Synovial inflammatory pathways characterize anti-TNF-responsive rheumatoid arthritis patients

Objectives: This study was undertaken to understand the mechanistic basis of response to anti-TNF therapies and determine if transcriptomic changes in the synovium are reflected in peripheral protein markers. Methods: Synovial tissue from 46 RA patients was profiled with RNA sequencing before and 12 weeks after treatment with anti-TNF therapies. Pathway and gene signature analyses were performed on RNA expression profiles of synovial biopsies to identify mechanisms that could discriminate among EULAR good, moderate and non-responders. Serum proteins encoded by synovial genes differentially expressed between EULAR response groups were measured in the same patients. Results: The gene signatures were able to predict good responder patients and pathway analysis identified elevations in immune pathways including chemokine signaling, Th1 and Th2 cell differentiation, and Toll-like receptor signaling uniquely in good responders. These inflammatory pathways were correspondingly down-modulated by anti-TNF therapy only in good responders. Based on cell signature analysis, lymphocyte, myeloid and fibroblast cell populations were elevated in good responders relative to non-responders, consistent with the increased inflammatory pathways. Cell signatures which decreased following anti-TNF treatment were predominately associated with lymphocytes and fewer were associated with myeloid and fibroblast populations. Following anti-TNF treatment and only in good responders, several peripheral inflammatory proteins decreased consistent with corresponding synovial gene changes. Conclusions: Collectively, these data suggest that RA patients with robust responses to anti-TNF therapies are characterized at baseline by immune pathway activation, which decreases following anti-TNF treatment. Understanding mechanisms that define patient responsiveness to anti-TNF may assist in development of predictive markers of patient response and earlier treatment options. Overall design: Forty-six RA patients fulfilling 2010 ACR/EULAR RA Classification Criteria were enrolled at the Centre for Experimental Medicine and Rheumatology, Barts and The London School of Medicine, Queen Mary University of London, UK. The study received ethical approval from the local UK Health Research Authority (10/H0801/47) and all patients gave written informed consent. Patients had clinically defined synovitis and were eligible to start anti-TNF according to UK NICE guidelines (failure of at least 2 csDMARDs and DAS28 >= 5.1). Upon enrolment and acquisition of demographics, current medications and clinical disease parameters, including CRP, ESR, RF/ACPA positivity/titer and DAS28, patients underwent minimally invasive US-guided synovial biopsy (17) of the most inflamed joint (ultrasound synovial thickening score >=2) (17) at baseline and 12 weeks after treatment with anti-TNF (19 etanercept, 27 certolizumab pegol). Duration of treatment was ensured up to primary endpoint and there were no treatment interruptions.