Abstract
Background/Hypothesis
Experimental provocation studies have yielded important insights in migraine pathophysiology. Levcromakalim has been previously shown to induce migraine-like attacks with and without aura. In this study, we aim to further explore the migraine aura-inducing potential of levcromakalim.
Methods
In a double-blind, randomized, placebo-controlled cross-over study, 27 adult participants with migraine with aura received intravenous infusions of levcromakalim and saline. Headache, aura and associated symptoms were evaluated for 24 hours following administration of the study drug. The primary endpoint was occurrence of migraine-like attacks with or without aura in the 24-hour observation period.
Results
Thirteen participants developed migraine-like attacks on the active day only (P = 0.0098), and four participants developed aura on the active day only (P = 0.68). The median time to onset of migraine-like headache was three hours, and the median time to onset of aura was 27.5 minutes.
Conclusion/Interpretation
Our findings affirm the potent migraine-inducing effect of levcromakalim. We observed a lower induction-rate of migraine aura than previously reported. Further studies are warranted to identify predictors of migraine aura following levcromakalim.
ClinicalTrials.gov identifier
NCT04905654
Introduction
In recent years, substantial advancements have been made in understanding the pathophysiology of migraine aura and the associated head pain. However, the exact relationship between these two entities remains ambiguous. The prevailing hypothesis suggests that migraine aura is explained by cortical spreading depression (CSD), a slowly spreading wave of intense neuronal and glial depolarization that leads to a temporary depression of electrical activity (1). This mechanism is closely linked with the activation of the trigeminovascular system (TVS), which involves arterial dilation and the release of vasoactive peptides, and is central to the genesis of migraine pain (2). Yet, the precise role of CSD in initiating this process during headaches associated with aura is still under investigation, with the potential involvement of either sequential or parallel processes in the manifestation of both aura and head pain (3).
The episodic nature of migraine necessitates detailed investigations of both the ictal and interictal phases. Our current understanding of migraine pathophysiology is derived from studies on both spontaneous, i.e., naturally occurring migraine attacks, and induced migraine attacks. The latter, utilizing pharmacological agents with well-characterized biochemical properties, has substantially advanced our understanding of migraine's molecular underpinnings (2,4). Furthermore, human migraine models have been pivotal in the development of migraine-specific drugs, including innovative therapies targeting the calcitonin gene-related peptide (CGRP) and its receptor.
Given the transient nature of migraine aura episodes, which typically last 20 to 30 minutes (5,6), developing reliable methods to induce these auras is of critical importance. Initial attempts using various pharmacological substances to specifically induce migraines with aura have been somewhat unsatisfactory (7). However, recent studies suggest that levcromakalim, an agent that activates adenosine triphosphate-sensitive potassium (KATP) channels, demonstrates potential in triggering migraine attacks both with and without aura (8,9). The present study aims to further investigate the potential of levcromakalim in inducing migraine aura and headache in individuals with migraine with aura.
Methods
Design
The study was designed as a double-blinded, randomized, placebo-controlled cross-over study. Each participant attended a total of three visits, comprising a preliminary screening session and two experimental days.
Participants
We recruited participants aged 18–70 years and weighing 50–100 kilograms with a diagnosis of migraine with aura according to the International Classification of Headache Disorders 3rd edition (10) (ICHD-3) and who reported at least six aura episodes in the preceding 12 months. For inclusion, individuals were required to have a history of headache, defined as any pressing or painful sensation in the head accompanying aura, in more than 20% of aura episodes. Exclusion criteria included a history of tension-type headache for five or more days per month in the past 12 months, cluster headache, pregnancy or breastfeeding, blood pressure <90/50 mmHg or >150/100, substance abuse, or smoking. Any physical or psychiatric illness or medication use that could potentially influence the safety of the study led to exclusion. Female participants of child-bearing age were subjected to a urine pregnancy test on each study day. Recruitment methods included recruitment of patients from the Danish Headache Center outpatient clinic, social media advertisements, referrals from neurologists in private practice, and via invitations to participants in the Danish Blood Donor Study (11). All participants gave written informed consent before inclusion.
The Ethics Committee of the Capital Region, Denmark, approved the study (H-20049785). The study conformed to the Declaration of Helsinki of 1964 with later revisions. It was registered at ClinicalTrials.gov (NCT04905654).
