OP0185 IMMUNOPHENOTYPIC ABNORMALITIES REMINISCENT OF ESTABLISHED RHEUMATOID ARTHRITIS ARE PRESENT IN ACPA(+) AT-RISK INDIVIDUALS

Background

RA-associated autoantibodies (RF, ACPA) are detectable before the onset of clinically-apparent inflammatory arthritis (IA) and classifiable RA (i.e. “Clinical RA”), defining a state known as ‘At-Risk’ for future RA. However, RA prevention trials in individuals in this At-Risk state have shown limited success in prevention. A challenge to developing more effective preventive interventions is that the immune phenotype of the At-Risk state remains incompletely understood.

Objectives

The purpose of this study was to (1) evaluate the hypothesis that ACPA(+) At-Risk individuals have immunopathologic features of Clinical RA, and (2) determine additional molecular features that define the At-Risk state. We propose that such features may help to identify biomarkers and causal pathophysiologic mechanisms that could be treatment or prevention targets. Furthermore, At-Risk immune features could improve prediction of future Clinical RA, as well as be used to monitor responses to therapeutic interventions.

Methods

Using a cross-sectional methodology, we analyzed ACPA(+) individuals without baseline IA (i.e. At-Risk) and age, sex and BMI matched ACPA(-) controls without IA (Table 1). We tested plasma proteomics, peripheral blood mononuclear cell (PBMC) composition by flow cytometry, and PBMC transcriptomes by single-cell RNA-sequencing (scRNA-seq). We further compared these findings to immunophenotypes reported in patients with Clinical RA. All studies were performed as part of a collaborative effort between the Allen Institute for Immunology, University of California San Diego, University of Colorado and Benaroya Research Institute.

Results

We observed significant differences in plasma levels of 138 proteins between ACPA(+) individuals vs. controls, including IL-6 (log2 fold change (FC) = 0.59-0.62, adjusted p value (padj) < 0.05), IL1B (log2 FC = 0.44, padj = 0.006), and CCL3 (log2 FC = 0.60, padj = 0.002). These changes also included additional elevated inflammatory chemokines and cytokines that have been reported in Clinical RA. B cell receptor signaling was one of the most significantly enriched protein pathways in ACPA(+) individuals. These proteomic alterations in ACPA(+) individuals were accompanied by numerous PBMC phenotypic changes, notably an increase in IgD- CD27- B cells (log2 FC = 0.814, padj = 0.002). Furthermore, compared to controls, scRNA-seq of IgD- CD27- B cells from ACPA(+) individuals expressed lower levels of genes previously shown to be increased in IgD- CD27- B cells or autoimmune processes, including SAMHD1 (log2 FC = -0.34, padj < 0.05), ARID3A (log2 FC = -0.32, padj < 0.05), and ZC3H12D (log2 FC = -0.30, padj < 0.05), suggesting dysregulation of this compartment at the preclinical stage.

Conclusion

ACPA(+) At-Risk individuals demonstrate numerous immune-related changes that are similar to changes in established Clinical RA, suggesting the ACPA(+) At-Risk state is on a continuum with Clinical RA. Highlighted are plasma proteomics demonstrating cytokine/chemokine elevations and B cell signaling pathways, and PBMC studies demonstrating specific B cell phenotypes and gene expression. These findings provide insights to potential immunopathologic mechanisms that may impact how we define the ‘At-Risk’ state, may improve prediction of future RA, and suggest targets for interventions or intermediate endpoints in trials for RA prevention.

REFERENCES:

NIL.

Acknowledgements:

NIL.

Disclosure of Interests

Mark Gillespie: None declared, Ziyuan He: None declared, Samir Rachid Zaim: None declared, Lauren Okada: None declared, Julian Reading: None declared, Alex Heubeck: None declared, Charles Roll: None declared, Palak Genge: None declared, Morgan Weiss: None declared, Cole Phalen: None declared, Regina Mettey: None declared, Cate Speake: None declared, Jane Buckner Consultant of: BMS, GentiBio, David Boyle: None declared, Kristen Demoruelle: None declared, Kristine A. Kuhn: None declared, Fan Zhang: None declared, Thomas Bumol: None declared, V.Michael Holers: None declared, Peter Skene: None declared, Xiao-jun Li: None declared, Gary Firestein Grant/research support from: Eli Lilly, Troy Torgerson: None declared, Kevin Deane Consultant of: Werfen, ThermoFisher, BMS, BI, Grant/research support from: Scipher Medicine, Gilead, Adam Savage: None declared.