Analysis of 2,876 adverse event reports for the CGRP receptor antagonist atogepant identified unexpected signals including abnormal dreams, self-injurious ideation, brain fog, and euphoric mood beyond the known side effects.
2,876 adverse event reportsReal-world FAERS data for atogepant revealed both expected GI/neurological effects and unexpected signals including self-injurious ideation and brain fog not prominent in clinical trials
What the researchers found
From 3,552,072 total FAERS reports, 2,876 specifically named atogepant. Women constituted the majority of reporters, concentrated in the 45–65 age group. The top adverse events by significant signal detection included expected effects: migraine (paradoxical worsening), constipation, nausea, vertigo, somnolence, decreased appetite, dizziness, and fatigue.
Notably, unexpected adverse event signals were detected for: abnormal dreams, self-injurious ideation, brain fog, tension headache, nightmares, brain neoplasm (likely coincidental), feeling abnormal, euphoric mood, hyperacusis (heightened sound sensitivity), and post-concussion syndrome. All signals were confirmed across four independent detection algorithms (ROR, PRR, BCPNN, EBGM).
Why it matters
CGRP-blocking drugs are the first migraine treatments specifically targeting the peptide pathway underlying migraine pathophysiology, and atogepant is one of the newest oral options. As millions begin using these drugs, identifying unexpected neuropsychiatric effects — especially self-injurious ideation — is critical for patient safety. The detection of mood and cognitive effects suggests CGRP peptide signaling may play broader roles in brain function than previously appreciated.
How the study worked
Researchers extracted adverse event data from the FDA Adverse Event Reporting System (FAERS), the largest spontaneous adverse event database in the United States. Four established pharmacovigilance algorithms were applied: Reporting Odds Ratio (ROR), Proportional Reporting Ratio (PRR), Bayesian Confidence Propagation Neural Network (BCPNN), and Empirical Bayes Geometric Mean (EBGM). Signals were considered significant when all four algorithms agreed. Demographic patterns and system organ class distributions were analyzed.
What this study cannot tell us
FAERS data has well-known limitations: reporting is voluntary and subject to reporting bias, reports don't prove causation (only association), and the database lacks denominator data (total number of patients taking the drug). The 'brain neoplasm' signal is almost certainly coincidental rather than causal. Some unexpected signals may reflect off-label use or pre-existing conditions. The statistical signals require clinical validation through controlled studies before being considered confirmed adverse effects.
How to read the evidence
This is a pharmacovigilance database analysis using four established signal detection algorithms. While it represents real-world safety data from a large database, FAERS analyses can only identify associations, not causation. The use of four concordant algorithms strengthens signal detection reliability, but clinical confirmation through controlled studies is needed.
When this study was published
Published in 2025, this is a timely study analyzing post-marketing safety data as CGRP-blocking drugs rapidly gain market share in migraine prevention.
The bigger picture
CGRP receptor antagonists (gepants) and CGRP-blocking antibodies represent the first disease-specific migraine therapies developed from understanding the peptide biology of migraine. This pharmacovigilance study adds real-world safety data as these drugs enter widespread use. The unexpected neuropsychiatric signals are scientifically plausible — CGRP is a neuropeptide with receptors throughout the brain, not just in pain pathways — and underscore the need for post-marketing surveillance as our understanding of CGRP's broader neurological roles evolves.
Questions still open
- Does the self-injurious ideation signal warrant a safety label update or specific screening recommendations for atogepant prescribers?
- Are the neuropsychiatric effects (dreams, brain fog, euphoria) seen across other CGRP-blocking drugs (rimegepant, ubrogepant, monoclonal antibodies)?
- What does the brain fog signal tell us about CGRP's role in cognitive function beyond pain processing?
Common questions
What is atogepant and how is it different from other migraine drugs?
Should people taking atogepant be worried about these findings?
Read the original research
A real-world pharmacovigilance study of FDA adverse event reporting system events for atogepant.
Expert opinion on drug safety, 24(6), 745-752
Citation
Wen, Heli; Ding, Yitian; Chen, Feichi. (2025). A real-world pharmacovigilance study of FDA adverse event reporting system events for atogepant.. Expert opinion on drug safety, 24(6), 745-752. https://doi.org/10.1080/14740338.2024.2377347