 64 weeks was linked to improvements in MSQ and MIDAS results compared with before treatment. This means the study participants reported their quality of life and migraine-related disability improved when they took rimegepant for the preventive treatment of migraine for up to 64 weeks.

Graphical Abstract

Introduction

Migraine is estimated to affect approximately 14% of the global population¹ and is two to three times more common in women than men.² The burden associated with migraine is significant. In terms of disability, migraine ranks eighth among all conditions worldwide.³ The high level of disability associated with migraine is reflected in an estimate that, globally in 2023, individuals with migraine collectively experienced 40.9 million years of life lived with disability.³ The substantial productivity losses and health economic costs associated with migraine impact the patient, their employer, and society.^4–6

Migraine can impact every aspect of the patient’s life, including daily activities; personal, family, and social relationships; finances; education; and career.^7–9 Compared with unaffected individuals, people with migraine have worse quality of life,¹⁰ and disability is increased with more frequent migraine attacks.^11–13 The impact of migraine can extend into the interictal period between attacks, resulting from anxiety about the next attack, avoidance or withdrawal from activities and commitments, and persistent, sometimes subtle, interictal symptoms.^14–16 This ongoing challenge underlines the importance of effective management and the inclusion of relevant patient-reported outcome measures (PROMs) in clinical trials.^17,18

Calcitonin gene-related peptide (CGRP) monoclonal antibodies are injectable preventive treatments for migraine.^19–22 Rimegepant is an orally administered small molecule CGRP receptor antagonist effective for both the acute^23–27 and preventive^28,29 treatment of migraine. The pivotal phase 2/3 trial evaluating the efficacy and safety of rimegepant for the preventive treatment of migraine in the USA comprised a 12-week, placebo-controlled double-blind treatment (DBT) phase followed by an optional 1-year open-label extension (OLE) phase.^28,30 The current secondary report focusses on the PROMs from both the double-blind and open-label phases of that study.

There are currently limited data assessing the effect of rimegepant on quality of life.^29,31,32 The aim of these analyses was to evaluate the impact of up to 64 weeks’ rimegepant treatment on PROMs in people with migraine. The hypothesis was that rimegepant would improve Migraine-Specific Quality of Life Questionnaire (MSQ) v.2.1 domain scores (including Role Function Restrictive) and Migraine Disability Assessment Scale (MIDAS) scores, and be associated with high levels of Satisfaction with Medication.

Materials and Methods

Trial Design

This study reports prespecified exploratory analyses of a phase 2/3, randomized, double-blind, placebo-controlled trial (study 305, NCT03732638) evaluating rimegepant for the preventive treatment of migraine, conducted in the USA. During an initial 12-week DBT phase, participants received rimegepant 75 mg tablet or placebo every other day. Subject to satisfactory laboratory test results, participants who continued to meet all inclusion/exclusion criteria and were compliant with the eDiary at the end of the DBT phase could enter the optional OLE phase lasting up to 52 weeks. During the OLE phase, participants took rimegepant 75 mg every other day but could also take rimegepant 75 mg on nonscheduled dosing days if they experienced a migraine attack. The OLE phase permits longer-term assessments than would be feasible in a DBT setting, although open-label treatment is associated with inherent bias. Full methodology and primary results for the study have been reported previously.²⁸

The study was conducted in compliance with the Declaration of Helsinki, Good Clinical Practice, and all applicable local regulations.²⁸ The protocol for this multicenter study was approved by a central Institutional Review Board (IRB) (Advarra IRB, Columbia, MD, USA), and by the IRB at one study site (MedStar Georgetown Headache-Georgetown University, McLean, VA, USA).²⁸ The IRB approval numbers from Advarra (current IRB organization 0000635 and IRB registration number 00000971) and MedStar/Georgetown University (current IRB organization 0000193, IRB registration committee not available due to administrative changes) are not retrievable due to administrative changes (the study was sponsored by Biohaven which was acquired by Pfizer in October 2022). All participants provided written informed consent before taking part in the study.

