Thienopyridines May Help ID Migraine Patients Best Suited to PFO Closure

Two studies demonstrate hypothesis-generating evidence that would support an upcoming randomized trial, experts say.

Thienopyridines May Help ID Migraine Patients Best Suited to PFO Closure

The hypothesis that migraine headaches may be linked to patent foramen ovale (PFO)—meaning PFO closure could be considered as a treatment—is getting renewed support from two new studies that show promise in identifying the subgroup of patients with migraine who might clearly benefit.

This is not the first time this procedure has been studied in migraineurs, but none of the previous trials have confirmed a link between PFO closure and improved outcomes. But according to Robert Sommer, MD (NewYork-Presbyterian Hospital/Columbia University Irving Medical Center), who served as lead and senior author on the two studies published in Neurology last month, the population of patients enrolled in those studies was too broad.

“The big issue has been that we know that some people respond really well and some people don't respond at all,” he told TCTMD. “And the question was: how do you pick the right patients [for closure]?” By screening patients with PFO and migraine to see if their headaches go away following thienopyridine treatment, Sommers said they can narrow down the overall population to choose patients who will benefit from closure.

“What this really allows us to do is to pick out the PFOs that are causing the headaches from the PFOs that are simply incidental,” he explained. “That's why the other studies failed—because they really couldn't make that selection. So that's what we hope, that the extra screening step with the thienopyridines is going to enrich the population for [a future] trial so that it will be successful.”

Why not just keep migraine patients on thienopyridines long-term? Sommer said it’s an option, though most migraine patients are younger women “who get horrible periods” on blood thinners and also can’t get pregnant on them. Also, in general patients would have to stop the drugs if they ever were to get surgery and their headaches would temporarily come back. Ultimately the point of future trials will be first to prove that thienopyridines reduce headaches, he said, but that’s not a long-term solution if patients can simply have their PFOs closed once and for all.

Investigating P2Y12 Inhibitors

For the first study, Sommer and colleagues retrospectively looked at 136 patients who were prescribed clopidogrel to prevent migraines between 2011 and 2017. Fifty-nine percent of the population were deemed responders, meaning that they had a reduction in at least 50% of monthly migraine days. Of the nonresponders, 40% were deemed to have inadequate platelet inhibition based on VerifyNow (Accriva Diagnostics) platelet reactivity unit levels, and 62% of them were eventually responders to prasugrel.

In all, 62% of the thienopyridine responders—both clopidogrel and prasugrel—underwent PFO closure with drug discontinuation after 3 months, and 94% had ongoing migraine relief. All eight of the responders who stopped thienopyridine treatment without closure reported resumed migraine symptoms.

“If you close the PFO in the patients who respond to the drugs, the relationship is almost completely synonymous,” Sommer summarized. “If you respond to the drugs, you basically respond to the closure.”

For the second study, known as TRACTOR, researchers led by Adam Reisman, BS (NewYork-Presbyterian Hospital/Columbia University Irving Medical Center), took things a step further to see if a similar cohort of patients would respond to ticagrelor, a nonthienopyridine P2Y12 inhibitor. They included 40 patients with PFO, documented right-to-left shunt, and at least 6 monthly migraine days, giving them ticagrelor for 28 days.

Not quite half of the population (43%) responded to the treatment, but migraine symptom reduction continued through 3 months for those that did respond. One-third of patients had medication side effects like shortness of breath and presyncopal events, but there were no serious adverse events.

“One of the main reasons for undertaking the TRACTOR Pilot Trial was to help select a P2Y12 agent for use in a blinded, randomized trial,” Weisman and colleagues write. “With respect to planning such a trial, one of the most significant differences between treatment with ticagrelor and with the thienopyridines was the relative rate of medication side effects.” Because of this, they add, ticagrelor is “potentially an inferior choice for a national trial.”

