Background Finding prognostic factors for treatment failure on synthetic disease modifying antirheumatic drugs (DMARD) in early treatment naive rheumatoid arthritis (RA) is a challenge.

Objectives To investigate whether baseline characteristics and disease activity variables add value to predict failure of classical DMARD treatment in a multivariable model.

Methods The study included DMARD naive patients consecutively diagnosed with RA according to the ACR/EULAR criteria 2010. The patients were enrolled between 1.10.2009 and 1.11.2012. They were followed for one year. Disease activity was registered in the DANBIO registry: number of swollen joints (NSJ) (38 joints), number of tender joints (NTJ) (40 joints), Health Assessment Questionnaire (HAQ), visual analog scales (VAS) 0-100 were used to assess pain, fatigue, patient and physician global assessment and DAS28-CRP, CRP, IgM-RF and anti-CCP. Treatment was oral methotrexate 15 mg per week initiated at time of diagnosis increased to 20 mg per week at week 6. If DAS28-CRP at this point was higher or equal to 3.2 and one or more swollen joints were present, treatment was intensified according to guidelines aiming at triple therapy (if tolerated). Intra-articular glucocorticoid injections were given in swollen joints. Treatment with biologics was applied according to guidelines. COX regression analysis was used to investigate if baseline characteristics and disease activity variables could predict failure of DMARD treatment requiring substitution or addition of biologic treatment. Primary outcome was failure DMARD defined as the need for addition of biologic treatment or change to biologic monotherapy.

Results 230 patients (41% males), with a mean age of 59 yrs (range: 16-85 yrs) were enrolled at baseline. Anti-CCP positive: 65%. IgM-RF positive: 60%. A total of 16 patients (7%) initiated biologic treatment during the first treatment year. Univariate COX regression analysis showed significant association between the need for biologic treatment and age (hazard ratio (HR) 0.95 (95% CI: 0.92 to 0.98), p<0.001; i.e. inversely associated), CRP (HR 1.01 (95% CI: 1.00 to 1.02), p=0.04), VAS fatigue (HR 1.03 (95% CI: 1.01 to 1.05), p=0.01)) and VAS global (HR 1.04 (95% CI: 1.01 to 1.06), p=0.007). In the subsequent multivariable analysis (incl. the above mentioned variables simultaneously) age and CRP remained significantly associated (HR 0.95 (95% CI: 0.91 to 0.98), p=0.002; HR 1.03 (95% CI: 1.01 to 1.04), p=0.002, respectively). After one year of follow-up no significant differences in disease activity was found between the patients still on DMARD treatment (DAS28-CRP: 2.1) compared to the 16 patients that had initiated biologic therapy (DAS28CRP: 2.3).

Conclusions In a pragmatic, single centre inception cohort, using the ACR/EULAR 2010 criteria for the diagnosis of early RA, regression analyses showed that young age and high CRP was significantly associated with failure on initial classical guideline applied DMARD strategy (p=0.002). The implications for clinical practise could be a necessity for more frequent visits for these individuals while treated with synthetic DAMARDs.

Disclosure of Interest None declared

DOI 10.1136/annrheumdis-2014-eular.4070