Ocular adverse event profiles associated with endocrine therapies for breast cancer: a global pharmacovigilance study using FAERS and VigiAccess.
Wang Xiaoping X, Zuo Wenzi W, Zhang Chengzhi C, Liu Ben B et al.
Ocular adverse events may affect tolerability during long-term endocrine therapy for breast cancer, but comparative real-world evidence remains limited. We characterized ocular adverse-event reporting patterns using the FDA Adverse Event Reporting System (FAERS) and VigiAccess. FAERS reports from Q1-2004 to Q1-2026 involving seven endocrine therapies were analyzed. Drug-specific signals were assessed using reporting odds ratio (ROR) and information component (IC) within a breast cancer indication-restricted, primary-suspect reference population. Class comparisons used multivariable logistic regression with Benjamini-Hochberg false-discovery-rate correction. Age, ovarian function suppression (OFS), and time to onset (TTO) were evaluated; VigiAccess was descriptive. Among 42,851 eligible reports, 2,528 contained ocular adverse events. Tamoxifen showed the broadest signal spectrum, while aromatase inhibitors (AIs) showed ocular-surface and posterior-segment signals. No ocular Preferred Terms met signal criteria for toremifene or elacestrant. FDR-adjusted comparisons showed broader retinal, macular, optic nerve, corneal, and vitreous reporting with selective estrogen receptor modulators (SERMs), particularly versus AIs. Among SERM users, age ≤45 years showed lower reporting odds than age 45-60 years (aOR 0.67, 95% CI 0.46-0.94). Concomitant OFS showed lower reporting odds among AI users (aOR 0.43, 95% CI 0.28-0.63), interpreted as exploratory. Median TTO was 79, 77, and 491 days for AIs, selective estrogen receptor degraders (SERDs), and SERMs. Ocular adverse-event reporting differed across endocrine therapies. These findings generate hypotheses for future pharmacoepidemiological, clinical, and mechanistic investigation but should not be interpreted as incidence, causality, or differential clinical risk.