A Migraine Drug Cut Headache Days From 20 to 9 in a Group That Had Already Tried Almost Everything

Imagine having a headache on 20 days of every month, then being told that many of the standard treatments have already failed. That was close to the starting point for hundreds of patients in a new Spanish real-world study of atogepant. Among those who remained on treatment for a full year with complete follow-up, median headache frequency fell from 20 days per month to 9. Migraine days fell from 13 to 5.

What makes that especially striking is who these patients were: they had typically already failed six preventive treatments, most had chronic migraine, 84% had failed Botox and more than half had previously tried anti-CGRP antibodies.

These were unusually difficult migraine cases

The prospective multicentre GEMA Project enrolled 513 patients from 15 specialist headache units in Spain.

This was not a typical first-line migraine population. The median patient had already failed six preventive therapies. Seventy-four percent had chronic migraine. Eighty-four percent had previously failed onabotulinumtoxinA, better known as Botox, and 270 patients had already been exposed to anti-CGRP monoclonal antibodies.

In other words, many had reached the point where another preventive treatment might reasonably be expected to produce diminishing returns.

Atogepant targets the CGRP pathway too, but unlike monoclonal antibodies it is a small-molecule oral CGRP receptor antagonist, or gepant, taken daily.

Randomized trials have already shown that it can prevent episodic and chronic migraine. The new question was harder: how well does it hold up for a full year in patients who have already been through multiple treatment failures?

Headache days fell from 20 to 9

Among patients who stayed on atogepant and had complete data at every follow-up, median monthly headache days dropped from 20 at baseline to 10 after three months.

By six months they had fallen to 9, and they remained at 9 after one year.

Migraine days showed a similarly large shift, falling from a median of 13 per month to 6 at three months and 5 at both six and 12 months.

That translates into roughly 11 fewer headache days and 8 fewer migraine days each month for the continuing-treatment group.

For someone starting with headaches on two-thirds of the month, that is not a subtle statistical improvement. It potentially means reclaiming more than a week of each month from migraine.

At 12 months, 52.3% of patients in the observed-case analysis had reduced their headache days by at least half. For migraine days, the figure was 57.8%.

Improvements were also reported in disability and medication overuse.

The difficult part of the result is who stayed on treatment

There is an important second half to the story.

About 80% of the original group were still taking atogepant at six months, but persistence fell to around 60% at one year. Lack of effectiveness was the main reason people stopped.

That means the spectacular 20-to-9 result should not be interpreted as what happened to every person who started the drug.

When the researchers used a deliberately conservative analysis that counted missing patients as non-responders, about 31% of the entire original cohort achieved at least a 50% reduction in headache days and about 33% achieved that reduction in migraine days.

Those numbers are less dramatic, but the context matters. This was an unusually refractory population in which repeated treatment failure was the norm.

The study also found exactly what clinicians might expect: the more preventive treatments a person had already failed, the lower their probability of reaching a 50% response after one year.

Why the result still matters

Atogepant was not being tested here as a miracle treatment for newly diagnosed migraine. It was being pushed into some of the hardest cases seen in specialist headache clinics.

And among patients for whom it worked well enough to continue, the benefit arrived early and remained strikingly stable through 12 months.

That durability is important. A treatment that produces a good three-month response but fades by month nine is very different from one that maintains the reduction for a year.

Safety was also broadly reassuring. Most reported adverse events became less frequent at later follow-up visits, and researchers identified no new long-term safety signal.

Previous randomized trials already established that atogepant can prevent migraine, including chronic migraine and migraine after prior preventive failures. What this study adds is a harsher real-world stress test.

For some people who had already tried Botox, CGRP antibodies and multiple other preventives, another CGRP-targeting treatment still produced a major improvement.

That is the most hopeful part of the result: failing several migraine therapies does not necessarily mean the remaining options will only produce marginal gains.

Scientific References & Further Reading

• Anchor Study: “Long-term atogepant for treatment-resistant migraine in real-world clinical practice: 12-month results (GEMA project).” (Gago-Veiga AB, López-Rodríguez AB, Hoyas Cruz B, et al. J Headache Pain. 2026. doi:10.1186/s10194-026-02509-6.)
• Phase 3 Evidence: “Atogepant for the Preventive Treatment of Migraine.” (Ailani J, Lipton RB, Goadsby PJ, et al. N Engl J Med. 2021;385:695-706. doi:10.1056/NEJMoa2035908.)
• Chronic Migraine Trial: “Atogepant for the preventive treatment of chronic migraine (PROGRESS).” (Pozo-Rosich P, Ailani J, Ashina M, et al. Lancet. 2023;402:775-785. doi:10.1016/S0140-6736(23)01049-8.)
• Prior-Failure Trial: “Safety and efficacy of atogepant for episodic migraine in adults for whom conventional oral preventive treatments have failed (ELEVATE).” (Tassorelli C, Nagy K, Pozo-Rosich P, et al. Lancet Neurol. 2024;23:382-392. doi:10.1016/S1474-4422(24)00025-5.)

---

Disclaimer: This article is for informational purposes only and does not constitute medical advice. It is not a substitute for diagnosis, treatment, or individualized medical care.