Ocular Surface Disease as an Immune-related Adverse Event of Immune Checkpoint Inhibitor Therapy.
Abstract
Purpose
Ocular surface disease, encompassing scleritis, episcleritis, keratitis, and sicca syndrome, is an increasingly recognized immune-related adverse event of immune checkpoint inhibitors (ICIs), yet its pharmacovigilance profile has not been systematically characterized across all approved ICI classes. We conducted a disproportionality analysis of the FDA Adverse Event Reporting System to quantify reporting odds ratios (RORs) for ocular surface end points across the PD-1, PD-L1, CTLA-4, and LAG-3 inhibitor classes.
Methods
Adverse event reports were extracted from FDA Adverse Event Reporting System using OpenVigil 2.1 from database inception through March 2026. Disproportionality was assessed using the ROR with 95% confidence intervals for FDA-approved ICI agents. Primary end points were immune-mediated scleritis, episcleritis, sicca syndrome, and keratitis. Secondary end points included scleritis (NOS, not otherwise specified), ulcerative keratitis, dry eye, and superior limbic keratoconjunctivitis.
Results
The most notable signals were for immune-mediated scleritis: pembrolizumab generated an ROR of 353.57 (95% CI: 114.03-1096.33, n = 6) and nivolumab an ROR of 120.84 (95% CI: 32.71-446.40, n = 3). Sicca syndrome signals were significant for nivolumab, atezolizumab, durvalumab, and ipilimumab. Keratitis signals were significant for nivolumab, pembrolizumab, and relatlimab. Ulcerative keratitis was disproportionately reported for nivolumab and pembrolizumab.
Conclusion
Immune-mediated scleritis generated the strongest disproportionality signal in this analysis among PD-1 inhibitors, although these estimates are based on small case counts and wide confidence intervals and should be regarded as hypothesis-generating. Sicca syndrome and keratitis were reported across multiple ICI classes, with the keratitis signals for nivolumab and pembrolizumab supported by the largest case counts. As a disproportionality analysis, this study cannot establish incidence, comparative risk, or causality; the findings support clinical vigilance and further epidemiologic study rather than definitive causal conclusions. Physicians managing ICI-treated patients should have a low threshold for ophthalmological referral when ocular symptoms arise.