Inflammatory and bone biomarkers/composites as a predictive tool for clinical characteristics of rheumatoid arthritis patients

Abstract

Rheumatoid arthritis (RA) is related to alterations in different inflammatory and connective tissue biomarkers. The diagnostic values and the factors affecting these biomarkers are conflicting. In the present study, a bone-related composite (B-composite), made from the z-score of stromelysin-1 (MMP3), colony-stimulating factor 2 (CSF2), and osteopontin (OPN), and I-composite, reflecting immune activation, made from the z-score of tumor necrosis factor-α (TNFα), interferon-γ (INFγ), and vascular endothelial growth factor-A (VEGF) were examined in RA patients. The biomarkers were measured by ELISA technique in 102 RA patients and 58 age-matched healthy control subjects. Serum MMP3, TNFα, IFNγ, and CSF2 showed significant elevation in RA patients. Multivariate general linear model (GLM) analysis revealed a significant high effect of diagnosis on biomarkers' level (partial η2 = 0.415). Duration of disease is significantly associated with VEGF, OPN, and B-composite and negatively correlated with TNFα. B-composite is significantly associated with CRP. A significant fraction of the DAS28 score variance can be explained by the regression on zlnINFγ. The variance in the CRP was explained by zlnOPN and B-composite. More than half of anti-citrullinated protein antibodies (ACPA) variation can be explained by the regression on serum MMP3 and I-composite. The top 3 sensitive predictors for RA disease are INFγ, MMP3, and TNFα. B-composite is associated with the duration of disease and CRP. At the same time, I-composite is negatively associated with the ACPA level. The biomarker composites have potential use as RA disease characteristic biomarkers.