AN EVALUATION OF FACTORS CONTRIBUTING TO PRIMARY NONRESPONSE TO BIOLOGICAL THERAPY IN PATIENTS WITH SPONDYLOARTHROPATHY: A RETROSPECTIVE COHORT STUDY

Emel Oğuz KÖKOĞLU, Hüseyin KAPLAN, Melih KIZILTEPE, Tuğba Kahraman DENİZHAN, Celil Barlas CENGİZ, Abdurrahman Soner ŞENEL

Archives of Rheumatology - 2026;41(3):201-211

Division of Rheumatology, Department of Internal Medicine, Kayseri City Education and Research Hospital, Kayseri, Türkiye

 

Background/Aims: Primary nonresponse to biological therapy in patients with spondyloarthropathy (SpA) is a considerable challenge. This study sought to identify the factors associated with primary nonresponse to biological therapy in a cohort of patients diagnosed with SpA. Materials and Methods: A retrospective analysis was performed on 261 patients with SpA who underwent at least 1 biological therapy. The incidence of primary nonresponse to any biological treatment was evaluated, and various clinical and demographic factors were analyzed for their potential associations with treatment outcomes. Results: The study found that the frequency of primary nonresponse to any biological disease-modifying antirheumatic drugs (bDMARDs) was 15.3% in the study population. In the logistic regression analysis of baseline parameters, presentation with arthritis at diagnosis was identified as a significant predictor of primary nonresponse (odds ratio: 2.113, 95% CI: 1.069-4.175, P = .031). Beyond this baseline predictor, primary non-responders significantly differed from responders by having higher frequencies of concomitant fibromyalgia (FMS) treatment (P = .047) and higher terminal C-reactive protein (CRP) levels (P = .030), as well as variations in acute-phase reactants, presence of psoriasis or enthesitis at diagnosis, and treatment-related factors such as bDMARDs duration. Conclusion: This study identifies baseline arthritis as a significant predictor of primary nonresponse to bDMARDs in SpA patients. Additionally, the presence of FMS and elevated CRP levels during follow-up are key clinical characteristics associated with the non-responder phenotype. These findings highlight the need for more objective assessment tools to distinguish true biological treatment failure from comorbid pain syndromes and the influence of disease phenotypes.