Predictive VALUE of CD19 SERUM Levels for LONG TERM Therapeutic Response and Utility As Biomarker for Optimization, in Rheumatoid Arthritis Patients Treated with Rituximab

Background Rituximab (RTX) is a chimeric monoclonal antibody directed against CD20 antigens in B lymphocytes (BL) surface, producing selective depletion of BL, without affecting mature plasmatic cells. Current guidelines recommend optimisation of biologic therapy once remission has achieved in rheumatoid arthritis (RA) patients. A serum biomarker able to predict what patients would maintain remission after optimisation in RTX treated RA patients is lacking. Objectives To analyse the predictive value of CD19 serum level prior to RTX infusion for the long term maintenance of therapeutic response, and it’s utility as a biomarker for optimisation in RA patients treated with RTX. Methods all RA patients treated with RTX in our centre during 2016 and 2017 for at least 6 months, and with 12 months follow-up, were included. Demographic data, clinical data related to RA, including activity parameters (DAS28, HAQ, ESR, RCP), number of infections, optimisation and serum levels of lymphocytic subpopulations (CD19, CD3 and CD56) and immunoglobulins (Ig) prior to each RTX cycle were collected. Optimisation was defined as any dose decrease (lower dose for cycle and/or increase interval between cycles). Recurrent infections were defined as three or more infections in one year. Descriptive statistics, correlation between basal levels of CD19, CD3 or CD56, and activity parameters or Ig levels at 6 and 12 months, and association studies between lymphocyte subpopulations and optimisation or recurrent infections were performed. Results Thirty patients (25 females, 60±12 year) with RA (24 RF/aCCP +, 17±11 years from disease onset), treated with RTX during 52±40 months, with an accumulated dose of 13±10 g were included. At study inclusion, DAS28 was 3,77±1,65, HAQ was 1,28±0,78 and 22 had been optimised. CD19 levels were lower than 2% in 22 patients (79%), including 10 with undetectable levels. CD19 levels (cels/mm³) were lower in optimised patients (73 vs 14, p=,03), and were correlated with DAS28 at 6 months (r=0,5; p=,007) and 12 months (r=0,44; p=,03), and with ESR at 6 m (r=0,6; p=,002). CD19 levels did not correlate with IgG levels, and were not associated with recurrent infections. CD3 and CD56 levels did not show any relevant association or correlation. Conclusions In RA patients treated with RTX for more than 6 months, CD19 levels correlate with long term therapeutic response, being low or undetectable levels predictors of good outcome, without association with Ig levels or increased infections. Our results suggest that CD19 levels before every RTX cycle might be a useful biomarker to select candidate patients for optimisation with this therapy. Disclosure of Interest None declared