Immune checkpoint inhibitor associated uveitis rates varies by medication class in international database.
My Session Status
Authors: Raheem Remtulla , Karin M. Oliver. McGill University.
Author Disclosures: R. Remtulla: None. K.M. Oliver: None.
Abstract Body:
Purpose: Immune checkpoint inhibitors (ICIs) inhibit the regulators of the immune system and have novel application in the treatment regimen of certain cancers. The three primary classes of ICIs are cytotoxic T lymphocyte associated protein 4 inhibitors (CTLA-1), programmed cell death protein 1 inhibitors (PD-1) and programmed cell death ligand 1 inhibitors (PD-L1). Unfortunately, ICIs have been associated with non-infectious uveitis. As occurrences of ICI associated uveitis are rare, there is little information on how the different classes of ICIs impact the presentation of uveitis. Using publicly available adverse event reporting, the rates of uveitis among the different classes were compared.
Study Design: Retrospective case control study of publicly available and anonymized data.
Methods: International adverse event reporting to the Food and Drug Administration between January 2010 and December 2021 of all patients with adverse reactions to CTLA-1, PD-1 and PD-L1 inhibitors were collected from the FDA Adverse Event Reporting System. Patient data collected included drug of interest, other medications adverse reactions, severity of reaction, reason for medication use, biologic sex and age. Patients treated with more than one class of ICI were not included in the analysis. To account for confounding factors Cochran-Mantel-Haenszel testing was employed.
Results: Data from a total of 39,190 patients were collected of which 4,508 were on CTLA-1s, 29,670 were on PD-1s and 5,012 were on PD-L1s. A total of 124 patients were identified as having ICI associated uveitis. The rate of uveitis was statistically higher among CTLA-1s (OR=4.74, p<0.001) and PD-1s (OR=4.36, p<0.001) than PD-L1s. Patient groups treated with different classes of ICIs, had significant differences in parameters including indication for medication (p<0.0001), biologic sex (p<0.0001) and age (p<0.0001). Despite accounting for these variations, the rate of uveitis was still statistically higher among those on CTLA-1 and PD-1 inhibitors over PD-L1 inhibitors (p<0.05).
Conclusions: The available data suggests that PD-L1s appear have a higher safety threshold for uveitis than CTLA-1s and PD-1s, even when accounting for confounding variables. Given that the different classes of ICIs inhibit different cellular targets, this study provides further insight in the mechanism and potentially the management of ICI associated uveitis.