Procedures
An initial screening visit involved a detailed medical and neurological history was taken, electrocardiogram (Cardiofax V, Nihon-Koden) and a medical examination were also performed. Participants were randomly assigned to receive an intravenous infusion of 0.05 mg/min (20 mL of 50 µg/mL total dose 1 mg) levcromakalim (manufactured by Tocris, Bio-Techne Ltd., UK under regulations required by good manufacturing practice) or a placebo (20 ml of 0.9 mg/ml saline) over 20 minutes using an electronic infusion pump (B.Braun Infusomat). The Capital Region Central Pharmacy managed the preparation and randomization of the study drug. The randomization code was inaccessible to investigators until study completion.
Participants arrived non-fasting at the Danish Headache Center between 8.00 a.m. and 2.00 p.m. Participants were required to abstain from caffeine at least 12 hours before the experiment and to be headache- and aura-free at least 48 hours before the experiment. Participants were placed in a bed with the headrest at a 45-degree angle, and an intravenous catheter was inserted in the right or left antecubital vein. Following a 20-minute resting period, we took baseline measurements (at time = −10 minutes) of blood pressure and pulse. The infusion process was initiated at time = 0. We monitored and documented pulse and blood pressure at 10-minute intervals for the next 120 minutes after the start of the infusion (Philips IntelliVue MP5). At each interval, we recorded migraine characteristics including the intensity and features of the headache, presence of nausea/vomiting, photo-/phonophobia, and aura. We also recorded whether the headache resembled a typical migraine attack for the participant. Additionally, flushing, heat sensation, palpitations and any other unusual symptoms were recorded. In the event of aura, we collected detailed data about its duration, character, location and spread. We also asked participants to draw visual and/or sensory symptoms every five minutes. We discharged participants 120 minutes post the initiation of infusion. Subsequently, for the next 22 hours (for a total observation period of 24 hours from the start of the infusion), participants were asked to record headache and aura features and any presence of unusual fatigue, neck stiffness, yawning, mood swings, concentration difficulties, hunger, thirst, redness, heat sensation and palpitations. Participants had free access to acute medication, the usage of which was also recorded. We requested that participants adhere to their regular sleep schedule and document symptoms only when awake. To ensure the wash-out of the study drug, we maintained at least a seven-day interval between the two experimental days.
Outcomes and measures
Our primary endpoints were the differences in the incidence of migraine-like attacks with and without aura, migraine attacks with aura, any headache within the 24-hour period following the start of the infusion. Key secondary endpoints were differences in the area under the curve (AUC) for headache intensity scores (0–24 hours), for heart rate (HR) (0–120 min) and mean arterial pressure (MAP) (0–120 minutes) between levcromakalim and placebo days. Additional secondary endpoints included incidence of adverse events (0–24 hours) and acute medication use. Migraine aura was considered present if symptoms fulfilled ICHD-3 criteria B and C for 1.2 migraine with aura. Migraine headache was considered present if ICHD-3 criteria C and D for 1.1 migraine without aura were fulfilled (moderate/severe headache defined as a pain intensity of 4 or above on a numeric rating scale (NRS) from 0–10), or if participants reported that headache mimicked usual migraine and used acute medication. Presence of any migraine was defined as either set of criteria being fulfilled.
Data and statistical analysis
The power calculation for our study was based on McNemar's test. Using the estimates from a previous study (8), we projected that 50% of participants would experience an aura following levcromakalim infusion, whereas 10% would have the same outcome after placebo. Power was set at 80% with a significance level of 5%. Consequently, our estimations indicated that a total of 27 participants were needed. Therefore, participant recruitment was sustained until 27 individuals had finished both experimental procedures.
Continuous data are presented as mean values ± standard deviations (SDs) or median (range). Other data are reported as categorical data. The incidence of migraine aura, all types of migraine, and any type of headache, were analyzed as binary categorical data using McNemar’s test. The Area Under the Curve (AUC) was calculated according to the trapezoidal rule (12).
We performed paired two-way t-tests for the AUC values related to heart rate and MAP, while AUC values for headache scores were compared using Wilcoxon signed rank tests. To test for period and carry-over effects, we employed unpaired t-tests and the Mann-Whitney U-test. All these analyses were conducted using GraphPad Prism v. 9.5.1, and the power calculation was conducted in R v. 4.2.0 using the MESS-package. A P-value of <0.05 was considered significant. As our endpoints were clearly delineated in the study protocol, we did not adjust for multiple comparisons.
Data availability
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.
