119. Single-cell transcriptome profiling of rituximab-resistant tissue B cells in ANCA-associated vasculitis

Background: Failure of rituximab to adequately deplete tissue B cells may contribute to remission failures and high relapse rates seen after a single induction course of rituximab for ANCA-associated vasculitis (AAV). There is very limited information on the impact of rituximab on tissue B cells in AAV. Methods: We performed flow cytometric assessment (n=7) and transcriptomic profiling (n=2) of blood and nasal tissue lymphocytes before and after rituximab 1g x 2 on a prospective cohort of patients with PR3-AAV. Nb: this is an interim report of early data from an ongoing project. Results: Flow cytometric and scRNA-seq analysis of sequential nasal mucosal biopsies revealed incomplete depletion of B cells in the tissue after rituximab. Both analytical methods indicated that the residual nasal B cells were predominantly antigen-experienced mature B cells (memory and plasma cells), suggesting that antigen-inexperienced cell types (naïve and transitional) were more susceptible to rituximab-induced depletion in the tissue. This was in contrast to the effect of rituximab on the circulating peripheral B cell pool for which there was near-complete depletion of B cells, by standard FACS, without preferential targeting of specific subtypes. Persistent nasal B cells had expression profiles suggesting the capacity for antigen presentation and production of proinflammatory cytokines. Quantification of predicted cell-cell interactions based on receptor-ligand expression revealed several predicted interactions between B and T cell ligands and receptor gene expression, highlighting a complex interplay of stimulatory and inhibitory signalling pathways within the tissue microenvironment, and also revealing some potential mechanisof rituximab resistance. Conclusion: Examination of tissue lymphocytes demonstrates a failure of a single course of rituximab (1g x 2) to deplete B cells and a change in the repertoire to a memory phenotype likely to be relevant to early relapse after rituximab. Disclosures: The authors have no financial disclosures relevant to the submitted wor