PT  - JOURNAL ARTICLE
AU  - Dafne Capelusnik
AU  - Daniel Aletaha
TI  - Baseline predictors of different types of treatment success in rheumatoid arthritis
AID  - 10.1136/annrheumdis-2021-220853
DP  - 2021 Oct 03
TA  - Annals of the Rheumatic Diseases
PG  - annrheumdis-2021-220853
4099  - http://ard.bmj.com/content/early/2021/10/03/annrheumdis-2021-220853.short
4100  - http://ard.bmj.com/content/early/2021/10/03/annrheumdis-2021-220853.full
AB  - Objective To perform a comprehensive analysis on predictors of achieving disease activity outcomes by change, response and state measures.Methods We used data from three rheumatoid arthritis (RA) trials (one for main analysis, two for validation) to analyse the effect of patient and disease characteristics, core set measure and composite indices on the achievement of different outcomes: response outcomes (% of patients achieving a relative response margin); state outcomes (remission or low disease activity, LDA) and change outcomes (numerical change on metric scales).Results We included patients from the ASPIRE trial (for analysis) and from the ATTRACT and GO-BEFORE trials (for validation). While lower disease activity components at baseline—except acute phase reactants—were associated with achievement of state outcomes (such as LDA by the Simplified Disease Activity Index, SDAI), higher baseline values were associated with change outcomes (such as SDAI absolute change). A multivariate analysis of the identified predictors of state outcomes identified best prediction by a combination of shorter disease duration, male gender and lower disease activity. For prediction of response, no consistently significant predictors were found, again, with exception of C reactive protein, for which higher levels at baseline were associated with better responses.Conclusion Prediction of treatment success is limited in RA. Particularly in early RA, prediction of state targets can be achieved by lower baseline levels of diseases activity. Gender and disease duration may improve the predictability of state targets. In clinical trials, included populations and choice of outcomes can be coordinated to maximise efficiency from these studies.All data relevant to the study are included in the article or uploaded as supplemental information.