OP0328 A UNIQUE PD1+CD38+ CD8+ T CELL POPULATION CHARACTERIZES CHECKPOINT INHIBITOR-ASSOCIATED INFLAMMATORY ARTHRITIS. (2nd June 2020)

• Record Type: Journal Article
• Authors: Wang, R.; Chan, K. K.; Cunningham-Bussel, A.; Donlin, L.; Vitone, G.; Tirpack, A.; Benson, C.; Keras, G.; Jonsson, A. H.; Brenner, M.; Bass, A.; Rao, D.
• Is Part Of: Annals of the rheumatic diseases. Volume 79(2020)Supplement 1
• Page Start: 201
• Page End: 202
• Publication Date: 2020-06-02
• Subjects: Rheumatism -- Periodicals; 616.723005
• DOI: 10.1136/annrheumdis-2020-eular.1075
• Languages: English
• ISSNs: 0003-4967

Abstract

Abstract : Background: Immune checkpoint inhibitors (CI) are monoclonal antibodies that block CTLA-4, PD-1 or PD-L1, resulting in cytotoxic T cell activation in the tumor microenvironment. They have revolutionized the management of metastatic cancer but unleash "immune related adverse events" in > 80% of treated patients, including inflammatory arthritis in ~4% 1 . CI-associated arthritis (CI-A) often presents as a symmetrical polyarthritis, phenotypically indistinguishable from rheumatoid arthritis (RA), but whether it shares cellular and molecular features of RA has not been determined. Objectives: To compare synovial fluid (SF) T cell populations from CI-A patients to those in patients with RA, phenotypically and functionally. Methods: We immunophenotyped SF mononuclear cells from patients with CI-A caused by anti-PD-(L)1 therapy (n=9), seropositive RA (n=5), and psoriatic arthritis (PsA) (n=5) using a 39-marker mass cytometry (CyTOF) panel. FlowSOM was used to cluster CD4 and CD8 T cells into 15 'metaclusters' based on multidimensional phenotypes. We used Kruskal-Wallis or Mann-Whitney tests to identify significantly altered populations (p<0.05), which we confirmed by biaxial gating. Flow cytometry was used to confirm SF findings in an independent cohort, and to identify cells of interest in peripheral blood. Cytokine staining was performed on sorted T cells populations after CDCD3/CD28 stimulation for 72 hours, followed by 4 hour PMA/ION+BRA/MON restimulation. Results: In CI-A patients, T cells represented 50% of SF mononuclear cells (53% CD4, 40% CD8), followed by monocytes (24%) and NK cells (8%), comparable to RA and PsA. However, FlowSOM analysis revealed expansion of a distinct population of PD-1 + CD38 hi CD127 - CD8 T cells (CD8 metacluster2) (Fig. 1 ). These cells comprised 30% of CD8+ SF T cells in CI-A, a 3.4-fold increase over RA/PsA, p=0.0002 (Fig. 2 ). Over 40% of these cells expressed Ki67 in CI-A, suggesting active proliferation. Flow cytometry on SF cells from an independent cohort of CI-A patients (n= 5) and RA/PsA comparators (n= 9) confirmed our findings. PD-1 + CD38 hi CD127 - CD8 T cells were also expanded in the blood of CI-A patients, where they represented 4.6% of CD8 Tcells, a 2.8-fold increase over RA, p = 0.0057. In addition to expressing high levels of PD-1, CD38 hi CD127 -, these CD8 T cells express other immune checkpoint receptors including ICOS and TIGIT. After in vitro stimulation, CD38 hi CD127 - CD8 T cells produced granzyme B along with TNF and IFN-γ at comparable levels to other CD8 populations, suggesting that they are not functionally exhausted. FlowSOM analysis of SF CD4 T cells in CI-A patients revealed the expansion of a subpopulation of CD4 cells with a similar surface phenotype of PD-1 + CD38 hi CD127 - (metacluster2, 10% of CD4s in CI-A, a 2.4-fold increase over RA/PsA, p=0.0047). In contrast, RA patients had a significantly expanded population of PD-1 hi ICOS+ CD4 T peripheral helper (Tph) cells (metacluster5, 30% of CD4s in RA, p=0.006), but these cells were not expanded in CI-A (Fig 3 ). Conclusion: CyTOF analysis of SF revealed a uniquely expanded PD-1 + CD38 hi CD127 - CD8 T cell population in CI-A not present in RA or PsA, and a similar PD-1 + CD38 hi CD127 - CD4 T cell population. These cells may contribute to the amplified immune response seen in CI-A patients. Further functional and transcriptional analysis of these cells will help to elucidate their function may reveal key mechanisms driving CI-associated immune related adverse events. References: [1]Kostine M. Ann Rheum Dis 2018;77(3):393-398 Disclosure of Interests: Runci Wang: None declared, Karmela Kim Chan: None declared, Amy Cunningham-Bussel: None declared, Laura Donlin Consultant of: Consultant – Genentech/Roche, Gregory Vitone: None declared, Aidan Tirpack: None declared, Caroline Benson: None declared, Gregory Keras: None declared, A. Helena Jonsson: None declared, Michael Brenner: None declared, Anne Bass: None declared, Deepak Rao Grant/research support from: Has received research grant support from Celgene and Merck., Consultant of: Has received consulting fees or honoraria from Merck, Pfizer, GlaxoSmithKine, Bristol-Myers Squibb, Janssen, and Scipher Medicine