THU0097 PREDICTIVE VALUE OF IMMUNOLOGICAL AND IMAGING BIOMARKERS ON ACHIEVING GOOD CLINICAL RESPONSE AT 6 MONTHS IN RHEUMATOID ARTHRITIS PATIENTS TREATED BY INTRAVENOUS BDMARDS

Abstract

Background:RA is the most prevalent chronic inflammatory rheumatism, responsible of functional impairment.Objectives:To investigate the value of biological and imaging biomarkers on predicting good clinical response at 6 months, in RA patients initiating IV bDMARD.Methods:From 2008 to 2017, 317 RA patients fulfilling ACR 1987 and/or ACR-EULAR 2010 criteria for RA, initiated IV bDMARDs in our department of Rheumatology. Patients were excluded in cases of lack of information on disease activity assessment before and at 6 months of treatment and on immunological status and titers (ACPA, RF, ANA) at baseline. For patients receiving successive IV bDMARDs during this time period (n=30), a randomization permitted to select 1 treatment sequence for the analysis. On 173 patients eligible to the study, 4 were loss to follow-up and 14 stopped treatment due to adverse events before 6 months. Clinical, biological and imaging (US and RX) data were collected when available at baseline. US examination was performed on 12 joints (wrist, MCP2-3-5 and MTP2-3-5) with qualitative and quantitative evaluation on B mode and Power Doppler (PD) for synovitis, tenosynovitis and erosion. The modified Sharp/van der Heijde erosion score was performed by 2 independent readers blindly from clinical and US informations. Good clinical response was defined by a DAS 28 < 3.2 and/ or DAS 28 decrease > 1.2 at 6 months. Only variables with a p<0.2 in univariate analysis were included in the multivariate model.Results:On 155 RA patients, 11 present a disease duration < 2 year, 44 (28.3%) were on first line of IV bDMARDs and 111 patients received at least one IV bDMARD (mean 2.5 (1.3)).Table 1.Characteristics of the patients (n=155) at baselineVariablesN (%)Mean (SD)Clinical characteristicsAge (years)54.8 (12.2)Female113 (72.9)Disease duration (months)166.9 (118.8)DAS 285.2 (1)TreatmentCorticosteroids / dose (mg/day)99 (85.3)10.9 (6)Monotherapy56 (36.1)IV bDMARDAbatacept27 (17.4)Infliximab11 (7.1)Rituximab84 (54.2)Tocilizumab33 (21.3)ImmunologyACPA + /titer(IU)132 (85.2)618.5 (791.0)RF + /titer (IU/ml)114 (74.5)184.7 (351.3)ANA + / level87 (56.1)1453 (3836)RadiographySharp’s erosion score (n=110)49.4 (46.2)USNb Erosion (n=95)3.0 (2.3)Nb B mode Synovitis (n=128)6.0 (4.1)Nb PD+ Synovitis (n=130)4.8 (3.8)Nb B mode Tenosynovitis (n=129)1.6 (2)Nb PD+ Tenosynovitis (n=129)1.3 (2.1)At 6 months, 87 patients (56.1%) were in good clinical response. Predictive values of biomarkers are presented in table 2.Table 2.Variables predictive of a good clinical response at 6 monthsBiomarkersResponseMultivariate Logistic regression AnalysisAllN = 101Response(N=60)OR (CI95%)P valueImmunology RF +7551 (68.0%)5.1 (1.8-14.4)0.002 ACPA +8756 (64.4%) ANA +5536 (65.5%)Radiography Erosive RA7448 (64.9%)Ultrasonography Erosive RA8855 (62.5%) Nb B mode synovitis10160 (59.4%)1.2 (1.1-1.4)0.002 Nb PD+ synovitis10160 (59.4%)All qualitative variables with a p value <0.2 on bivariate analysis were incorporated in the multivariate model (RF +, ACPA +, US erosive RA, Nb B mode synovitis, Nb PD+ synovitis, RX erosive RA). Only patients with all data available are incorporated in the multivariate logistic regression analysis (n=101/155)Conclusion:We showed that positive RF was predictive of good clinical response to IV bDMARDs. For the first time, we demonstrated that number of US B-mode synovitis was also predictive to good clinical response.Disclosure of Interests:Stephane Giuliani Grant/research support from: BMS, Benjamin Laurent Grant/research support from: BMS, Hella MEZGHANI Employee of: BMS, Isabelle Duprat-Lomon Employee of: BMS, Amandine Luc Grant/research support from: BMS, Marcelo De carvalho Bittencourt Grant/research support from: BMS, Cedric BAUMANN Grant/research support from: BMS, Isabelle CHARY VALCKENAERE: None declared, Damien LOEUILLE: None declared