AB0267 Baseline predictors of response to methotrexate in early rheumatoid arthritis

Abstract

Background: The disease activity score (DAS28) is widely used to assess response to treatment in early rheumatoid arthritis. Few studies have looked at the individual components of the DAS at baseline to predict drop in DAS28 in early disease. Specifically at 3, 6 and 12 months. Predicting response early in disase would enable more targetted treatments to be given and patients selecteed early for more aggressive disease control. Although certain biomarkers have been advocated for use and have shown some promise.1,2 there is still a place for exploring using more clinically derived indicators. Objectives: To determine whether DAS28 response at 3, 6 and 12 months could be predicted by baseline components of the DAS28 Methods: The study used patients from the multicentre UK based RAMS study. All patients with early rheumatoid arthritis starting methotrexate were enrolled. We used data from a single centre in the North west of England for the analysis. DAS28 and it's components were recorded at baseline and at 3 6 and 12 months. Baseline components of the DAS28 were used to predict the change in DAS from baseline to 3 6 and 12 months. Linear regression was used with subsequent adjustment for age and gender. Results: 120 patients were enrolled in the study, median age was 62.4 years (IQR42.1, 72.5), 81 (67%) were female. Median das28 at baseline was 5.3 (IQR 4.2,6.2). Duration of symptoms was 9 months (IQR 2,11). There was a drop of DAS28 of 1.49 at 3 months (IQR 0.57,2.45) and at 6 and 12 months it was 1.84 (IQR 0.48,2.91) and 1.63 (IQR 0.72,2.85) respectively. At three months the only baseline predictor of change in DAS28 was the patient global assessment unadjusted and adjusted for age adjusted beta 0.012 95% CI 0.002, 0.023 (p=0.02). At six months the baseline tender joint count adjusted beta 0.08 95% CI 0.015, 0.154 (p=0.01) as well as swollen joint count adjusted beta 0.05 95% CI 0.004,0.095 (p=0.02) also predicted reposnse in addition to the patients global assessment adjusted beta 0.016 95% CI 0.002, 0.03 (p=0.01), but not the ESR or CRP at baseline (p=NS). At 12 months the only predictors of response was the baseline CRP adjusted beta 0.02 95% CI 0.005, 0.04(p=0.02). Conclusions: In this cohort of patients with early disease, response to methotrexate could not be reliably predicted using baseline measures and the ability to predcit patients who would improve could not be found in this cohort. Further work on other clinical biomarkers to predict response is needed.