Participants

As reported previously,²⁸ eligible participants were adults ≥18 years of age with a history of migraine (with or without aura) for at least 1 year, consistent with the International Classification of Headache Disorders 3rd edition (ICHD-3) migraine criteria.³³ Participants had to have experienced the onset of migraine before 50 years of age, with untreated migraine attacks lasting 4–72 hours, on average. They needed to have 4–18 migraine attacks per month (4 weeks) of moderate or severe pain intensity in the 3 months prior to screening, at least 6 migraine days during the observation period, and no more than 18 headache days during the observation period. Participants had to be able to distinguish migraine attacks from tension-type or cluster headache. Those with contraindications for triptans could be included if they met all other criteria.

Key exclusion criteria were basilar migraine and hemiplegic migraine. Participants were allowed to remain on one prophylactic migraine medication (excluding CGRP antagonists) if the dose had been stable for at least 3 months prior to the observation period and was likely to remain stable through the study. Concomitant use of CGRP antagonists was prohibited. Permitted rescue medications included aspirin, nonsteroidal anti-inflammatory drugs, acetaminophen, antiemetics, and muscle relaxants. In addition, triptans were allowed as a rescue medication during the DBT phase but were prohibited during the OLE phase.

Assessments

Study visits occurred at baseline (randomization) and Weeks 2, 4, 8, and 12 during the DBT phase; at Weeks 14 and 16 and then every 4 weeks through Week 64 during the OLE phase. An eDiary was used daily to record headache frequency and severity over time, as reported previously.²⁸ Additionally, PROMs related to quality of life, functional disability, and satisfaction with treatment were assessed and are the focus of this analysis.

The MSQ v.2.1³⁴ was administered by study investigators and completed, on paper, at baseline and Weeks 12, 24, and 64. The 14-item MSQ v.2.1 is a validated tool that assesses the impact of migraine during the past 4 weeks on three domains: Role Function Restrictive, Role Function Preventive, and Emotional Function. For each item the possible responses are (1) none of the time, (2) a little bit of the time, (3) some of the time, (4) a good bit of the time, (5) most of the time, and (6) all of the time. Domain scores range 0–100, and an increase in MSQ score indicates improvement. When comparing treatment groups, the minimal important differences in MSQ score that can be considered clinically relevant are 3.2 for Role Function Restrictive, 4.6 for Role Function Preventive, and 7.5 for Emotional Function.³⁵ For within-group comparisons, clinically meaningful improvements are at least 5.0 points for Role Function Restrictive, 5.0–7.9 for Role Function Preventive, and 8.0–10.6 for Emotional Function.³⁵

The MIDAS³⁶ was administered by study investigators and completed, on paper, at baseline and Weeks 12, 24, and 64. The MIDAS is a retrospective, 5-item questionnaire that measures headache-related disability as the number of days lost due to headache from paid work or school, household work, and nonwork activities over a 3 month-period. Total Score, Absenteeism (missed days from work, school, household work, or nonwork activities due to headache), and Presenteeism (days at paid work, school, or household work in which productivity was reduced by at least half due to headache) were calculated. A decrease in MIDAS score indicates improvement. A decrease of ≥4.5 points³⁷ to 5.0-points³⁸ represents a clinically important change for patients.

Satisfaction with Medication data were collected via eDiary at Weeks 12, 24, and 64, based on a participant-rated, 7-point response option, single-question scale that measures satisfaction with the study medication. This scale has not been validated. Participants were asked “How satisfied are you with the study medication in treating your migraines?” The response options were as follows: completely satisfied, could not be better; very satisfied; somewhat satisfied; neither satisfied nor dissatisfied; somewhat dissatisfied; very dissatisfied; and completely dissatisfied, could not be worse.

Statistical Analysis

Analyses were conducted for treated participants, that is, those who received at least one dose of study drug (DBT or OLE). For each treatment group at Weeks 12, 24, and 64, mean (95% CI) changes from baseline in MSQ and MIDAS scores were prespecified exploratory endpoints. Post hoc completer analyses were conducted for changes in MSQ and MIDAS scores for the subgroup of participants who completed the study (ie, with nonmissing values at Week 64). Satisfaction with Medication ratings at Weeks 12, 24, and 64 were prespecified exploratory endpoints and were summarized by treatment group as percentage (95% CI) of participants for each category of response. These exploratory descriptive analyses were based on observed data without imputation for missing data. There were no corrections for multiplicity. Statistical comparisons were not prespecified and were not conducted for these exploratory analyses. All analyses were performed using SAS statistical software (SAS Institute Inc., Cary, NC, USA) Version 9.4 or higher.

