Comparative post-marketing reporting signals of elagolix and myfembree in endometriosis: a FAERS pharmacovigilance study

Background Elagolix and Myfembree are gonadotropin-releasing hormone (GnRH)-pathway therapies used for endometriosis, but their post-marketing safety reporting patterns remain incompletely characterized. Because spontaneous reporting databases are susceptible to reporting bias and differential market exposure, comparative analyses require cautious interpretation. Methods We analyzed quarterly data from the FDA Adverse Event Reporting System (FAERS) from 2015Q3 to 2026Q1. Deduplicated female reports with endometriosis-related indications were identified and classified as elagolix, Myfembree, or other endometriosis-related reports. The primary analysis consisted of drug-specific case-noncase disproportionality analyses for elagolix and Myfembree separately within the female endometriosis reporting background. A direct elagolix-versus-Myfembree head-to-head analysis was performed as a secondary analysis. Reporting odds ratios (RORs), proportional reporting ratios (PRRs), and information components (ICs) were calculated at the preferred term (PT) level. Sensitivity analyses included serious-report-only, healthcare-professional-only, physician-only, complete-age, reporter-type-stratified, overlapping-market-period, and bootstrap analyses. Results A total of 4,428 deduplicated female endometriosis-related reports were included, comprising 1,744 elagolix reports, 280 Myfembree reports, and 2,404 other endometriosis-related reports. Serious reports accounted for 31.2% of elagolix reports and 26.8% of Myfembree reports. In drug-specific case-noncase analyses, elagolix showed robust disproportionality signals for hot flush, night sweats, and suicidal ideation. Myfembree showed distinct reporting signals for reproductive and bleeding-related PTs, including heavy menstrual bleeding and intermenstrual bleeding. In the secondary head-to-head analysis, selected PTs including hot flush, nausea, headache, depression, arthralgia, and suicidal ideation showed higher reporting signals for elagolix, whereas alopecia showed a lower reporting signal for elagolix. Sensitivity analyses using alternative algorithms, reporter-type restrictions, overlapping-market-period restriction, complete-age restriction, and bootstrap validation generally supported the direction of the main selected reporting patterns, although some estimates were limited by small cell counts. Conclusions Elagolix and Myfembree showed distinct post-marketing reporting signal profiles among female endometriosis-related FAERS reports. Elagolix was characterized mainly by vasomotor and selected neuropsychiatric reporting signals, whereas Myfembree was characterized mainly by reproductive and bleeding-related reporting signals. These findings represent hypothesis-generating reporting differences rather than clinical incidence rates or causal risk estimates. Further pharmacoepidemiologic studies with denominator data and adjustment for patient-level confounding are needed to clarify comparative safety profiles.