Effect of erenumab versus other migraine preventive medications on cardiovascular and cerebrovascular outcomes: A United States claims database-based observational cohort study

Objective: To estimate the real-world risk of cardiovascular events among patients with migraine treated with erenumab and other migraine preventive medications. BackgroundMigraine preventive treatment with calcitonin gene-related peptide (CGRP) pathway inhibitors, such as erenumab and others, may theoretically result in cardiovascular effects due to a lack of compensatory vasodilation with CGRP pathway inhibition. Methods: In this retrospective observational cohort study, we estimated the unadjusted cumulative risk (CR) of new-onset hypertension, acute myocardial infarction (MI), or stroke among patients with migraine newly treated with erenumab, other anti-CGRP pathway monoclonal antibodies (mAbs), standard oral preventive medications, and onabotulinumtoxinA using data from the MarketScan (R) Commercial and Medicare Supplemental medical claims database. Comparative analyses to assess the relative risk (RR) of vascular events were gated on the comparability of treatment groups with respect to baseline demographics and clinical characteristics. Potential bias due to unmeasured confounding was evaluated via negative control outcome (NCO) analyses. Confounding based on measured covariates and differential informative censoring were addressed with inverse probability weights. Results: A total of 108,019 new users of migraine preventive medications were included. Unadjusted CR (95% confidence interval [CI]) of hypertension at 12 months of treatment was: erenumab, 9.34% (8.79-9.89%); other anti-CGRP pathway mAbs, 9.42% (8.92-9.92%); standard oral preventive medications, 9.09% (8.77-9.41%); and onabotulinumtoxinA, 9.10% (8.39-9.81%). NCO analyses identified minimal concerns related to unmeasured confounding in erenumab versus other mAbs and erenumab versus onabotulinumtoxinA comparisons. Adjusted RRs (95% CIs) of acute MI and stroke, respectively, at 36 months of treatment were 1.02 (0.45-1.59) and 0.90 (0.56-1.25) for erenumab versus other mAbs and 0.87 (0.19-1.55) and 0.97 (0.42-1.52) for erenumab versus onabotulinumtoxinA. Conclusions: In this analysis of the MarketScan medical claims database, we found no difference in the risk of vascular events in patients treated with erenumab versus other anti-CGRP pathway mAbs or onabotulinumtoxinA.