Results

Disposition and Demographics

In total, 741 participants were treated with ≥1 dose of study drug in the DBT phase (rimegepant n=370, placebo n=371). Disposition and participant flow during the DBT phase were reported previously.²⁸ Subsequently, 603 participants were treated with rimegepant in the OLE phase (including 302 participants randomly assigned to rimegepant in the DBT phase [DBT rimegepant/OLE rimegepant] and 301 participants randomly assigned to placebo in the DBT phase [DBT placebo/OLE rimegepant]). Of the 603 treated with rimegepant in the OLE phase, 428 (71.0%) completed the OLE phase (including 209 DBT rimegepant/OLE rimegepant and 219 DBT placebo/OLE rimegepant). The reasons for not completing the OLE phase (n=175) were adverse event (n=19), death (n=2), failed eligibility for OLE due to laboratory test results (n=2), lack of efficacy (n=4), lost to follow-up (n=32), noncompliance (n=29), other (n=2), physician decision (n=18), pregnancy (n=4), protocol deviation (n=5), study terminated by sponsor (n=1), and withdrawal by participant (n=57).

As described previously,²⁸ the study population (n=741) was 82.7% female, 81.5% white, 15.0% Black or African American, and 3.5% were other races (including Asian, American Indian or Alaska Native, Native Hawaiian or Other Pacific Islander, or multiple). Ethnicity was Hispanic or Latino for 27.4% and not Hispanic or Latino for 72.6%. Participants had a mean (SD) age of 41.2 (13.06) years and body mass index of 26.4 (3.82) kg/m².²⁸ The primary migraine type was without aura for 60.2% of participants and with aura for 39.8%, and 24.0% had a history of chronic migraine. Mean (SD) age at migraine onset was 21.2 (11.00) years, the participants had a history of 7.8 (2.71) moderate to severe migraine attacks per month, and the duration of untreated migraine attacks was, on average, 29.0 (21.26) hours.²⁸ In the 4-week observation period prior to study drug treatment, efficacy-evaluable participants (n=695) reported a mean (SD) of 10.1 (3.1) migraine days per month.²⁸

Exposure

During the DBT phase, participants (n=741) spent a mean (SD) of 11.1 (3.02) weeks on treatment and, on average, used a mean (SD) of 13.8 (1.68) tablets (rimegepant or placebo) per month. During the OLE phase, participants (n=603) spent a mean (SD) of 42.5 (16.14) weeks on rimegepant and, on average, used a mean (SD) of 14.6 (2.45) rimegepant tablets per month.

MSQ

Baseline mean (SD) scores for MSQ Role Function Restrictive, Role Function Preventive, and Emotional Function domains were 50.9 (18.10), 65.3 (20.82), and 57.8 (27.14), respectively, for participants assigned to rimegepant and 49.7 (18.94), 64.1 (20.46), and 55.8 (27.00), respectively, for participants assigned to placebo.

Improvements from baseline in MSQ Role Function Restrictive domain scores were maintained through the 64-week study, with greater improvement observed with increased time on treatment (Figure 1A and Table S1). At the end of the DBT phase (Week 12), the mean (95% CI) improvement from baseline in Role Function Restrictive domain scores was 18.6 (16.24, 20.94) for participants assigned to rimegepant and 15.0 (12.60, 17.37) for participants assigned to placebo (Table 1). At Week 64 when all participants were receiving rimegepant, the mean (95% CI) improvement from baseline in Role Function Restrictive domain scores was 32.6 (29.68, 35.42) for the DBT rimegepant/OLE rimegepant group and 33.2 (30.03, 36.29) for the DBT placebo/OLE rimegepant group.

Figure 1.

Mean change from baseline in MSQ domains: (A) Role Function Restrictive, (B) Role Function Preventive, and (C) Emotional Function.

Notes: Treated participants. For sample sizes and additional data, see Table S1.

Abbreviations: BL, baseline; DBT, double-blind treatment; CI, confidence interval; MSQ, Migraine-Specific Quality of Life Questionnaire v.2.1; OLE, open-label extension; PBO, placebo; RIM, rimegepant.

