Thirty-four teenagers with migraine started a shot that mops up a pain-signaling molecule, and most of them felt clearly better within three months. About half of the ones who improved got there in four to six weeks.
That is the finding from a single-hospital cohort in Japan, published online July 8 in Cephalalgia ↗. It is small and it has no comparison group, but it fills a real gap: the drugs it tested are barely studied in this age band, and the outcome it tracked is the one that matters most to a 15-year-old.
The drugs are CGRP monoclonal antibodies. CGRP is short for calcitonin gene-related peptide ↗, a small protein that nerve cells release to widen blood vessels and carry pain. Blocking it, or soaking it up before it reaches its receptor, is the basis of a whole class of migraine drugs approved over the last several years. The three used here (galcanezumab, fremanezumab, and erenumab) are lab-made antibodies that do exactly that. They are given as an injection, usually monthly.
Almost all of the adult evidence for these drugs comes from adults. Regulators cleared them on trials that mostly enrolled people in their thirties and forties. Teenagers get prescribed them anyway, off the strength of that adult data, but the studies that would confirm the drugs work the same way in a still-developing brain are thin. This cohort, patients aged 15 to 17 treated between May 2021 and July 2025, is an attempt to see what actually happened in the clinic.
What the numbers say
The researchers measured migraine burden with the Headache Impact Test-6, a six-question survey that scores how much headaches interfere with work, school, and daily life. A drop of six points or more is the standard threshold for a change a patient would actually notice. The team used a survival-analysis method, the same statistical tool used to track how long until an event happens, to ask how quickly patients crossed that six-point line.
They crossed it fast. The cumulative response rate climbed from the first weeks of treatment and reached 68.3 percent within the 10-to-14-week window (95 percent confidence interval 44.6 to 81.8). Roughly half of the responders had already hit the mark by weeks four to six. Averaged across everyone and adjusted for starting scores, HIT-6 fell 9.4 points at 12 weeks (95 percent confidence interval -14.2 to -4.6), and the improvement held through follow-up.
Speed is the part worth arguing about. For an adult, a migraine drug that takes two months to work is an inconvenience. For a teenager missing school, two months is a chunk of a semester. Of the 27 patients who answered a functioning questionnaire, the activity most wrecked by migraine before treatment was school attendance or concentration in class (70.4 percent named it). After treatment, 88.9 percent said their main goals were mostly or partly met. That is the real endpoint hiding behind the HIT-6 math.
The caveats are large
This is 34 patients at one center, looked at after the fact, with no placebo arm and no randomization. Migraine in adolescents improves on its own more often than it does in adults, and a retrospective single-arm study cannot separate the drug from that natural drift, or from the placebo effect, which runs high in headache trials. The response numbers are almost certainly flattered by the absence of a control group.
Tolerability is the sturdier signal. Adverse events showed up in 40.7 percent of patients, mostly injection-site reactions (29.6 percent), and none were serious enough to stop or change treatment. For a drug given to minors, a clean side-effect profile in real use is worth as much as the efficacy curve.
The finding also sits oddly against the rest of the CGRP story. These same antibodies barely outperformed placebo in cluster headache ↗, a different headache disorder driven by the same peptide. The mechanism is shared; the clinical payoff is not. A molecule that is a clean target in one condition can be a weak one next door, and CGRP keeps demonstrating both.
The authors are direct about what this is: a real-world signal, not proof. They call for controlled prospective studies and for work to find which teenagers respond early. Until that arrives, this cohort is the closest thing to evidence that the shots adults have relied on for years do something useful, and something fast, in the age group that can least afford to wait.