AB0098 A unique subset of rheumatoid arthritis defined by a distinct serum cytokine profile

Background Research into the pathogenesis of rheumatoid arthritis (RA) has been focusing on identifying disease biomarkers that can predict response to therapy. Early work identified differing levels of pro-inflammatory cytokines within a cohort of RA patients [1]. A recent study looking at the role of a modified release glucocorticoid preparation in the management of RA may have identified a distinct subgroup of patients who are resistant to therapy [2] with raised circulating levels of the cytokines TNF-α, IL-4, IL-1 receptor antagonist (IL-1ra) and IL-1β compared to treatment responders. The concept of glucocorticoid resistance in RA has been observed in other studies and across the spectrum of inflammatory disorders [3,4]. Objectives This study explores the the possibility of identifying a treatment resistant phenotype subset of patients with RA through peripheral serum cytokine measurements. Methods 41 patients with RA were recruited from general rheumatology clinics in the South-West of the United Kingdom (Bristol and Plymouth). Additional exclusion criteria included active malignancy, active infection and co-existing auto-immune disease. Patient blood samples were taken in the morning to minimise diurnal cytokine variations and cytokine assays were performed using a MILLIPLEX® MAP multiplex assay kit at a commercial laboratory in Germany. Results Results were analysed according to the number of patients who had raised serum TNF-α, IL-4, IL-1ra and IL-1β above the cohort median for each cytokine. The frequencies of the number of patients who had 0,1,2,3 or all 4 cytokines raised were recorded. This revealed a biphasic distribution of patients with 16/41 patients having 3 or 4 raised cytokines versus 22/41 patients having only 1 or 0 cytokines raised. Only 3 patients had 2 raised cytokines. This is a much smaller figure than predicted, as we would expect the population to be normally distributed. The P value for the Chi squared test comparing the observed population versus the expected normal distribution was