FRI0290 FACTORS ASSOCIATED WITH PERSISTENCE OF GOLIMUMAB TREATMENT IN THE BIOBADASER REGISTRY, UP TO 7 YEARS OF FOLLOW-UP

Abstract

Background: Persistence in treatment with a biological drug can be considered an indirect measure of efficacy,safety and tolerability Objectives: We assessed the probability of persistence of golimumab treatment in patients with rheumatic diseases and the factors associated with persistence up to 7 years of follow-up Methods: BIOBADASER is the Spanish registry of biological drugs of the Spanish Society of Rheumatology and the Spanish Medicines Agency.A data-base analysis was done in Dec 2019 on all the patients aged 18 years or older who had initiated golimumab for rheumatoid arthritis (RA),psoriatic arthritis (PsA) and axial spondyloarthritis (SpA).The probability of persistence was calculated with a Kaplan-Meier test.Factors related to persistence were analyzed with a Cox-regression model Results: There were 685 patients included (195 [28.5%] RA, 294 [42.9%] axial SpA and 196 [28.6%] PsA) in the analysis (Mean age 51.3 [12.6] years, 53.7% women). Median duration of disease at the onset of golimumab therapy was 7.6 (3.0-14.5) years. Golimumab was prescribed as first biological drug in 36.4%, as second in 31.7% and as third or further in 31.9% of the patients. Concomitant medication at golimumab initiation included methotrexate (40.0%), corticosteroids (34.2%), leflunomide (17.8%) and sulphasalazine (8.1%). The probability of persistence with golimumab treatment was 71.7% (95% CI 68.1-74.9) at year 1,60.5% (56.5-64.2) at year 2, 55.6% (51.5-59.5) at year 3, 50.6% (46.2-54.8) at year 4, 45.1% (40.1-50.0) at year 5, 44.2% (39.0-49.3) at year 6 and 39.5% (32.8-46.2) at year 7. As the first biological agent the probability of persistence was 83.2% at year 1 and 60.0% at year 5. As a second biological drug persistence was 70.4% and 39.9% (year 1 and 7). Cox-regression analysis (table) showed that the probability of persistence with golimumab treatment was higher in first vs second or third biological line (Hazard Ratio[HR] for discontinuation: 1.45 for second and 3.04 for third or further vs first line), in SpA and PsA patients (HR discontinuation vs RA:0.56 and 0.49 respectively) and in patients with methotrexate (HR:0.61) and lower in those needing corticosteroids(HR:1.71) or DMARDs different to methotrexate (HR:1.88) and in patients with higher disease activity at golimumab onset (HR:1.45) Conclusion(s): The probability of persistence on golimumab therapy was high and remained relatively stable up to 7 years of follow-up. A lower risk of treatment discontinuation was observed in patients who received golimumab as first biological agent and in PsA and SpA patients. Persistence was lower in patients needing corticosteroids, DMARDs different to methotrexate and in those with higher disease activity at golimumab onset.