Results
Participants
The screening visit was attended by 33 participants, and 27 participants completed both experimental visits (Figure 1). Seventeen (63.0%) were females, the mean (SD) age was 32.8 (12.1) years, and the mean (SD) body-mass index (BMI) was 24.0 (3.3) kg/m2 (Table 1). The median (range) number of aura episodes in the last 12 months was 10 (6–120). Ten (37.0%) participants reported a history of migraine without aura. All 27 participants reported visual aura. Sensory aura was reported by 12 (44.4%), speech/language aura by 10 (37.0%), and one (3.7%) participant reported hemiplegic aura.
Flowchart of participants.
Baseline characteristics of participants.
BMI: Body-mass index, MA: Migraine with aura, MO: Migraine without aura.
Incidence of migraine attacks with and without aura, and any headache
Fourteen participants (51.9%) developed migraine attacks with and without aura following levcromakalim compared with three participants (11.1%) following placebo (P = 0.0098; Table 2). One participant reported migraine-like attacks on both experimental days. The median time to the onset of attacks was three hours. In an exploratory analysis 10 participants reported migraine-like attacks without aura following levcromakalim compared with one after placebo (P = 0.0077), with only four having a history of migraine without aura. Headaches of any type were reported by 23 after levcromakalim compared with 18 after placebo (P = 0.18). Acute medication was used by 16 participants on the active day after a median of five hours and by four on the placebo day after a median of three hours.
Clinical characteristics of headache and associated symptoms of reported migraine attacks (0–24 h observation period).
*Onset of migraine was defined as the onset of aura or the earliest time point when headache symptoms fulfilled migraine criteria.
**Localization/intensity/quality (throb = throbbing; pres = pressing; cut = cutting)/aggravation (by cough during in-hospital phase and by movement during out-hospital phase).
***Nausea/photophobia/phonophobia. Abbreviations: F: female; M: Male.
Incidence of migraine attacks with aura
Four participants experienced migraine attacks with aura following levcromakalim compared to two after placebo (P = 0.68; Table 3). The median time to the onset of aura was 27.5 minutes. Four participants reported visual aura (two after levcromakalim and two after placebo), one sensory aura (after levcromakalim), and one both visual and sensory aura (after levcromakalim). The median frequency of aura did not differ between those who did and did not develop aura (median = 10 episodes/year for both groups). Headache accompanying spontaneous aura met ICHD-3 criteria C and D for 1.1 migraine without aura in only two cases. In one case, aura preceded headache (after levcromakalim), in another, aura and headache presented simultaneously (after placebo), and in the remaining four participants headache preceded aura by 3–273 minutes (three after levcromakalim, one after placebo). Four auras were unilateral, and two of these were accompanied by ipsilateral headache, whereas the other two were accompanied by bilateral headache. Two auras were bilateral and were both accompanied by unilateral headache. Two participants who developed aura (one following levcromakalim and one following placebo) had a history of migraine without aura.
Clinical characteristics of reported aura episodes (0–24 h observation period).
*On placebo day.**Participant reported sensory aura during the hospital phase and then reported separate visual and sensory aura post-hospital phase.
Headache intensity and hemodynamic variables
The AUC for headache intensity was significantly larger after levcromakalim compared to placebo (0– 24 hours, P < 0.0001, Figure 2(a) and (b)). There was a significant increase in heart rate after levcromakalim compared to placebo (heart rate AUC 0-120 minutes, P < 0.0001, Figure 2(c)). MAP did not differ significantly between the two days (MAP AUC 0-120 minutes, P = 0.068). No period or carry-over effects were detected for any measure.
(a and b) Headache registrations from the levcromakalim day and the placebo day, respectively and (c) Hemodynamic data. Abbreviations: Bpm: Beats per minute, HR: Heart rate, MAP: Mean arterial pressure, NRS: Numeric rating scale.
Adverse events
The most commonly reported adverse events within the 24-hour observation period included tiredness (10 instances following levcromakalim, four following placebo), neck stiffness (seven following levcromakalim and one following placebo), heat sensation (seven following levcromakalim and two following placebo), yawning (seven following levcromakalim and 0 following placebo) and concentration difficulties (five following levcromakalim and 0 following placebo). The full list of adverse events is provided in the Online Supplemental Table S1.