Table 1.

Mean Changes from Baseline at Week 12 in MSQ and MIDAS Scores for Rimegepant and Placebo Groups

	DBT Rimegepant (N=370)				DBT Placebo (N=371)
	Score		Change from Baseline		Score		Change from Baseline
	n	Mean (SD)	n	Mean (95% CI)	n	Mean (SD)	n	Mean (95% CI)
MSQ Role Function Restrictive
Baseline	369	50.9 (18.10)	–	–	370	49.7 (18.94)	–	–
Week 12	308	69.9 (19.78)	308	18.6 (16.24, 20.94)	306	65.9 (21.44)	305	15.0 (12.60, 17.37)
MSQ Role Function Preventive
Baseline	369	65.3 (20.82)	–	–	370	64.1 (20.46)	–	–
Week 12	308	80.2 (18.82)	308	14.4 (12.18, 16.55)	306	77.1 (20.10)	305	11.5 (9.16, 13.79)
MSQ Emotional Function
Baseline	369	57.8 (27.14)	–	–	370	55.8 (27.00)	–	–
Week 12	308	77.4 (23.67)	308	18.9 (16.20, 21.68)	306	72.2 (25.97)	305	15.1 (12.14, 17.98)
MIDAS Total Score
Baseline	368	37.8 (30.83)	–	–	371	37.4 (30.76)	–	–
Week 12	307	23.3 (34.59)	306	−13.0 (−16.57, −9.52)	306	24.3 (26.43)	306	−12.0 (−15.52, −8.54)

Notes: Treated participants.

Abbreviations: CI, confidence interval; DBT, double-blind treatment; MSQ, Migraine-Specific Quality of Life Questionnaire v.2.1; MIDAS, Migraine Disability Assessment; SD, standard deviation.

Similarly, improvements from baseline in MSQ Role Function Preventive domain scores (Figure 1B) and MSQ Emotional Function domain scores (Figure 1C) were also maintained throughout the entire study, both DBT and OLE phases (Table S1). At the end of the DBT phase (Week 12), the mean (95% CI) improvements from baseline in Role Function Preventive domain scores were 14.4 (12.18, 16.55) for participants assigned to rimegepant and 11.5 (9.16, 13.79) for participants assigned to placebo, and improvements from baseline in Emotional Function domain scores were 18.9 (16.20, 21.68) and 15.1 (12.14, 17.98), respectively (Table 1). At Week 64 when all participants were receiving rimegepant, the mean (95% CI) improvements from baseline in Role Function Preventive and Emotional Function domain scores, respectively, were 24.7 (21.86, 27.58) and 29.3 (25.75, 32.82) for the DBT rimegepant/OLE rimegepant group and 25.5 (22.28, 28.67) and 31.0 (27.13, 34.83) for the DBT placebo/OLE rimegepant group.

MIDAS

Baseline mean (SD) MIDAS Total Score was 37.8 (30.83) for participants assigned to rimegepant and 37.4 (30.76) for participants assigned to placebo. For Absenteeism and Presenteeism, respectively, baseline mean (SD) scores were 19.5 (18.46) and 18.3 (15.80) for participants assigned to rimegepant and 19.0 (17.82) and 18.4 (15.96) for participants assigned to placebo.

Improvement from baseline in MIDAS Total Score was maintained through the 64-week study, with greater improvement observed with increased time on treatment (Figure 2A and Table S2). At the end of the DBT phase (Week 12), the mean (95% CI) improvement from baseline in MIDAS Total Score was −13.0 (−16.57, −9.52) for participants assigned to rimegepant and −12.0 (−15.52, −8.54) for participants assigned to placebo (Table 1). At Week 64 when all participants were receiving rimegepant, the mean improvement from baseline in MIDAS Total Score was −25.4 (−30.23, −20.48) for the DBT rimegepant/OLE rimegepant group and −24.4 (−28.29, −20.59) for the DBT placebo/OLE rimegepant group.

Figure 2.

Mean change from baseline in MIDAS (A) Total Score, (B) Absenteeism, and (C) Presenteeism.

Notes: Treated participants. For sample sizes and additional data, see Table S2.

