A BETTER LONG-TERM CLINICAL COURSE AND LESS RADIOGRAPHIC JOINT DAMAGE CAN BE PREDICTED BY AN EARLY GOOD RESPONSE TO THERAPY IN RA PATIENTS

Background: Methotrexate (MTX) is widely used as the disease modifying anti-rheumatic drug (DMARD) of first choice in the treatment of rheumatoid arthritis (RA) patients. Our strategy trial (CAMERA) already proved that intensive tight controlled MTX treatment is effective for RA patients and leads to better clinical effects after 2 years when compared to conventional MTX treatment. Objectives: To investigate whether an early good clinical response to treatment in patients with RA predicts a better long term clinical efficacy of medication and less radiographic joint damage compared to a less favourable early clinical response. Methods: Patients of a two-year multicentre randomized open label strategy trial (CAMERA study) were after 6 months of treatment classified as good responders (DAS28 ≤3.2 and improvement of >1.2) or non-responders (DAS28 >5.1 and improvement ≤1.2 or only improvement ≤0.6) according to the EULAR response criteria. The clinical and radiological variables over a period of 4 years as well as MTX use of the two groups were compared. Results: Of the 208 patients with complete 2 years data in the CAMERA study, at 6 months 74 (49 intensive vs. 25 conventional treated) patients fulfilled the EULAR good response criteria and 49 (12 intensive vs. 37 conventional treated) patients the non-response criteria. The difference in DAS28 between the early good and non-response groups decreased during the first two years, but after 4 years of treatment a difference still existed (of 0.8 units, p<0.001, although less than at 6 months (difference 3.3 units, p<0.001). During the entire 4 year period, the radiological progression of joint damage differed between both groups, the difference increasing over the years (at 4 yrs, it was 10 SharpvdHeijde units, p<0.01). Thus radiographic outcome was related to the classification of early good and early non-responders. Until 6 months after start of treatment the use of MTX was on average 15% lower (p<0.04) in the early non-response group when compared to the early good response group. In the period between 6 months and 2 years of treatment when clinical responses converged, the mean dose MTX in the early non-response group increased significantly and it decreased significantly in the early good response group, resulting in a mean 35% higher (p<0.001) MTX dose in the early non-response group. Conclusion: In both early good responders and early non-responders a significant decrease in clinical activity was seen after 2 years. However, the treatment effect was delayed and less in the early non-responders, which was still apparent after 4 years. Despite a higher MTX dose in the early non-response group over time, the clinical and radiological outcomes were worse compared to the early good response group at 4 years of treatment. These findings show that the early clinical response predicts the future course of disease and supports the concept of early aggressive treatment to prevent joint damage later on. References: 1. Verstappen, et al. Ann Rheum Dis. 2007; 66: 1443-9 Disclosure of Interest: none declaredCitation: Annals of the Rheumatic Diseases, volume 68, supplement 3, year 2009, page 135Session: Abstract Session: RA Can we predict treatment response? (Oral Presentations )