Background/Objectives: Response to TNF inhibitors in RA remains heterogeneous and reliable predictors of treatment response are still lacking. Biomarker-based stratification may improve therapeutic decision-making and aligns with the emerging paradigm of precision medicine. Methods: We conducted a prospective observational study including 64 biologic-naïve patients with active RA being inadequately controlled by csDMARDs. All patients initiated anti-TNF therapy and were followed for 12 months. Clinical response was assessed at 6 and 12 months using EULAR response criteria based on DAS28-CRP. Baseline serum levels of classical biomarkers (RF type IgM, RF type IgA, anti-CCP) and additional biomarkers (anti-MCV,14-3-3η protein, COMP) were evaluated. Logistic regression analyses were performed to identify predictors of treatment response. Results: At 6 months, 7 patients were classified as non-responders, 38 as moderate responders, and 19 as good responders Lower baseline levels of RF isotypes, anti-CCP antibodies, 14-3-3η protein, and COMP were associated with favorable clinical response at 6 months. Baseline anti-CCP and 14-3-3η protein levels emerged as significant predictors in univariate analysis. Multivariate logistic regression yielded a predictive model incorporating anti-CCP, 14-3-3η protein, and COMP, achieving an overall prediction accuracy of 89.1%. At 12 months, baseline RF isotypes remained associated with treatment response, whereas the predictive value of other biomarkers diminished. Longitudinal analysis demonstrated significant reductions mainly for classical autoantibody levels under anti-TNF α inhibitors. Conclusions: A combined serum biomarker panel may support early prediction of response to anti-TNF therapy in RA. These findings highlight the potential of integrated biomarker-based stratification to optimize therapeutic decisions and support the implementation of precision medicine approaches in RA.
Gavrilă et al. (Tue,) studied this question.