Association between immune checkpoint inhibitor medication and uveitis: a population-based cohort study utilizing TriNetX database

Objective: To explore the associations between the use of immune checkpoint inhibitors (ICIs) and the risk of developing uveitis among cancer patients. Methods: Cancer patients who received ICI therapy and a comparison group of cancer patients who did not receive ICI therapy were retrospectively recruited from the TriNetX electronic heath-record registry. The outcome of interest was the development of new-onset uveitis. Propensity score matching based on a 1:1 ratio was conducted in order to reduce bias. Multi-variate cox proportional hazard models and Kaplan Meier method were also utilized to assess for the risk of uveitis among cancer patients who received ICI therapy. Results: 71931 cancer patients (54.7% male; 76.5% white; mean age at index 63.6 +/- 12.2 years) who received ICI treatment (ICI group) and 71931 cancer patients (54.7% male; 77% white; mean age at index 63.5 +/- 12.4 years) who never received ICI (comparison group) were recruited. Associated Kaplan-Meier curves showed significantly increased uveitis risk among the ICI group for all follow-up years (p<0.001). The risk of uveitis was also higher among the ICI group during the 144-month follow-up period with a hazard ratio (HR) of 2.39 (95% CI: 2.07-2.75). Increased risk for specific uveitis diseases, such as iridocyclitis, chorioretinal inflammation, retinal vasculitis, unspecified purulent endophthalmitis, pan-uveitis and sympathetic uveitis were found. Subgroup analysis demonstrated an elevated hazard ratio for the development of uveitis among ICI recipients, spanning individuals below the age of 65 as well as those aged 65 and older. The elevated hazard ratio for uveitis development among ICI recipients was also observed across all genders, among those of white and Asian ethnicities, those with smoking history, and those with comorbid conditions such as hypertension and dyslipidemia, in comparison to their non-ICI counterparts. An additional subgroup analysis on monotherapy versus combinatory ICI regimens was also conducted. Individuals who received monotherapy from the class of anti-PD-1 (HR:1.98 [CI: 1.65-2.37]) and anti-CTLA-4 (HR:5.86 [CI:1.99-17.24]) exhibited elevated hazard ratios for uveitis development compared to their non-ICI comparators. Those exposed to combinatory ICI regimens, specifically a combination of anti-PD-1 and anti-CTLA4 (HR: 5.04 [CI:3.55-7.16]), showed increased hazard ratios for uveitis development compared to their non-ICI comparators. In contrast, individuals exposed to a combination of anti-PD-1 and anti-PD-L1 (HR: 2.47 [CI:0.81-7.50]) did not demonstrate an increased risk for uveitis compared to their non-ICI comparators. Conclusion: A significantly increased risk for uveitis diseases was found among the ICI group from the first year of follow-up. Increased awareness should be promoted on the occurrence of uveitis among cancer patients receiving ICI therapy.