Abstract
Objective
To investigate whether clinical and sociodemographic factors are associated with calcitonin gene-related peptide (CGRP) induced migraine attacks.
Methods
A total of 139 participants with migraine received a 20-minute intravenous infusion of CGRP (1.5 µg/min) on a single experiment day. The incidence of CGRP-induced migraine attacks was recorded using a headache diary during the 12-hour observational period post-infusion. Univariable and multivariable regression analyses were conducted to examine potential predictors' relationship with CGRP-induced migraine attacks.
Results
CGRP-induced migraine attacks were reported in 110 (79%) of 139 participants. Univariable analysis revealed that participants with cutaneous allodynia had higher odds of developing CGRP-induced migraine attacks, compared with those without allodynia (OR, 2.97, 95% CI, 1.28 to 7.43). The subsequent multivariable analysis confirmed this association (OR, 3.26, 95% CI, 1.32 to 8.69) and also found that participants with migraine with aura had lower odds of developing CGRP-induced migraine attacks (OR, 0.32, 95% CI, 0.12 to 0.84).
Conclusion
Our results suggest that cutaneous allodynia and aura play a role in CGRP-induced migraine attacks, while other clinical and sociodemographic factors do not seem to have any noticeable impact. This indicates that the CGRP provocation model is robust, as the CGRP hypersensitivity remained unaffected despite differences among a heterogeneous migraine population.
Trial Registration: ClinicalTrials.gov Identifier: NCT04592952
Introduction
Calcitonin gene-related peptide (CGRP) has emerged as a key signaling molecule in migraine pathogenesis, attracting considerable attention in recent decades (1–3). Intravenous infusion of CGRP has been shown to induce migraine attacks in persons with migraine (4,5), and drugs against CGRP signaling are effective in the acute and preventive treatment of migraine (6–8).
The development of migraine attacks after CGRP infusion can vary from person to person, with some experiencing these attacks while others do not (4). Interestingly, around 20–40% of individuals with migraine do not experience migraine attacks following CGRP infusion, a phenomenon consistently observed in earlier experimental studies. The underlying causes of this variability remain incompletely understood, underscoring the significance of investigating potential associations between the provocation of CGRP-induced migraine attacks and specific clinical or sociodemographic factors. The establishment of such associations could provide further insights into the potential significance of these factors within the migraine pathogenesis. In future human provocation studies, the identified factors could potentially be considered as confounding variables or effect modifiers. Conversely, the absence of preconditioning factors might suggest that a person’s susceptibility to develop CGRP-induced migraine attacks relies on multiple factors, eluding straightforward identification. In either case, the subject merits investigation and might provide insights into the plethora of factors involved in the genesis of migraine pain.
In the present open-label trial, we aimed to investigate the potential relationship between clinical and sociodemographic factors and the development of CGRP-induced migraine attacks. A large sample of adult participants diagnosed with episodic and chronic migraine were enrolled. All participants underwent a semi-structured interview and received a continuous 20-minute intravenous infusion of CGRP.
Methods
The data presented in this paper was collected as part of the larger REgistry FOR Migraine (REFORM) study. Details of the REFORM study are available elsewhere (9). In brief, REFORM is a prospective, single-center study of participants who were screened for migraine at the Danish Headache Center. The protocol was approved by the relevant ethics committee (H-20047793). All participants provided written informed consent per Declaration of Helsinki principles.
Participants
Eligible participants had to be at least 18 years of age with a history of migraine with or without aura, as defined in the 3rd edition of the International Classification of Headache Disorders (ICHD-3) (10), for at least one year prior to enrollment. Furthermore, the onset of migraine had to occur before the age of 50 years, and the participants had to experience at least four monthly migraine days on average across the three months prior to study inclusion. In addition, the participants who received preventive migraine medication had to report a stable dose for at least two months prior to enrollment. Female participants also had to be on effective contraception at the time of inclusion. A complete list of inclusion and exclusion criteria has been published elsewhere (9).
Design and procedures
This was a non-randomized, open-label trial, in which the participants were assigned to receive a 20-min continuous intravenous infusion of CGRP on a single experiment day. The administered dose of CGRP (1.5 μg/min) was identical to the one used in previous provocation experiments in people with migraine, cluster headache, and post-traumatic headache (5,11,12). Of note, the participants were rescheduled if they reported any headache at the time of infusion start or had taken any acute headache medication within 24 hours of the planned infusion start.