Abbreviations: BL, baseline; DBT, double-blind treatment; CI, confidence interval; MIDAS, Migraine Disability Assessment; OLE, open-label extension; PBO, placebo; RIM, rimegepant.

Similarly, improvements from baseline in MIDAS Absenteeism (Figure 2B) and MIDAS Presenteeism (Figure 2C) scores as well as all individual MIDAS scores (Table S2) were maintained throughout the study. At Week 64, the mean (95% CI) improvements from baseline in Absenteeism and Presenteeism scores, respectively, were −13.2 (−15.48, −10.87) and −12.2 (−15.13, −9.24) for the DBT rimegepant/OLE rimegepant group and −13.0 (−15.30, −10.67) and −11.5 (−13.39, −9.52) for the DBT placebo/OLE rimegepant group.

Satisfaction with Medication

High levels of satisfaction with medication were maintained through the 64-week study, with the percentage of participants who were completely or very satisfied increasing with time on treatment (Figure 3 and Table S3). At the end of the DBT phase (Week 12), 49.2% (95% CI 42.9, 55.5) of participants assigned to rimegepant (n=240) were completely or very satisfied, compared with 39.3% (95% CI 33.4, 45.6) of participants assigned to placebo (n=239). At Week 24 when all participants were receiving rimegepant, 79.8% (95% CI 73.4, 84.9) of the DBT rimegepant/OLE rimegepant group (n=188) and 75.1% (95% CI 68.6, 80.7) of the DBT placebo/OLE rimegepant group (n=197) were completely or very satisfied. At Week 64, 87.4% (95% CI 80.9, 92.0) of the DBT rimegepant/OLE rimegepant group (n=143), 80.0% (95% CI 72.7, 85.7) of the DBT placebo/OLE rimegepant group (n=145), and 83.7% (95% CI 79.0, 87.5) of participants overall (n=288) were completely or very satisfied.

Figure 3.

Percentage of participants completely or very satisfied with medication.

Notes: Treated participants. The remainder of participants were somewhat satisfied, neither satisfied nor dissatisfied, somewhat dissatisfied, very dissatisfied, or completely dissatisfied, could not be worse (see Table S3).

Abbreviations: DBT, double-blind treatment; CI, confidence interval; OLE, open-label extension; PBO, placebo; RIM, rimegepant.

Post Hoc Completer Analyses

For the subgroup of participants with nonmissing values at Week 64, the changes from baseline in MSQ Role Function Restrictive domain score (Figure S1A and Table S4) and MIDAS Total Score (Figure S1B and Table S4) through the study were similar to those for all participants (Figures 1A and 2A). At Week 12, the treatment effect (separation between rimegepant and placebo groups) was greater for completers compared with all participants, although confidence intervals overlapped. At Week 12, the mean (95% CI) improvement from baseline in MSQ Role Function restrictive domain score was 18.9 (16.10, 21.74) for completers assigned to rimegepant (n=190) and 13.8 (10.90, 16.72) for completers assigned to placebo (n=198). At Week 12, the mean (95% CI) improvement from baseline in MIDAS Total Score was −15.4 (−19.40, −11.31) for completers assigned to rimegepant (n=189) and −11.0 (−14.86, −7.10) for completers assigned to placebo (n=199).

Discussion

These data show the improvements in migraine-related quality of life and disability as well as the high levels of satisfaction with rimegepant taken every other day, and up to once daily on days when medication was not scheduled, for up to 64 weeks. Compared with baseline, the benefits of rimegepant were observed at the end of the DBT phase (Week 12) and continued to improve over the course of the subsequent 52 weeks of open-label rimegepant treatment. These data extend the evidence base for rimegepant, increasing understanding of the impact of treatment on quality of life and productivity in people with migraine.

