FRI0118 INFECTION OUTCOMES IN PATIENTS WITH RA TREATED WITH ABATACEPT AND OTHER DMARDS: RESULTS FROM A 10-YEAR INTERNATIONAL POST-APPROVAL STUDY

Abstract

Background The abatacept global post-marketing epidemiology programme consists of observational studies based on biologic disease registries and healthcare claims databases to assess infection and malignancy risks associated with abatacept treatment, as used in routine clinical practice. Objectives To evaluate the risk of infection in patients with RA treated with abatacept versus conventional synthetic (cs)DMARDs and other biologic (b) or targeted synthetic (ts)DMARDs. Methods Data were analysed from five cohorts: two biologic registries (the Anti-Rheumatic Therapy in Sweden register and the Rheumatoid Arthritis Observation of Biologic Therapy German registry), a disease registry (FORWARD, The National Databank for Rheumatic Diseases in the USA) and two healthcare claims databases (the population-based British Columbia Canadian RA Cohort and the US Optum Research Database). Crude incidence rates (per 1000 patient-years of exposure) with 95% CIs were calculated for hospitalised infections and opportunistic infections. Adjusted risk ratios (RRs) with 95% CIs were estimated using multivariate models adjusting for demographics, co-morbidities and other potential cofounders within each database and were subsequently pooled using a random effects model for meta-analyses.1 Results From all cohorts, patients treated with abatacept (∼6400), csDMARDs (∼137K) and other b/tsDMARDs (∼54K) were followed up for a mean of 2.3–3.7, 2.2–6.2 and 2.3–4.7 years, respectively. Patients were mainly female (71–86%), with a mean age ranging from 49–63 years, and 3–18% had a history of prior severe infections across treatment groups/cohorts. A greater number of abatacept-treated patients had a history of ≥2 prior biologics (44–85% vs csDMARDs, 11% [FORWARD] and other b/tsDMARDs, 0–19%). In abatacept-treated patients, incidence rates for hospitalised infections ranged from 16–56 and opportunistic infections from 0.4–7.8 (Table). Adjusted RRs (95% CIs) for abatacept vs csDMARDs (range: 0.3 [0.2, 0.7] to 2.2 [1.3, 3.7]; pooled estimate: 1.2 [0.6, 2.2]) and abatacept vs other b/tsDMARDs (range: 0.5 [0.3, 0.8] to 1.3 [0.8, 2.1]; pooled estimate: 0.9 [0.6, 1.3]) showed no increased risk in hospitalised infections. Conclusion In this large, international post-marketing epidemiology programme, pre-specified infection risks were more frequent with abatacept than csDMARDs and b/tsDMARDs in some, but not all, cohorts. Increased risks for individual outcomes/cohorts may be attributed to prior biologic exposure, study design and practice pattern differences. These data are consistent with the current safety profile of abatacept.