Upon arrival on the experiment day, the participants underwent a neurologic examination and had a 12-lead electrocardiogram taken. Trained personnel conducted a semi-structured interview to record clinical and sociodemographic information. The participants were then placed in a supine position, and peripheral cannulation of the antecubital fossa was performed to obtain intravenous access.
At the time of infusion start, trained personnel recorded data on clinical features, vital signs, and adverse events using a headache diary. This procedure was repeated every 10 minutes until one-hour post-infusion start. Following this, the participants were discharged and instructed to fill out the headache diary on an hourly basis until 12 hours post-infusion start or until falling asleep.
Outcomes
The main outcome was to examine the association of 23 selected clinical or sociodemographic factors associated with CGRP-induced migraine attacks. All factors were obtained through the semi-structured interview process. The selected factors included age, sex, number of days since last attack, number of monthly migraine days, migraine with aura, current use of preventive migraine medication, medication-overuse headache, number of days since the start of the last menstruation, and cutaneous allodynia (full list of clinical and sociodemographic factors is in Table 1). Cutaneous allodynia was evaluated using the 12-item self-report Allodynia Symptom Checklist (ASC-12) (13). The range of the scale is 0 to 24, with scores of 3 to 5 indicating mild cutaneous allodynia, while scores of 6 to 8 represent moderate cutaneous allodynia, and scores of ≥9 are indicative of severe cutaneous allodynia.
Baseline characteristics and clinical characteristics after calcitonin gene–related peptide infusion.
Responder: developed migraine attack after CGRP infusion; Non-responder: did not develop migraine attack after CGRP infusion; IQR: interquartile range; SD: standard deviation; n: number; NRS: numeric pain rating scale.
Cutaneous allodynia was evaluated by Allodynia Symptom Checklist-12 (ASC-12).
Univariate analyses of clinical and sociodemographic features with a significance level of P < 0.05. None of the above analyses were significant after Bonferroni correction.
A CGRP-induced migraine attack was defined as a migraine headache occurring within the 12-hour observation period following the start of infusion. The definition was based on the fulfillment of ICHD-3 criterion C and D for migraine without aura or if it closely resembled the participant’s typical spontaneous migraine attack (10). In addition, instances where the migraine attack was treated with an acute headache medication were also classified as CGRP-induced migraine attacks.
A CGRP-induced episode of migraine aura was defined as fully reversible neurologic symptoms that met ICHD-3 criteria B and C for migraine with aura or closely resembled the participant’s usual aura symptoms (10).
Statistical analysis
Data conforming to a normal distribution are presented as the mean value, accompanied by the standard deviation (SD), whereas data not following a normal distribution are reported as the median value with the associated interquartile range (IQR). Assessment of normality was performed using the Shapiro-Wilk test. For sample sizes exceeding 30, normal distribution was assumed in accordance with the central limit theorem. Further validation of this assumption was conducted by visual inspection of the data distributions. Categorical variables are presented by the count and corresponding percentage of participants.
The univariate analysis was performed using the Mann-Whitney U test for the comparison of non-normally distributed data, while the Student’s two-sample t-test was applied for the comparison of normally distributed data. For categorical data comparisons, the Chi-squared test was used for group sizes ≥50 participants, while Fisher’s exact was applied for group sizes <50. A multivariable binomial regression analysis was carried out to identify predictors of CGRP-induced migraine attacks. The chosen covariates for this model included number of days since last attack, number of monthly migraine days, current use of preventive migraine medication, medication-overuse headache, and number of days since the start of the last menstruation. Any variables that were found significant in the univariate analyses were also included. Assumptions of the model were validated by demonstrating linear or near-linear relationships between continuous independent variables and the logit transformed migraine headache response (Online Supplemental Figure 1). For all analyses, statistical significance was set at P < 0.05. Results are provided both uncorrected and after Bonferroni correction for multiple comparisons. All statistics were performed in R version 4.1.0.
