Background Approximately one-third of patients with ankylosing spondylitis (AS) treated with tumor necrosis factor inhibitors (TNFi) need to switch their biological therapy. However, a significant clinical response only occurs in half of switcher patients.
Objectives To determine predictive factors of treatment response at 12 weeks in AS patients switching treatment to a second TNFi.
Methods Patients with AS (1984 modified New York classification criteria) switching treatment to a second TNFi were identified in the Rheumatic Diseases Portuguese Register, Reuma.pt. Data was extracted at baseline (at second TNFi start) and at 12 weeks of follow-up. Variables included in the analysis were age, race, gender, BMI, education level, work status, tobacco and alcohol consumption, disease duration, time to diagnosis, peripheral arthritis, extra-articular manifestations, ESR, CRP, HLA-B27 positivity, previous DMARD and steroid therapy, BASDAI, ASDAS, BASFI, BASMI, first TNFi used, reason and time to switch. Univariable logistic regression analysis of baseline predictors of BASDAI response (improvement ≥2 units or ≥50%) was performed and variables with a p-value<0.25 were re-tested in multivariable regression models.
Results Of the 334 AS patients treated with biologicals, 85 (25.4%) switched to a second TNFi. Reasons for switch included treatment failure (16; 18.8%), adverse events (25; 29.4%) and other reasons (44; 51.8%) like refractory extra-articular manifestations, surgery or patient preferences. Patients were included in the analysis if they stayed in therapy at least for 3 months and have BASDAI at baseline and at 3 months available. Forty-nine patients were analyzed, 22 responders (44.9%) and 27 non-responders (55.1%). BASDAI clinical response at 12 weeks was predicted by BASMI (p=0.04; OR 3.8 [95% CI 1.0, 14.1]) adjusting for gender, erythrocyte sedimentation rate and age at biological treatment onset. No other statistical significant demographical, clinical or laboratorial predictors were found.
Conclusions We found 44.9% switchers who responded to the second TNFi. An increased BASMI score was an independent predictor of BASDAI clinical response to a second TNFi in AS patients. The inclusion of a greater number of patients in future studies may allow the determination of further predictive factors.
Disclosure of Interest None declared