FRI0143 Influence of low-dose glucocorticoid treatment on persistence on biologic dmards therapy: real-life data from the italian gisea registry

Background The use of glucocorticoid (GC) in Rheumatoid arthritis (RA) is recognised by the current treatment approach as a valid adjunct to DMARDs therapy. Despite its efficacy, safety of GC is still an issue and the best strategy of use is still debated, including patients under bDMARDs therapy Objectives To analyse in RA the differences of GC users versus non-users of baseline features, response to therapy and persistence in bDMARDs from the Italian biologics registry GISEA (Italian Group for the Study of Early Arthritis) Methods Consenting patients satisfying 1987 or 2010 ACR criteria for RA included in the Italian GISEA registry were enrolled. Data recorded comprehend demographic and clinimetric variables. Data are collected at baseline and 6 monthly during follow-up. To be included in the study patients needed a minimum follow-up time of 12 months and if data were not updated after 2012 patients were considered lost to follow-up. EULAR and HAQ responses were calculated. Statistical analysis included descriptive measures, parametric and nonparametric comparisons between groups and univariate analysis of survival on therapy Results A total of 8545 patients were enrolled, of them 4193 (49%) using a variable dose of GC. In 3035 (72%) the dose was ≤5 mg. Baseline demographic and disease-specific features at the start of bDMARD therapy were not different between GC users and non-users, both in 1 st and 2nd line bDMARDs RA patients. EULAR response rates were generally better in GC users at 6 and 12 months, but without statistical significance: good/moderate EULAR responses at 6 months were attained in 76,5% of GC users versus 67% in non users, while at 12 months in 81,5% vs 73% respectively (both Ps not significant). Similarly, HAQ responses ( Conclusions Our data show that GC are used in a high percentage of RA patients on bDMARD therapy. GC significantly improve the persistence on bDMARDs therapy in 1 st and 2nd line. No obvious other differences are evident in baseline, EULAR and HAQ response rates. This fact should be kept in mind when evaluating the persistence on bDMARD treatment reported in different registries. Safety evaluations in individual patients should be further analysed to guide the use of GC in this setting Disclosure of Interest None declared