Results
The study population encompassed 139 participants diagnosed with migraine, both with and without aura. The participants were on average 44.1 years of age (SD = 18.0), and predominately female (n = 118; 85%). Male participants accounted for 15% of the study sample (n = 21). Each participant completed a screening visit and was subjected to CGRP infusion. The average body mass index of the cohort was 24.6 kg/m2 (SD = 4.4). Furthermore, the mean number of monthly headache days was 13.8 (SD, 5.8), while the mean number of monthly migraine days was 10.6 (SD, 4.3). Among the participants, 130 (94%) reported current use of triptans, and 84 (60%) were currently using preventive migraine medication. All participants were headache-free at the time of infusion start. Table 1 presents the clinical and sociodemographic factors of the study population.
CGRP-induced migraine attacks
During the 12-hour observation period after CGRP infusion, 110 (79%) of 139 participants reported experiencing migraine attacks. The median time to onset was 50 min (IQR, 20 to 240) after the infusion started (Figure 1). Among those who developed CGRP-induced migraine attacks (n = 110), 86 (78.2%) reported using acute headache medication to treat their headache. Moreover, 13 (38%) of 34 participants diagnosed with migraine with aura developed aura symptoms during the 12-hour observation period.
Baseline-Corrected Median Headache Intensity Scores after Calcitonin Gene–Related Peptide Infusion.
Predictors of CGRP-induced migraine attacks
The univariate analysis revealed a significant association between CGRP-induced migraine attacks and cutaneous allodynia, with an OR of 2.97 (95% CI, 1.28 to 7.43; P = 0.01) in the uncorrected analysis. No associations were found between CGRP-induced migraine attacks and mean monthly migraine days (OR, 1.03; 95% CI: 0.93–1.14; P = 0.60) or the number of days since the last migraine attack (OR, 0.91; 95% CI, 0.82–1.01; P = 0.08). The use of preventive migraine medication also did not demonstrate any association with CGRP-induced migraine attacks (OR, 1.32; 95% CI, 0.57 to 3.01; P = 0.51), nor did the number of days since the onset of the last menstruation (OR, 0.98; 95% CI, 0.87 to 1.05; P = 0.34). Additional variables were also investigated but were likewise found not to be associated with CGRP-induced migraine attacks (Table 2). Of note, cutaneous allodynia did not retain its statistical significance after Bonferroni correction.
Univariate analyses comparing clinical features and CGRP hypersensitivity.
Asterisk indicate the results are significant (P < 0.05).NRS: Numeric pain rating scale.
Cutaneous allodynia was evaluated by Allodynia Symptom Checklist-12 (ASC-12).
Univariate analyses of clinical and sociodemographic features with a significance level of P < 0.05. None of the above analyses were significant after Bonferroni correction.
In the multivariable regression model, cutaneous allodynia (OR, 3.26; 95% CI, 1.32 to 8.69; P = 0.013) and aura (OR, 0.32; 95% CI, 0.12 to 0.84, P = 0.02) were the only variables significantly associated with CGRP-induced migraine attacks (Table 3). This model also included monthly migraine days, number of days since last migraine attack, current use of preventive migraine medication, medication-overuse headache, and number of days since the onset of the last menstruation. Of note, both cutaneous allodynia and aura failed to retain their statistical significance after Bonferroni correction.
Multiple binomial regression analyses comparing clinical features and CGRP hypersensitivity.
Asterisk indicate the results are significant (P < 0.05).Multivariable binomial regression analyses of clinical features with a significance level of P < 0.05. None of the above analyses were significant after Bonferroni correction.
Adverse events
Following the CGRP infusion, the most commonly reported adverse events included flushing (n = 134), warm sensations (n = 133), and palpitations (n = 110). Less frequently reported adverse events included fatigue (n = 26), abdominal pain (n = 12), nasal congestion (n = 10), sensation of fulness in the ear (n = 10), concentration difficulties (n = 7), dizziness (n = 7), and lacrimation (n = 3). The adverse events linked to CGRP infusion were not specifically associated with the initiation of a migraine episode, as their onset occurred both before and during the migraine attack.
