Clinical Research Update: Lots of Good News for the Headache Field, and Some Disappointment, Too

Highlights include machine learning studies, anti-CGRP trials, combination treatment studies, and more.

Editor’s note: This is Part 1 of a two-part update on recent findings from clinical studies in the headache field since March 2025. See Part 2 here.

A surprising finding about stigma
Many people with migraine report being stigmatized by their healthcare providers, co-workers, and even family and friends. Previously, data from the ObserVational survey of the Epidemiology, tReatment and Care Of MigrainE (OVERCOME) study showed that people with migraine reported experiencing stigma often or very often, and that feeling stigma was associated with greater disability, reduced quality of life, and a greater interictal burden (see previous MSC coverage).

In a recent paper published in BMC Neurology, OVERCOME investigators including Robert Shapiro and colleagues reported findings from an observational, population-based survey study of nearly 12,000 participants who did not have active migraine, asking about their beliefs about people living with migraine. Roughly 70% of study participants reported at least one stigmatizing attitude, most frequently the belief that people living with migraine should be able to easily treat their condition. The investigators also found something surprising: Stigma towards people with migraine was higher among those who had closer and more frequent relationships with people who had migraine, and among those who had previously had migraine themselves.

The study also found that stigma towards people with migraine was much higher than stigma directed towards people with epilepsy, though similar to stigma directed towards individuals with chronic low back pain. For instance, 45% of participants thought people living with migraine should be able to easily treat their condition, whereas roughly 30% held the same attitude towards people who had epilepsy. Interestingly, according to an analysis comparing the sexes, both men and women held the same top three stigmatizing attitudes towards those with migraine.

The authors say their research is the first population-based study designed to query the stigmatizing attitudes of people without active migraine in the US. They concluded: “Our findings highlight the need to increase advocacy and awareness in order to reduce these stigmatizing attitudes towards people with migraine and as a result increase accommodations, support, and opportunities for people with migraine.”

Putting machine learning to the test in post-traumatic headache, and in migraine
As in many areas of medicine, those in the headache field are working to capitalize on the potential that artificial intelligence, including machine learning, has to improve prediction, diagnosis, and treatment of migraine (see MSC related podcast). One recent advance toward this aim is a Headache study from Mayo Clinic researchers, who asked if machine learning could help predict how those with acute post-traumatic headache (PTH) after mild traumatic brain injury (mTBI) would progress over time (see MSC related news).

Making use of electronic diary data, from 73 participants, on the occurrence and severity of headache, as well as on disability severity and difficulty in performing usual and physical activities, the investigators found that a machine learning model accurately classified those with PTH after mTBI into four different subgroups with different headache trajectories. The subgroups included those who had severe symptoms that did not improve over the three months of the study; those who had severe symptoms with mild improvement; individuals with milder symptoms that got substantially better; and people who had the mildest symptoms with mild improvement.

In their paper, the authors said their findings could help doctors determine how to best treat each individual with PTH – a step towards personalized medicine. The work could also help investigators figure out which patients will be most suitable to take part in clinical trials.

Another recent study, published in Brain, also used machine learning, in this case to identify genes that might contribute to migraine but that have potentially been missed in more traditional genome-wide association studies (GWAS); one estimate has found that genes implicated in migraine from GWAS only account for roughly 11% of the expected heritability of the condition – hence the search for the “missing heritability.”

The researchers used a machine learning model that could handle not only the additive effects of individual genes identified in GWAS (this is known as polygenic risk scoring), but also how genes interact with each other (referred to as non-additive effects, for which standard GWAS are not designed to uncover). The key question was whether the machine learning models could outperform polygenic risk scoring, using genome-wide genotyping data from more than 10,000 people with migraine, and roughly 33,000 headache-free controls.

That turned out to be the case: Machine learning models performed better than polygenic risk scoring in distinguishing those with migraine from those without headaches, in all four datasets used in the study. Importantly, while the machine learning models identified many of the genes that previous GWAS had found, the models also identified many new genes and pathways that had previously eluded GWAS; those pathways mainly had to do with signal transduction and neurological function. Further, the best machine learning model in the study pinpointed some familiar players, including pathways related to the calcitonin gene-related peptide (CGRP) receptor, and to botulinum toxins.

The authors pointed to a number of potential implications of their work: “First, future models with improved performance could serve as an objective measure of migraine. Second, the modeling approach is transferrable and could prove a valuable risk scoring tool for other phenotypically diverse, idiopathic neurological traits of non-additive genetic architecture. Finally, further unraveling of the model architecture could help elucidate the underlying etiology and pathophysiology of migraine, paving the way for clinical and therapeutic markers.”

Migraine progression
What is “migraine progression”? Traditionally, it’s been defined as going from episodic migraine (14 or fewer monthly headache days) to chronic migraine (at least 15 monthly headache days, with at least 8 linked to migraine). But what if progression was defined differently – would the rates and risk factors for progression be the same, compared to the usual definition?

Those questions were the focus of a study published in Headache earlier this year, with interesting results. Researchers including Dawn Buse, Richard Lipton, and colleagues looked to data from OVERCOME to investigate progression rates and risk factors using three different definitions of progression: the traditional definition; an increase of five or more monthly headache days (termed monthly headache day progression); or an increase of five or more points on the Migraine Disability Assessment (MIDAS) scale (MIDAS progression).

