Divergent Eosinophilic and Paradoxical Inflammatory Adverse-Event Signatures Across Five Type 2 Asthma Biologics: A FAERS Active-Comparator Disproportionality Analysis

BackgroundFive biologics target sequential nodes of the type 2 (T2) inflammatory axis in severe asthma and have each been anecdotally linked to eosinophilic or paradoxical inflammatory adverse events. Whether these reflect a class effect or drug-specific signatures is untested. MethodsWe performed a case-noncase disproportionality analysis of FAERS, 2022-2024. Asthma-indication reports naming tezepelumab, dupilumab, mepolizumab, benralizumab, or omalizumab as primary suspect were compared against the pooled other four (active-comparator design) across 7 pre-specified eosinophilic, vasculitic (including EGPA), cutaneous, and ocular outcomes. A signal was robust if the reporting odds ratio (ROR) 95% CI lower bound exceeded 1 and a Bayesian criterion was also met. ResultsAmong 50,360 reports (dupilumab n=29,652; mepolizumab n=8,846; omalizumab n=5,484; benralizumab n=4,372; tezepelumab n=2,006), benralizumab showed robust signals for EGPA (ROR 5.06), vasculitis (ROR 4.60), and peripheral eosinophilia (ROR 1.80). Mepolizumab showed a robust vasculitis signal (ROR 1.85); its similarly sized EGPA association narrowly missed the robustness criterion despite significance after correction for multiple testing. Dupilumab showed no EGPA/vasculitis excess but a robust ocular-inflammation signal (ROR 7.81). Tezepelumab and omalizumab showed no robust signal in any domain. EGPA/vasculitis onset occurred earlier after dupilumab (median 20-44 days) than anti-IL-5-axis agents (median 195-550 days; p[≤]0.012); mepolizumabs signal did not reproduce over the full 2015-2024 window (calendar-time confounding). ConclusionFAERS active-comparator analysis reveals drug-specific, not class-uniform, adverse-event reporting patterns: anti-IL-5-axis agents disproportionately report EGPA/vasculitis, and dupilumab disproportionately reports ocular inflammation with markedly earlier onset. These hypothesis-generating signals cannot establish causality and require confirmation through case-level or comparative pharmacoepidemiological studies.