Discussion
The major outcome of the present study was that approximately 50% of participants developed migraine-like attacks, with or without aura, on the active day, while 15% experienced aura. This observation confirms the migraine-inducing effect of levcromakalim, albeit at a lower induction rate compared to previous studies (8,9). The divergence in the induction rate of migraine with aura, which was non-significant compared to a previously reported 59% (8), raises questions about variability in response or potential methodological differences. Despite our methodology's consistency with past research, the results contrast with earlier findings, suggesting possible limitations in statistical power or an overestimation of effect size in our initial studies. A notable aspect of our study is the unexpectedly high nocebo response rate for migraine aura (7.4%), contrasting with earlier studies where no aura symptoms were reported post-placebo (8,13,14). Interestingly, one of the two participants reporting aura following placebo reported that the aura did not mimic spontaneous attacks.
Migraine induction
Among the participants who developed migraine-like attacks without aura, it is noteworthy that only four had a prior history of such conditions, suggesting a predisposition in the others. This finding could also highlight the limitations of retrospective reporting in diagnosing migraines. The underlying mechanisms by which levcromakalim induces migraine with aura are still unknown. The pathophysiology of migraine involves the interplay of the trigeminovascular system and central structures, with particular focus on the role of CSD and its relationship to migraine headache (3). The hypothesis that levcromakalim acts either on structures within or outside of the central nervous system (CNS) leads us to explore both central and peripheral mechanisms.
Central mechanisms
The blood-brain-barrier (BBB) penetrance of levcromakalim, which could theoretically allow it to affect central structures, remains unconfirmed in humans. It is thus unknown whether levcromakalim reaches central structures at all. Considering the hyperpolarizing effect of potassium channel opening, a direct induction of CSD by levcromakalim on central neurons appears counterintuitive. The potential role of KATP-channels in glial cells leading to increased extracellular potassium and triggering CSD lacks substantial evidence. An alternative hypothesis is the selective inactivation of inhibitory interneurons by levcromakalim, raising the likelihood of CSD. However, rodent studies have provided mixed results (15,16). One must consider, however, the inherent limitations in translating findings from rodent models to humans. If the effect of levcromakalim is indeed central, it remains to be demonstrated that CSD leads to headache. If CSD were the precipitating cause of headache, we would propose that either 1) CSD would activate ipsilateral central structures in the brain involved in the perception of pain and thus cause headache ipsilateral to the aura; or 2) CSD would activate extra-axial structures such as meningeal nociceptors ipsilateral to the CSD and thus cause headache contralateral to the aura. The clinical correlation between CSD and headache manifestation, particularly the varied presentation of headache in relation to aura (17,18), remains an area requiring further investigation.
Peripheral mechanisms
The vasodilatory effect of levcromakalim, particularly on intracranial arteries, may activate perivascular trigeminal nociceptive fibers, central to migraine pathophysiology (8,9,19,20). Substances not presumed to cross the BBB, such as CGRP and pituitary adenylate cyclase activating peptide, have proven potent in inducing migraine attacks without aura (4). Importantly, a recent study from our group demonstrated that 13 of 34 participants developed migraine attacks with aura following infusion of CGRP, corroborating previous findings (21,22). Based on this, our group has proposed that mechanical activation of first-order trigeminal neurons consequent to vasodilation provides an excitatory stimulus reaching cerebral cortex and reducing the threshold of CSD initiation. As the net-result of levcromakalim on smooth muscle cells is the same as that of CGRP, it is possible that levcromakalim may cause migraine aura via this mechanism. It is conceivable that potassium efflux from smooth muscle cells diffuses to and depolarizes trigeminal nerve endings. However, this hypothesis, although plausible, lacks direct evidence (2). The frequent occurrence of headache prior to aura in participants suggests a primary activation of the trigeminovascular system, potentially leading to CSD.