This study of more than 11,000 people with migraine showed that rates of progression depended on how progression was defined. Specifically, using the current definition, 4.7% of those who had episodic migraine at baseline progressed to chronic migraine, at one-year follow-up. But 9.6% of study participants progressed when progression was defined as monthly headache day progression, and 21.7% of study participants progressed according to the MIDAS progression definition.

One commonality among groups was the role of ever taking preventive medications for migraine, with all three groups showing lower odds of progressing. Another similarity: The presence of depression increased the odds of progressing in all of the groups.

The authors concluded that the new definitions “provide alternative broader ways to conceptualize the patient journey and define progression in clinical practice. The risk factors may inform clinical care if physicians use them to identify patients at high risk for progression. Our hope is to facilitate movement toward evidence-based, personalized approaches to preventing progression.” The new research could also aid clinical trial design.

More good news from studies of anti-CGRP treatments
Several recent clinical studies expand the “use cases” of migraine therapies targeting the CGRP system.

In one study appearing in Nature Medicine, researchers led by Peter Goadsby tested ubrogepant, a CGRP receptor antagonist, to alleviate prodromal symptoms of migraine (see MSC related news). This Phase 3, placebo-controlled, double-blind crossover study of more than 500 patients found that the drug was indeed successful at doing so, with significant improvements in key prodromal symptoms including photophobia, fatigue, neck pain, phonophobia, and other symptoms at various time periods from two to 24 hours post-dose.

In a different study published in JAMA Neurology, another treatment targeting the CGRP system, this time the monoclonal antibody fremanezumab, provided a two-for-one benefit (see MSC related news). Here, researchers led by Richard Lipton reported that the drug improved not only migraine, but also depression, a common comorbidity of migraine, over a 12-week double-blind study period, and during a 12-week open-label extension period. The trial included roughly 350 people with chronic or episodic migraine.

Yet another randomized trial, this time a small study of 30 participants, provided evidence that anti-CGRP treatment could be combined with neuromodulation to produce promising results (see MSC related news). Here, co-first authors Raffaele Ornello and Aurora D’Atri reported in Cephalalgia that active transcranial direct current stimulation combined with anti-CGRP antibodies resulted in fewer migraine days in those who hadn’t fully responded to the antibodies, compared to those receiving sham stimulation, and compared to baseline, at one-month of follow-up.

While the results were mixed – the study did not meet its primary outcome, which was the number of headache days – the positive results on the secondary outcome of migraine days provides hope that neuromodulation might provide a boost to those who don’t show as large a benefit as one would like from anti-CGRP treatments.

Potassium channels
Researchers led by Messoud Ashina recently built on their previous work examining the role of ATP-sensitive potassium (K_ATP) channels in migraine. In a small trial of 20 adults with migraine without aura who completed the study, the investigators asked if intravenous sumatriptan could prevent migraine attacks caused by levcromakalim, a compound that opens K_ATP channels and dilates cranial arteries.

They reported that sumatriptan did not prevent levcromakalim-induced migraine, with no significant differences in the incidence of migraine between those who received sumatriptan and those who received a placebo (saline), at 12 hours after the start of levcromakalim. While headache intensity was lower after sumatriptan compared to placebo, that particular result should be interpreted with caution considering it came from a post-hoc analysis.

One important lesson the authors draw from their findings is that, since levcromakalim still induced migraine even though sumatriptan constricted the superficial temporal artery, this supports the idea that it was increases in extracellular potassium at the neurovascular interface that played a key role here, and not vasodilation; the elevated potassium activates and sensitizes perivascular nociceptors. The authors also appear optimistic about the potential to treat migraine by targeting the K_ATP channels.

Some disappointing findings
Unfortunately, the migraine field has also recently seen some disappointing findings, too.

In a randomized trial appearing in JAMA Neurology, researchers reported that one of the triptans, in this case rizatriptan, did not differ from placebo in treating vestibular migraine, a type of migraine still without firm evidence-based treatments.

In this double-blind study of 222 participants, the drug did not show efficacy on the trial’s primary outcomes, which included reducing vertigo, and reducing unsteadiness/dizziness, at one hour after treatment. Results also disappointed on the study’s secondary outcomes, such as reductions in headache and use of rescue medications. The picture was a bit brighter when considering outcomes after 24 hours, but even in that case, rizatriptan showed a limited benefit, on just two out of seven secondary efficacy outcomes.

As the authors concluded, “Results of this study do not support routine use of rizatriptan to treat vestibular migraine attacks. It was not efficacious for any symptoms at one hour and demonstrated no effects on vertigo. Adverse effects were minor and consistent with previous reports.”

Another disappointment came from a clinical trial of M-sense Migräne, a digital app that had been developed by a German company to help people with migraine manage their condition. The app included not just an electronic healthcare diary, but also other features including a trigger diary, trigger analysis, and non-pharmacological preventive intervention capabilities.

The trial compared the full app to a limited version of the app that included only a headache and medication intake diary. Results from the 477 adults in the study were clear: Compared to the limited app version, the full app showed no advantage on the study’s primary outcome, which was the number of migraine days per month. The full app performed no better on most secondary outcomes, which included the number of headache days per month, use of acute medication, quality of life, headache management self-efficacy, and headache-attributed lost time.

The good news: Those who spoke to Migraine Science Collaborative for a news story on the study remain optimistic that digital apps could still play an important role in migraine management, particularly if they are integrated with other treatments and are appealing for patients to use.

Neil Andrews is a science journalist and executive editor of the Migraine Science Collaborative.

Image credit: ID 198082021 © Putilich | Dreamstime.com