Abstract
BACKGROUND: Roflumilast is an add-on therapy for COPD following exacerbations, but real-world safety data in the United States are limited. OBJECTIVE: This study aimed to identify safety signals associated with roflumilast initiation through a high-throughput signal detection algorithm. METHODS: Using sequence symmetry analysis (SSA), we analysed Marketscan databases for new roflumilast users (2011-2021). We screened for adverse effects across 211 therapeutic classes within 365 days of initiation. Sensitivity analyses were conducted by sex, age and observation period. Crude and adjusted sequence ratios (cSR and aSR) were reported with 95% confidence intervals (CIs). RESULTS: Among 11 091 patients (53% aged 65+, 52% female), 32 safety signals were identified. Strong associations were observed with antithyroid agents (aSR, 4.18; 95% CI: 1.66-11.95), parathyroid hormones (aSR, 3.09; 95% CI: 1.56-6.44), haematopoietic agents (aSR, 2.55; 95% CI: 1.07-6.49) and meglitinides (aSR, 2.37; 95% CI: 1.15-5.35). While many signals aligned with prior clinical trial data, novel associations with antithyroid and parathyroid agents were discovered. CONCLUSION: In our study, we detected 32 safety signals for roflumilast, including notable associations with antithyroid agents and parathyroid hormones. Future investigations using more robust study designs are warranted to evaluate those signals.