THU0119 Smoking is negatively associated with clinical response to concomitant prednisone use in early rheumatoid arthritis

Abstract

Background: Smoking and overweight increase the risk of development of rheumatoid arthritis (RA) and have been associated with a reduced clinical response to several biological disease modifying anti-rheumatic drugs (bDMARDs) and methotrexate (MTX).1-6 The effect of smoking and overweight on clinical response to prednisone (pred) has not been investigated Objectives: To determine whether smoking and overweight are negatively associated with clinical response to pred in RA, as assessed by the disease activity score using 28 joints (DAS28). Second, to similarly determine the effect of smoking and overweight on the four individual components of DAS28. Methods: In the second Computer Assisted Management in Early Rheumatoid Arthritis trial (CAMERA-II), patients with early, DMARD naïve RA were randomised to an MTX based treatment strategy with addition of 10 mg pred or placebo daily for 2 years. We used data from the patients concomitantly treated with pred, n=127. We performed linear mixed model analyses with DAS28 over the 2 year trial period as outcome variable and smoking status (current yes/no) and BMI (in mg/kg2) as predictor variables, corrected for baseline DAS28, gender, rheumatoid factor (RF-test negative or positive), MTX administration route (subcutaneous or oral), MTX dosage (mg/week) and bDMARD use (yes/no), time (in months) and time2. For the second aim, we performed an extended mixed model analysis with a multivariate outcome, consisting of the individual components of the DAS28 (tender joint count 28, swollen joint count 28, erythrocyte sedimentation rate and VAS general health) and as predictor variables smoking status and BMI, correcting for important confounders. Results: Smoking data was available of 104/127 patients. Smoking was significantly associated with higher DAS28 over time, corrected for baseline DAS28, compared to non-smoking (table 1). No significant associations were found between BMI and response to pred, nor between smoking or BMI and any of the individual components of DAS28. Table 1 The association between smoking and DAS28 in patients treated with concomitant prednisone for 2 years (Table Presented) Outcome measure: DAS28 over 2 years; CI: confidence interval; MTX: methotrexate; bDMARD: biological disease modifying anti-rheumatic drug; BMI: body mass index; RF: rheumatoid factor. A positive estimate reflects a higher DAS28 over time, corrected for baseline DAS28, and thus a lesser effect of the concomitant prednisone therapy Conclusions: To our knowledge, this is the first time that smoking is shown to be a negative predictor for clinical response to pred in RA patients. Cessation of smoking needs to be encouraged in patients initiating bDMARDS, MTX and pred and in those already on these drugs.