Serum levels of 14-3-3η protein supplement C-reactive protein and rheumatoid arthritis-associated antibodies to predict clinical and radiographic outcomes in a prospective cohort of patients with recent-onset inflammatory polyarthritis

Background: Age, C-Reactive Protein (CRP) and autoantibodies (Abs) are associated with worse prognosis in patients with recent-onset inflammatory polyarthritis (EPA). Serum 14-3-3 eta protein is a joint-derived biomarker that up-regulates cytokines and enzymes that perpetuate local and systemic inflammation and may contribute to joint damage. Our objective was to evaluate, over a 5-year prospective period of observation, the additional prognostic potential of serum 14-3-3 eta protein in EPA patients. Methods: Clinical variables, serum and radiographs (scored according to the Sharp/van der Heijde (SvH) method) were collected serially. Relationships between serum 14-3-3 eta protein and other biomarkers were computed with Spearman correlations. Outcomes were Simple Disease Activity Index (SDAI) scores and joint damage progression:Delta SvH for SvH score and Delta Erosion for its Erosive component. The additional predictive contribution of 14-3-3 eta was defined using generalized estimating equations (GEE) and generalized linear mixed models (GLMM). Results: Among 331 patients, baseline 14-3-3 eta was >= 0.19 and >= 0.50 ng/ml in 153 (46.2 %) and 119 (36.0 %), respectively; CRP was > 8.0 mg/L in 207 (62.5 %), and at least one Ab (Rheumatoid Factor, anti-CCP2 or anti-Sa/citrullinated vimentin) was positive in 170 (51.5 %). Elevated 14-3-3 eta levels moderately correlated with positive Abs, but not with elevated CRP. Baseline 14-3-3 eta >= 0.19 ng/ml was associated with more radiographic progression over 5 years. The optimal levels of baseline 14-3-3 eta to predict radiographic progression was defined by ROC curves at 0.50 ng/ml. Levels of 14-3-3 eta >= 0.50 ng/ml at baseline were associated with lower likelihoods of ever reaching SDAI remission (RR 0.79 (95 % CI 0.64-0.98), p = 0.03) and higher subsequent progression of Total and Erosion SvH scores. Elevated levels of 14-3-3 eta during follow-up also predicted higher subsequent progression, even in patients in SDAI remission. Decreases of 14-3-3 eta levels by at least 0.76 ng/ml and reversion to negative during follow-up associated with less subsequent radiographic progression. In multivariate models, elevated 14-3-3 eta interacted with positive Abs, elevated CRP and older age to predict subsequent radiographic progression. Conclusions: Levels of 14-3-3 eta protein >= 0.50 ng/ml predict poorer clinical and radiographic outcomes in EPA, both at baseline and after initiation of treatment, even in SDAI remitters. 14-3-3 eta, CRP, age and Abs represent independent predictors of subsequent joint damage.