The impaired quality of life of the study population at baseline (mean MSQ Role Function Restrictive domain scores of 50.9 in the rimegepant group and 49.7 in the placebo group) was typical for a population with migraine.^39,40 Quality of life measures may be expected to take some time to respond to treatment as patients gradually experience the benefits on their lifestyle, and greater improvements from baseline were observed with increased time on treatment. The improvements in quality of life observed through the current study were clinically meaningful. At the end of the DBT phase (Week 12), the changes from baseline in MSQ Role Function Restrictive domain scores for both rimegepant and placebo groups (18.6 and 15.0, respectively) exceeded the meaningful within-group threshold of 5.0.³⁵ The numerical difference between the rimegepant group and the placebo group at Week 12 was 3.6, which exceeded the between-group minimal important difference of 3.2,³⁵ supporting the secondary endpoint result reported previously.²⁸ Changes from baseline in MSQ Role Function Restrictive domain scores continued to exceed the meaningful within-group threshold³⁵ at Weeks 24 and 64. Changes from baseline in MSQ Role Function Preventive and Emotional Function domain scores exceeded meaningful within-group thresholds³⁵ at all timepoints.

The study population reported severe disability^41,42 at baseline (mean MIDAS Total scores of 37.8 in the rimegepant group and 37.4 in the placebo group). From Week 12 through Week 64, the changes from baseline in MIDAS Total Score for both groups exceeded the within-group minimal important change threshold of a 4.5-point³⁷ to 5.0-point decrease.³⁸ However, the numerical difference between the rimegepant group and the placebo group at Week 12 (1.0) was small, in line with the secondary endpoint result reported previously.²⁸ This might indicate no real difference. However, complicating factors potentially include a large placebo response, delayed functional adaptation despite reduced migraine frequency, and recall bias. The MIDAS questionnaire administered at 12 weeks assessed headache-related disability over the previous 3-month rimegepant treatment period. It is possible that rimegepant treatment initiation requires time before it starts to impact disability outcomes. Conversely, the MIDAS may require more than 12 weeks to show a treatment difference in this population. Once attack frequency starts to decrease it can take time for patients to evaluate the impact on their lifestyle. An instrument with a shorter recall period may be preferable to reduce the inevitable recall bias when asking patients to consider symptoms in the past 3 months. There are few comparable data evaluating treatment effects (CGRP antagonist vs placebo) on MIDAS at 3 months, although atogepant treatment resulted in a greater reduction from baseline in MIDAS Total Score than placebo.⁴³

Improvements in both MIDAS Absenteeism and Presenteeism scores were maintained in both treatment groups throughout the study. This is important considering the detrimental effect of migraine on work and productivity.⁴⁴ Further research is warranted using a dedicated instrument, such as the Work Productivity and Activity Impairment (WPAI)-migraine questionnaire.⁴⁵

After Week 12, participants assigned to placebo in the DBT phase could switch to open-label treatment with rimegepant, and participants assigned to rimegepant in the DBT phase could remain on rimegepant. During the OLE phase when all participants were receiving rimegepant, the mean scores from Week 24 across all PROMs suggested participants in the DBT placebo group were able to “catch up” with the participants in the DBT rimegepant group, once they commenced active treatment. At Week 24, although the differences were small, there was a greater response to open-label rimegepant in the DBT placebo/OLE rimegepant group compared with the DBT rimegepant/OLE rimegepant group, which was in contrast to the data at Week 12 (Figures 1 and 2).

As is common to all studies of pain⁴⁶ the placebo response can be considerable in migraine trials.^47,48 Like pain itself, PROMs are self-reported and subjective (that is, they cannot be objectively measured) and are similarly susceptible to the placebo effect.⁴⁹ A sizable placebo response was observed across all outcomes in the current study during the DBT phase. Placebo effects can be exacerbated by participants’ expectation that the intervention will work. Expectations can increase during longer term studies, for example due to interactions with the researchers, and experience of the benefits and/or adverse effects of the intervention.

Rates of satisfaction with medication were high throughout the study; >80% of participants overall reported being completely or very satisfied with rimegepant at Week 64. The longer the participants took rimegepant, the greater the percentage that reported being satisfied.

The trajectory of continuing improvement versus baseline on the PROMs continued through the end of the study. However, it should be noted that changes in the population through the study may have influenced the results; participants who were benefiting from treatment may be more likely to remain in the study, introducing the potential for responder bias at later timepoints. However, discontinuation rates through the OLE phase were low. Further, the post hoc completer analyses conducted for changes in MSQ Role Function Restrictive domain score and MIDAS Total Score showed patterns of improvement similar to those observed in the overall analyses, suggesting informative loss to follow-up did not lead to overestimation of long-term treatment benefits.

