INFLAMMATORY CYTOKINE PROFILING IN RHEUMATOID ARTHRITIS PATIENTS UNDERGOING BIOLOGIC THERAPY

Abstract

Rheumatoid arthritis (RA) is a chronic autoimmune disorder characterized by systemic synovitis and progressive joint damage, driven by complex cytokine-mediated immune responses. This study aimed to identify inflammatory cytokine signatures predictive of therapeutic response in RA patients undergoing biologic therapy. A prospective cohort of 120 RA patients was evaluated through multiplex cytokine assays at baseline, 3 months, and 6 months, alongside clinical and serological assessments. Key findings revealed that responders exhibited significantly lower levels of TNF-α, IL-6, IL-1β, and IL-17A, and elevated IL-10 levels at follow-up compared to non-responders. Multivariate logistic regression confirmed IL-6 and TNF-α as negative predictors, and IL-10 as a positive predictor of response. Machine learning models, particularly the random forest classifier, achieved high predictive accuracy (AUC = 0.90). IL-6 inhibitors yielded the highest treatment response rate (79.5%), while cytokine correlation analyses revealed biologically relevant interactions. The integration of serological markers such as RF with cytokine profiling further improved predictive power. These findings suggest that cytokine signatures can serve as robust biomarkers for tailoring RA treatment, enabling a shift toward personalized therapeutic strategies. Cytokine profiling not only predicts therapeutic efficacy but also offers insights into RA pathophysiology, potentially informing the design of novel biologics. This study supports the implementation of immunoprofiling as a routine component of RA management and calls for extended validation in larger, multicenter trials.