SAT0032 Ex vivo analysis of autoantigen-specific T cell responses using a multi HLA-CLASS II tetramer approach

Background HLA class II tetramers allow the direct ex vivo enumeration and phenotypic characterization of antigen-specific T cells and have proven a useful tool in different settings, e.g. allergy and vaccination. In rheumatoid arthritis (RA), many candidate auto-antigens have been characterized and are believed to play a role in the disease. Objectives Based on flow cytometry we developed a multi HLA-class II tetramer approach that renders it possible to look at numerous specificities simultaneously and is sufficiently sensitive at the same time as auto-reactive T cells are likely to be rare. We focused on validated citrullinated T cell epitopes previously studied in RA, namely a-enolase, fibrinogen and CILP. To test the robustness and sensitivity of our multi-tetramer assay, we performed repeated experiments on PBMCs of HLA-DRB1*04:01-positive healthy controls as well as HLA-DRB1*04:01-positive RA patients. Results First, we examined the sensitivity of the panel by assessing PBMC from healthy donors and we could detect low frequencies of auto-reactive T cells in these samples (1–10 per million CD4), mostly displaying a naive phenotype, which was in sharp contrast to influenza-specific T cells in the same donors which were 10–20 fold higher in numbers and of memory phenotype. Next, we tested the robustness of the panel by running technical repeats of all healthy donor samples, which yielded similar frequencies. Thereafter, we focused on RA patient PBMC obtained from repeated blood draws (2–4 weeks apart). Also in these samples, frequencies ranged between 1–10 tetramer-positive cells per million CD4+ T cells, i.e. similar as in healthy controls. Moreover, not all T cell specificities were present in all patients. Still, we found the frequencies to be stable in the repeated blood draws in approximately half of the individuals, implicating that frequencies close to 1 cell per million CD4+ T cells is borderline of what we can stably detect. Importantly, the patient samples utilized were not taken from time points of active disease, where we hypothesize frequencies to be elevated. This may be the case also in synovial fluid and as a proof of principle, we assessed synovial fluid samples of HLA-DRB1*04:01+ RA patients and were able to detect auto-reactive T cells at frequencies comparable to the ones found in peripheral blood. Conclusions In conclusion, we developed a sensitive tetramer panel allowing the simultaneous enumeration and phenotypic characterization of antigen-specific T cells in ex vivo samples from RA patients that can be used for monitoring and in-depth studying of these cells during the course of disease and treatment in individual patients. Disclosure of Interest None declared