Randomized Trial Ahead

In an editorial accompanying the studies, Jérôme Mawet, MD (Lariboisiere Hospital, Paris, France), and Matthias Karst, MD, PhD (Hannover Medical School, Germany), write: “These studies open a new diagnostic and therapeutic window for a subgroup of migraineurs with PFO in which altered platelet activity may play a major role in their migraine activity. Low-dose acetylsalicylic acid did not show the same effects, suggesting that attenuation of the proinflammatory capacity of activated platelets via the inhibition of P2Y12 receptors mediates the observed effects on migraine.”

But migraine is a heterogeneous condition both in pathophysiology and manifestation, stressed Rebecca Burch, MD (Brigham and Women’s Hospital, Boston, MA), a neurologist who commented on the study for TCTMD. “I think what the authors here are attempting to do is to identify a subgroup of patients with migraine with aura in whom the PFO is part of the underlying pathophysiology of their migraine, and I think it's probably going to be more complicated than that,” she observed.

The limitations of both studies lead them to only be “hypothesis-generating” at this time, Burch said. “The nature of the study design and the methodological flaws in these studies mean that we cannot draw any conclusions from them. . . . I think it provides support for the possibility of going forward with a larger placebo-controlled, well-blinded, randomized controlled trial of these treatments. But until we see results from a high-quality study, I don't think this should change clinical practice or change how we refer patients for PFO closure.”

Sommer told TCTMD that his group recently received funding from WL Gore—the manufacturer of the Cardioform septal occluder that was approved by the FDA for prevention of recurrent ischemic stroke in April—to do just that. The RELIEF Migraine study, he explained, will first compare the results of thienopyridines against placebo to show efficacy in reducing migraines, and then the estimated 150 thienopyridine responders will be randomized to PFO closure or a sham procedure followed by a stopping of the thienopyridine.

“We expect there to be a very, very large clinical difference between the patients who get their PFOs closed and the sham patients,” Sommer said.

“Among headache specialists, this is an ongoing area of kind of intellectual curiosity,” Burch said. “I think that there are a few people who are still very interested in the idea of PFO closure and whether we could offer a sort of single procedure that would have lifelong preventive effects for our patients. I think there are some people who are involved in the trials of PFO closure for migraine who feel that there is a subset of patients who have migraine with aura who probably would benefit if we could identify them, so with that perspective I think there will be interest in the headache community in the results of this trial.”

However, “the results would need to be conclusively positive and the safety of both the thienopyridine and PFO closure would need to be clearly demonstrated in order for us to consider this as an appropriate preventive strategy going forward,” she added. “We very much want to reduce the disability associated with migraine and the pain associated with migraine, but nobody will want to do that at the potential risk of intracranial hemorrhage or cardiac problems from the procedure.”

The editorialists say the idea of targeted therapy is appealing and express optimism about what’s ahead in terms of research. “Let us hope that this new approach of targeting a subset of migraineurs will be shared, discussed, duplicated, and challenged, and that it will lead to RCTs,” Mawet and Karst write. “These upcoming studies may not only help to determine the efficaciousness of PFO closure in this subset of migraineurs but also delineate the large heterogeneous spectrum of migraines.”


  • Reisman and Burch report no relevant conflicts of interest.
  • Sommer reports receiving investigator-initiated grant funding by AstraZeneca and using ticagrelor off-label through and IDE with the FDA for this trial; developing a National Migraine/PFO protocol based on this trial with WL Gore and Associates; serving as PI on the REDUCE, ESCAPE, and ASSURED ASD Closure trials; and serving as a physician trainer for Abbott Medical and Boston Scientific.
  • Mawet reports receiving travel, accommodation, and meeting expenses from SOS Oxygen, Air Liquide, Amgen, and Homeperf; served as a consultant from Air Liquide; and serving on the redaction committee of La Lettre du Neurologue.
  • Karst reports receiving travel expenses and speaker honoraria from Gruentrhal GmbH and Bionorica Ethics GmbH.