Persistence to OnabotulinumtoxinA or Calcitonin Gene-Related Peptide Monoclonal Antibody Therapy Among Patients With Migraine: A Retrospective Cohort Study
Brian Talon1, Christine Sullivan1, Meghana Karnik-Henry1, Seema Soni-Brahmbhatt1, Carlton Anderson1, Steven Kymes1, Stephane A Regnier2
1Lundbeck LLC, 2H. Lundbeck A/S
Objective:
To compare treatment persistency among patients with migraine on calcitonin gene-related peptide (CGRP) monoclonal antibodies (mAbs; erenumab, fremanezumab, galcanezumab, or eptinezumab) or onabotulinumtoxinA.
Background:
To date, real-world evidence on persistence to CGRP mAbs or onabotulinumtoxinA have excluded eptinezumab.
Design/Methods:
This retrospective study used IQVIA PharmMetrics data. Adult patients with migraine treated with a CGRP mAb or onabotulinumtoxinA who had 12 months of continuous insurance enrollment before starting treatment were included. A “most recent treatment episode” analysis was used in which the most recent episode was defined as the latest treatment period with the same drug (CGRP mAb or onabotulinumtoxinA) without a 15-day gap in medication supply on/after 25June2020‒31December2021. Patients were indexed at the start of their most recent episode. Patients were non-persistent and discontinued the therapy associated with their most recent episode if there was ≥15-day gap in medication supply. A Cox proportional-hazards model estimated the discontinuation hazard between treatments. The gap periods and cohort definition were varied in sensitivity analyses.
Results:
The study included 66,576 patients (median age 46 years, 88.6% female). More eptinezumab-treated patients had chronic migraine (727/1074), ≥3 previous uses of acute (323/1074) or preventive (333/1074) therapies, and more prior treatment episodes than other treatment groups (3 [Q1-Q3, 2-5]). Based on a 15-day treatment gap, patients on subcutaneous CGRP mAbs had 32% (95%CI: 1.185, 1.494; erenumab) to 58% (95%CI: 1.418, 1.796; fremanezumab) higher discontinuation hazard than those receiving eptinezumab, with this relationship attenuated but significant with 30-day and 60-day treatment gaps. There was no significant difference in the discontinuation hazard between eptinezumab and onabotulinumtoxinA. Based on a 15-day treatment gap among patients who newly initiated therapy, the discontinuation hazard of subcutaneous CGRP mAbs remained significantly higher compared to eptinezumab and onabotulinumtoxinA.
Conclusions:
Eptinezumab had similar persistency with treatment to onabotulinumtoxinA, which was higher than subcutaneous CGRP mAbs.