BACKGROUND: To describe the 12-month Clinical Disease Activity Index (CDAI) trajectory patterns and their determinants in patients with rheumatoid arthritis (RA) treated with subcutaneous abatacept. METHODS: This was a post hoc analysis of adult patients with RA from the Abatacept Best Care (ABC) study. A growth mixture model (GMM) with Bayesian imputation for missing endpoints was used to define CDAI trajectory groups over 12 months. Determinants of trajectory patterns were identified among baseline patient and disease characteristics using bivariate comparisons and multivariable logistic regression with stepwise selection. RESULTS: For the overall study sample (n = 256) mean (SD) age was 60.1 (11.5) years, 191 (75%) were female, and baseline CDAI was 30.1 (10.6). The GMM identified 3 trajectory groups that were classified as Rapid Responders (RR; n = 125 49%), Late Responders (LR; n = 96 38%), and Non-Responders (NR; n = 35 14%). RR had lower baseline CDAI, Disease Activity Score 28 - C-Reactive Protein, Patient Global Assessment (PtGA), Routine Assessment of Patient Index Data 3, Simplified Disease Activity Index, and Tender Joint Count compared to LR and NR (p < 0.001). Lower baseline PtGA, Rheumatic Disease Comorbidity Index, and lack of prior biologic use were significant predictors of RR (p < 0.03). CDAI scores at 3 months were not predictive of 12-month response to treatment. CONCLUSIONS: Approximately 85% of the abatacept-treated patients with RA in this study showed Rapid or Late CDAI reduction over 12 months. Trajectory-based analyses are informative and may have implications for clinical practice and research. TRIAL REGISTRATION: This study was registered on 06 September 2017 in ClinicalTrials.gov under the registration number NCT03274141.
Pope et al. (Tue,) studied this question.