Harnessing quantitative proteomics to identify biomarkers of treatment response to etanercept in patients with rheumatoid arthritis

Abstract Background/Aims Tumour necrosis factor inhibitors (TNFi) are commonly prescribed to treat rheumatoid arthritis (RA), but response to TNFi is not universal and there are no validated pre-treatment biomarkers of treatment response at present. Proteins have many features that make them ideal potential biomarkers, as they carry out diverse biological processes, interact with drugs and can reflect post-translational modifications. This study aimed to identify protein biomarkers associated with response to TNFi in patients with RA. Methods Participants were recruited from a UK-based prospective multi-centre study of patients fulfilling either the 1987 ACR or 2010 ACR/EULAR classification for RA starting biologic treatment for the first time. Quantitative proteomics was performed using Sequential Window Acquisition of all THeoretical fragment ion spectra mass spectrometry (SWATH-MS) in sera from study participants before and after three months of treatment with the TNFi etanercept. Linear regression models were used to study the association between protein levels and the composite 28-joint count Disease Activity Score (DAS28) and its sub-components, adjusting for baseline covariates including age, biological sex disease duration and baseline DAS28. Sub-network analysis was carried out using the DIAMOnD algorithm, followed by functional enrichment analysis. Results Of the total 180 participants recruited, 134 (74.44%) were female, with a median age of 56.90 years and median pre-treatment DAS28 of 5.9. Eight individual proteins were found to be significantly associated with RA clinical outcome measures (see Table). Sub-network analysis identified the ontological theme with the strongest association as being positive regulation of response to endoplasmic reticulum (ER) stress (-log10(P) ∼ 7). Conclusion This longitudinal study of 180 patients with RA commencing etanercept has identified several putative protein biomarkers of treatment response and identified ER stress as an underlying pathway. The ER stress pathway has been previously reported to contribute to RA pathogenesis via synoviocyte proliferation and stimulation of pro-inflammatory cytokine production. Disclosure S.F. Ling: None. C. Yap: None. N. Nair: None. J. Bluett: None. A.W. Morgan: None. J.D. Isaacs: None. A.G. Wilson: None. K.L. Hyrich: None. A. Barton: None. D. Plant: None.