Other perspectives
Comparative studies examining other migraine-inducing agents (such as glyceryl trinitrate and sildenafil) and conditions (like normobaric hypoxia) have provided important insights into migraine pathophysiology (13,23,24). Notably, two key observations emerge from these studies on experimentally induced migraine auras: 1) The temporal relation between migraine aura symptoms and headache in provoked migraine-like attacks with aura is identical to the participants’ usual attacks. This similarity suggests that both aura and headache are provoked through a shared mechanism, similar to what occurs in spontaneous migraine with aura attacks. This mechanism likely operates at a common site, rather than involving distinct actions at separate locations such as one on smooth muscle cells or peripheral nerves and another in the CNS. 2) Provocation experiments typically result in a phenotype of migraine-like attacks without aura in the majority of participants. This is particularly interesting as it occurs even among those who report aura symptoms in all their previous migraine attacks. Conversely, only a few participants report migraine aura symptoms following provocation. This observation implies that while experimentally induced vasodilation can efficiently trigger migraine attacks without aura, it is less effective in inducing migraine aura. Consequently, this suggests that the initiation of spontaneous attacks of migraine with aura involves a distinct mechanism that more potently triggers CSD. These findings collectively indicate that the spontaneous onset of migraine attacks with aura likely involves an intrinsic mechanism leading to both CSD and activation of the trigeminovascular system. In contrast, experimental migraine provocation methods appear to predominantly affect the trigeminovascular system. The causal relationship between aura and headache in spontaneous migraine attacks remains complex. It may involve an initial event of CSD leading to the subsequent activation of the trigeminovascular system. Alternatively, based on these aura provocation studies, a primary activation of the trigeminovascular system could be the initial event, subsequently leading to CSD initiation in susceptible individuals.
Further research is needed to elucidate this intricate causal relationship between aura and headache in migraine. In this context, patients who report migraine aura without headache are of particular interest. Pathophysiologically, these patients are expected to experience CSD without significant activation of the trigeminovascular system. If substances like levcromakalim have distinct actions, one triggering CSD within the CNS and another inducing migraine headache outside the CNS, they might be expected to induce a phenotype of migraine aura without headache in this unique group of patients. Conversely, if CSD induction by levcromakalim is contingent upon activation of the trigeminovascular system, these individuals might exhibit a clinical response to levcromakalim infusion similar to healthy volunteers, without the triggering of migraine aura or headache.
Based on the current study and previous human studies, opening of KATP channels induced migraine attack in 51–100% of individuals with migraine with and without aura. Furthermore, the induction of migraine attacks via opening of KATP channels appears independent of CGRP-signaling (25). Thus, KATP-channels are a key target for development of medical treatments for migraine.
Conclusions
Our findings, while aligning with the known effects of levcromakalim, also highlight discrepancies with prior research. This underscores the need for further studies to clarify these differences and to understand the individual variability in response to levcromakalim. Exploring these variations can offer deeper insights into the mechanisms of migraine induction and aura manifestation.
Clinical implications
Our study affirms the potent migraine-inducing effect of levcromakalim. Levcromakalim likely induces migraine aura via effects on the trigeminovascular system. Further studies are needed to identify predictive factors in developing migraine aura following levcromakalim infusion.
Supplemental Material
sj-pdf-1-cep-10.1177_03331024241237247 - Supplemental material for Investigations of the migraine-provoking effect of levcromakalim in patients with migraine with aura
Supplemental material, sj-pdf-1-cep-10.1177_03331024241237247 for Investigations of the migraine-provoking effect of levcromakalim in patients with migraine with aura by Andreas Vinther Thomsen, Mohammad Al-Mahdi Al-Karagholi, Anders Hougaard, Sisse Rye Ostrowski, Ole Birger Pedersen Thomas Folkmann Hansen and Messoud Ashina: on behalf of the Chinese Cluster Headache Alliance (CCHA) in Cephalalgia
Footnotes
Acknowledgements
The authors wish to thank all the study participants and the research staff at the Danish Headache Center. The authors give thanks to doctors Jakob Møller Hansen, Lili Kokoti, Peter Thede Schmidt-Hansen, Janu Thuraiaiyah and Zixuan Alice Zhuang.
Author contributions
M.M.A-K., A.H. and M.A. conceived and initiated the study. M.M.A-K, A.V.T., T.F.H, O.B.P and S.R.O. participated in recruitment. A.V.T. and M.M.A-K. participated in data acquisition. A.V.T. analyzed the data and drafted the first version of the manuscript. All authors revised the manuscript for intellectual content, approved the final version of the manuscript and agree to be accountable for all aspects of the work.
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: A.V.T., M.M.A-K., O.B.P, S.R.O. and T.F.H. declare no competing interests.
A.H. reports receiving personal fees from AbbVie, Eli Lilly, Lundbeck, Novartis, Teva and Pfizer. A.H. also serves as an associate editor of Headache.
M.A. reports receiving personal fees from AbbVie, Amgen, Astra Zeneca, Eli Lilly, GlaxoSmithKline, Lundbeck, Novartis, Pfizer, and Teva Pharmaceuticals outside of the submitted work. M.A. also serves as an associate editor of The Journal of Headache and Pain, and an 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 was funded by a research grant from the Lundbeck Foundation (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.
