Background/Purpose: The analysis of synovial tissue has been instrumental in advancing our understanding of Rheumatoid Arthritis (RA) pathogenesis. As an alternative, the identification of serum biomarkers that mirror synovial activity offers a promising and less invasive strategy for diagnosing and monitoring RA. This approach could address existing challenges and significantly enhance the implementation of precision medicine in the management of RA. The objectives of this study were: 1- To characterize synovial tissue-secreted proteins in early RA patients that contribute to serum inflammatory profiles, 2- To identify subsets of early RA patients with distinct serum levels of the protein signature and evaluate their association with disease activity and clinical response to conventional DMARDs (cDMARDs).

Methods: Synovial explants obtained via ultrasound-guided biopsies from 15 early RA patients were cultured in vitro for 24 hours to profile secreted proteins using proximity extension assay (PEA) technology (Olink), targeting 92 inflammation-related proteins. Proteomic profiles were also evaluated in serum samples from the same subjects, and correlation studies were performed. Validation of the identified protein signature was performed in an independent cohort of 86 early RA patients, aiming to assess its association with clinical response to cDMARDs.

Results: Eleven synovial-secreted proteins (CXCL6, CXCL9, CXCL11, DNER, FGF-5, IL1α, MCP-1, βNGF, STAMBP, TNF, TNFβ) showed significant correlations with their serum levels in early RA patients. These proteins are involved in biological processes closely linked to RA pathophysiology, such as cell proliferation, cytokine/chemokine signaling, neutrophil chemotaxis and inflammation. Clinically, higher levels of this inflammatory signature were observed in RA patients with active disease, elevated acute-phase reactants and high titres of autoantibodies. Unsupervised clustering in an independent cohort of 86 early-stage RA patients treated with cDMARDs identified two patient groups with distinctive baseline serum protein profiles. The group with higher levels of the inflammatory signature exhibited higher disease activity and a better clinical response to cDMARDs after 6 and 12 months of treatment. Machine learning algorithms identified clinical and molecular signatures as potential predictors of response to cDMARDs treatment with high accuracy, which was further increased when both features were integrated in a mixed model (AUC: >0.8).

Conclusion: This study identifies a synovial tissue–secreted protein signature that contributes to both the circulating inflammatory profile and clinical characteristics of early RA patients. This signature stratifies early RA patients into subgroups with distinct clinical responses, demonstrating its potential as a robust predictor of therapeutic response to cDMARDs. These findings underscore the value of serum biomarkers reflecting the molecular profile of the synovium to advance personalized medicine in RA. Funded by Instituto de Salud Carlos III (ISCIII) and co-funded by the European Union. Supported by EU/EFPIA IMI-JU 3TR, ISCIII (PI24/00959, CD21/00187, RICOR-21/0002/0033) and RYC2021-033828-I.

« Back to ACR Convergence 2025

ACR Meeting Abstracts - https://acrabstracts.org/abstract/integrating-synovial-protein-signatures-and-serum-profiles-to-predict-disease-activity-and-treatment-response-in-early-rheumatoid-arthritis/