Abstract
To evaluate the risk of ocular adverse events (AEs) associated with crizotinib using real-world data from FAERS and VigiBase, and to characterize signal patterns through disproportionality and time-to-onset (TTO) analyses. Reports from FAERS and VigiBase (2011-2025) were analyzed. Disproportionality was assessed using reporting odds ratio (ROR) at both system organ class (SOC) and preferred term (PT) levels. TTO analysis was estimated based on FAERS data. Crizotinib was consistently associated with ocular AEs across both databases (FAERS ROR = 3.46, 95% confidence interval [CI]: 3.30-3.63; VigiBase ROR = 3.34, 95% CI: 3.17-3.51). Frequent PTs included visual impairment, blurred vision, and photopsia. High RORs were also observed for common events such as photopsia (FAERS: ROR = 42.5; VigiBase: ROR = 49.42) and visual brightness (FAERS: ROR = 31.67; VigiBase: ROR = 214.25). The median TTO was 14 days, suggesting early onset during treatment. Crizotinib is associated with a distinct profile of ocular AEs, typically mild and early in onset. The detection of rare but strongly associated PTs underscores the need for routine ophthalmologic monitoring. These findings are biologically plausible, as crizotinib inhibits the MET and ROS1 signaling pathways, both of which are expressed in retinal tissue, thereby supporting the need for enhanced clinical vigilance in patients receiving crizotinib.