0279: Therapeutic Exploration of Immune Checkpoint Inhibitor Induced Inflammatory Arthritis In Vivo

Sang Kim¹, Rene Rico², Naimah Turner², Trupti Trivedi², Theresa Guise² and Roza Nurieva², ¹The Univesrity of Texas MD Andesron Cancer Center, Pearland, TX, ²The University of Texas, MD Anderson Cancer Center, Houston, TX

Background/Purpose: Despite unprecedented clinical success, immune checkpoint inhibitors (ICIs) are associated with immune-related adverse events (irAEs), including arthritis (ICI-arthritis). Management of ICI-arthritis is challenging because immunomodulatory therapy for arthritis should not impede antitumor immunity. Understanding of the mechanisms of ICI-arthritis is critical to overcome this challenge, but the precise pathophysiology remains unknown due to lack of pre-clinical murine models.

Methods: We immunized type II chicken collagen emulsified in Complete Freund Adjuvants. From Day 19, we implemented PD-1 inhibitor or combined CTLA-4 and PD-1 inhibitors (combined ICIs) and measured arthritis score until sacrifice. At sacrifice, we harvested joint (knee), bone (tibia), and serum to perform immunologic, histologic, and micro CT analyses.

Results: Four to five days after initiation of the ICIs, mice started to develop arthritis de novo. Clinical and histologic, and micro CT analyses revealed more severe arthritis in mice receiving combined ICIs compared to mice receiving PD-1 inhibitor alone. Importantly, like humans, collagen-specific T helper (Th)17/T cytotoxic (Tc)17 cell signatures were enhanced in inflamed joints of mice with combined ICI arthritis. Kinetic analysis revealed that upon ICI therapy, CII-specific Th1/Tc1 cells undergo plasticity into Th17/Tc17 cells which is correlated with disease severity. Inhibition of cytokines promoting Th17 cell development, function, and/or survival, including IL-1b, IL-6, IL-17A, IL-23 and TNFa, successfully ameliorate arthritis disease progression. Finally, microbial dysbiosis contributes to ICI-arthritis disease pathogenesis by enhancing production of Th17 related cytokines.

Conclusion: By modifying the collagen-induced arthritis model, we observed that mice develop arthritis de novo with ICI implementation. This model recapitulates ICI-arthritis and will provide insights of ICI-arthritis disease pathogenesis as well as preclinical supports to develop ideal therapeutic strategies for ICI-arthritis without abrogating ICI-antitumor efficacy.


S. Kim: None; R. Rico: None; N. Turner: None; T. Trivedi: None; T. Guise: None; R. Nurieva: None.