Discussion
In this open-label trial, we investigated the association of clinical and sociodemographic factors with CGRP-induced migraine attacks. Among the 139 participants diagnosed with migraine, 110 (79%) experienced migraine attacks during the 12-hour observation period after CGRP infusion. Our results suggest that the presence of cutaneous allodynia might increase the odds of experiencing CGRP-induced migraine attacks, while the presence of aura appears to reduce the odds. However, these interpretations merit careful consideration, as their statistical significance was not retained after correction for multiple comparisons. Future studies are thus needed to substantiate our findings and to better delineate the influence of these factors on hypersensitivity to CGRP in persons afflicted with migraine.
The factors contributing to CGRP-induced migraine attacks are of great scientific interest, in particular given the effectiveness of drugs against CGRP signaling for the treatment of migraine (6–8). Our findings suggest that cutaneous allodynia might be pathogenic contributor to the onset of CGRP-induced migraine attacks. Cutaneous allodynia is defined as normally innocuous stimuli being perceived as painful (14) and has been linked to various chronic pain disorders, including chronic migraine (15,16). The neurobiologic substrate of cutaneous allodynia is thought to involve central sensitization, a phenomenon wherein the central nervous system (CNS) becomes increasingly responsive to sensory stimuli, leading to amplified pain perception (17). This suggests that persons with migraine who report cutaneous allodynia might have a lowered threshold for the onset of CGRP-induced migraine attacks. The observed correlation underscores the importance of future research to elucidate the role of cutaneous allodynia in migraine pathogenesis and its potential implications for treatment with drugs targeting CGRP signaling.
Another interesting finding was the present results showing that aura might lower the odds of developing CGRP-induced migraine attacks. This puzzling observation eludes a straightforward explanation. The neurobiologic substrate of aura is thought to be cortical spreading depression (CSD), a phenomenon characterized by a slow, self-propagating waves of neuronal and glial depolarization (18). CSD is thought to modulate nociceptive messaging and activate the trigeminovascular system (18,19). Thus, it is unclear how and if the presence of aura lowers the odds of developing CGRP-induced migraine attacks. In this context, it also merits mention that the statistical significance did not survive multiple comparisons correction. Further investigations are necessary to establish any potential relationship between the occurrence of aura and CGRP-mediated migraine attacks, as well as to explore the impact of aura on CGRP-targeted signaling pathways.
The concept of migraine as a sensory threshold disease has gained increased attention in recent years (5). If so, one might assert that the responsiveness to experimental migraine inducers – such as CGRP – is greater in those with a high frequency of migraine attacks. However, our results showed no such association, as the incidence of CGRP-induced migraine attacks was comparable between those with episodic and chronic migraine (78% versus 81%, respectively). The lack of correlation between headache frequency and the initiation of migraine also remains consistent for both migraine with and without and without aura. These findings align well with previous studies, in which the induction of CGRP-induced migraine attacks was evaluated in persons with episodic and chronic migraine (5). However, the assertion of migraine being a sensory threshold disease cannot be dismissed based on our findings. Other endogenous pro-nociceptive agents are likely involved in the genesis of migraine pain (20–23), and there is no evidence to support that migraine is exclusively a CGRP-mediated disease.
Earlier studies have postulated that CGRP-induced migraine attacks might depend on various factors, such as the time elapsed since the last attack, current use of preventive migraine medication, and the number of days since the onset of the last menstruation (5). However, in the present study, no significant association was identified between these factors and the occurrence of CGRP-induced migraine attacks. It should, nonetheless, be noted that our study included few participants who were in their perimenstrual period. As such, we cannot exclude the potential importance of transient changes in sensitivity to CGRP during this period.
Considering the present results, it seems reasonable to assert that the human provocation model is quite robust, as the induction of CGRP-induced migraine attacks remained unaffected by clinical and sociodemographic data after multiple comparison correction. Further studies might explore the potential contributions of genetics, blood-based biomarkers, and neuroimaging in CGRP-induced migraine attacks. Moreover, future research should delve into the potential of responsiveness to CGRP infusion serving as predictive biomarker for the effectiveness of drugs against CGRP signaling. If so, this might prove useful in clinical practice and add therapeutic decision-making.