The current data support and extend previous findings of the benefits assessed by PROMs for rimegepant.^29,31,32 Similarly, improvements in MSQ and MIDAS at Week 12 were greater than placebo with another oral CGRP antagonist indicated for migraine prevention, atogepant, in a double-blind trial.⁴³ Matching-adjusted indirect comparisons of rimegepant with atogepant found a greater treatment effect for atogepant than rimegepant on MSQ.⁵⁰ Matching-adjusted indirect comparisons of rimegepant with galcanezumab and erenumab found no statistically significant differences for MIDAS, whereas for MSQ rimegepant was favorable compared with erenumab and similar to galcanezumab.⁵¹ A head-to-head, double-blind trial comparing rimegepant and galcanezumab found improvements from baseline in MSQ Role Function Restrictive domain score at Month 3 of 31.9 points in the galcanezumab group and 26.7 points in the rimegepant group.⁵²

The limitations of the study include the inherent bias associated with open-label treatment and the lack of an active comparator. There also may be responder bias, in that responders may have elected to participate in and/or continue in the OLE phase of the study at higher rates than nonresponders or low responders. The completer analysis suggests that informative loss to follow up most likely did not materially influence these results. At minimum, these results reflect the long-term benefits of rimegepant in those who stay on treatment for 64 weeks. There may have been survivorship bias at the later time points, although the endpoints were based on mean scores. There was no correction for multiplicity on these prespecified exploratory endpoints. There was a notable placebo response. The current study population, like other migraine trials, reflects the higher prevalence of migraine in women;² sex-based subgroup analyses are precluded by the smaller size of the male subgroup and would require further research. The strengths of the current study include its long-term, prospective nature and the dosing regimen during the OLE phase (every other day plus as needed on nonscheduled dosing days to a maximum of one dose of rimegepant 75 mg per calendar day), which closely reflects real-world practice.

Conclusions

Treatment with rimegepant for up to 64 weeks was associated with marked improvements from baseline in MSQ and MIDAS, and high levels of satisfaction with medication. Notable numerical benefit versus placebo was observed at the end of the DBT phase (Week 12) for MSQ Role Function Restrictive domain score. During open-label treatment with rimegepant, improvements from baseline in MSQ domain scores and MIDAS Total Score were clinically meaningful at Weeks 24 and 64. Participants who received placebo in the DBT phase achieved similar benefit when they moved into the OLE phase as those who took rimegepant from the start of the study.

Funding Statement

This study was sponsored by Biohaven which was acquired by Pfizer in October 2022.

Data Sharing Statement

Upon request, and subject to review, Pfizer will provide the data that support the findings of this study. Subject to certain criteria, conditions and exceptions, Pfizer may also provide access to the related individual de-identified participant data. See https://www.pfizer.com/science/clinical-trials/trial-data-and-results for more information.

Author Contributions

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Disclosure

Richard B. Lipton: Research support: the National Institutes of Health and the US Food and Drug Administration; Consultant for and/or advisory board member of, or honoraria or research support: AbbVie/Allergan, Amgen, Axsome, Biohaven, Eli Lilly, GlaxoSmithKline, Lundbeck, Merck, Novartis, Pfizer, Teva, Vector, and Vedanta Research; Stock/options: Axon, Biohaven, CoolTech, and Manistee. Dawn C. Buse: Grant support and honoraria: AbbVie, Amgen, Biohaven, Collegium, Eli Lilly, Ipsen, the Headache and Migraine Policy Forum, Lundbeck, the National Headache Foundation, Pfizer, Teva, Theranica, and the US Food and Drug Administration; Editorial board: Current Pain and Headache Reports. Vincent Martin: Grant support: AbbVie, Amgen, Eli Lilly, and Teva; Consultant: AbbVie, Pfizer, and Tonix; Speaker: AbbVie and Pfizer. Dave Semel: employee of Pfizer with shares/share options. Joseph C. Cappelleri: employee of Pfizer with shares/share options. Alexandra Thiry: employee of Pfizer with shares/share options. Lucy Abraham: employee of Pfizer with shares/share options. At the time of the study, Terence Fullerton was an employee of Pfizer and has shares/share options in Pfizer. The authors report no other conflicts of interest in this work.