Strengths and limitations
The current study represents the largest human provocation study to date (n = 139). To ensure the integrity of our data, we established strict eligibility criteria: participants were required to have no headache and to abstain from using any acute headache medication within the 24 hours leading up to the CGRP infusion. Moreover, we used a semi-structured interview to collect detailed data on a range of clinical and sociodemographic characteristics.
This study also had several limitations. First, the information on headache and migraine frequency was assessed retrospectively and might be subject to recall bias. A preferable approach would have been prospective data collection using an electronic headache diary with time stamps. Second, a considerable proportion of participants reported current use of preventive migraine medication. This might have influenced the responsiveness to CGRP infusion, although no such association was found in our analysis. Third, most participants were recruited from a tertiary referral center, which might limit the generalizability of our findings to the broader population of people with migraine. Lastly, the absence of a placebo arm might have influenced participants’ expectations regarding the development of CGRP-induced migraine attacks.
Conclusion
Among persons diagnosed with migraine, our findings hint at the potential influence of cutaneous allodynia and aura on CGRP-induced migraine attacks, while other clinical or sociodemographic factors appear to have no discernible effect. However, cutaneous allodynia and aura did not retain their association after correction for multiple comparisons. This underscores the robustness of human provocation model using intravenous infusion of CGRP, as the response was unaffected by variations in clinical and sociodemographic characteristics across a diverse migraine population. Future studies should investigate whether hypersensitivity to CGRP infusion can predict the effectiveness of treatment with medications against CGRP or its receptor in persons afflicted with migraine.
Key findings
Cutaneous allodynia and aura significantly contribute to CGRP-induced migraine attacks, highlighting their crucial role in this context. Despite variations among a diverse migraine population in terms of clinical and sociodemographic factors (n = 139), the CGRP provocation model remains robust, as CGRP hypersensitivity remains unaffected.
Supplemental Material
sj-pdf-1-cep-10.1177_03331024231206375 - Supplemental material for An exploratory analysis of clinical and sociodemographic factors in CGRP-induced migraine attacks: A REFORM study
Supplemental material, sj-pdf-1-cep-10.1177_03331024231206375 for An exploratory analysis of clinical and sociodemographic factors in CGRP-induced migraine attacks: A REFORM study by Haidar M. Al-Khazali, Håkan Ashina, Rune Häckert Christensen, Astrid Wiggers, Kathrine Rose, Afrim Iljazi, Henrik W. Schytz, Faisal Mohammad Amin and Messoud Ashina in Cephalalgia
Footnotes
Acknowledgment
The authors thank all participating participants, medical students Mohammed Bakir Ahmad Lafta, Sarra Al-Khazali, and Amir Al-Saoudi, headache research laboratory technician Merete Bak Bertelsen, and headache nurse Janne Jensen for their expert assistance.
Availability of data and materials
Upon reasonable request, the corresponding author will provide the necessary data and materials to interested researchers for the purpose of academic scrutiny, reproducibility, and further scientific investigation.
Author contributions
Concept and design: MA, FMA, HA, AI, HWS.
Acquisition, analysis, or interpretation of data: HMA, RHC, HA, FMA, MA.
Drafting of the manuscript: HMA, RHC, HA, AW, KR, MA, HWS.
Critical revision of the manuscript for important intellectual content: HMA, RHC, HA, FMA, AI, AW, KR, HWS, MA.
Statistical analysis: HMA, RHC, HA.
Administrative, technical, or material support: MA, FMA, HA.
Supervision: MA, FMA, HA, HWS.
Declaration of conflicting interests
The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: HA reports personal fees from Teva, outside of the submitted work. HS has received personal fees from AbbVie, Teva, Lundbeck, Novartis, Eli Lilly, outside of the submitted work. FMA has received personal fees from Pfizer, Teva, Lundbeck, Novartis, Eli Lilly, outside of the submitted work. MA reports receiving personal fees from AbbVie, Amgen, Eli Lilly, Lundbeck, Novartis, Pfizer and Teva Pharmaceuticals outside of the submitted work. MA received institutional grants from Lundbeck Foundation, Novo Nordisk Foundation, and Novartis. MA reports serving as associate editor of Cephalalgia, associate editor of The Journal of Headache and Pain, and associate editor of Brain.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study received funding from the Lundbeck Foundation professor grant (R310-2018-3711).
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
