(216) Comparative safety of baricitinib versus TNFi in Rheumatoid Arthritis management in the French population: a propensity score-matched cohort study
Scientific Officer Bordeaux PharmacoEpi, INSERM CIC-P1401, Univ. Bordeaux, 33000 Bordeaux, France Bordeaux, France
Background: An international post-authorization safety study using real-world data was set up to assess the safety profile of baricitinib for the treatment of rheumatoid arthritis (RA). Results issued from the French nationwide healthcare database (SNDS) are presented here.
Objectives: To compare the risk of venous thromboembolism (VTE), major adverse cardiovascular events (MACE), and serious infection (SI) between RA patients treated with baricitinib and similar patients treated with tumor necrosis factor inhibitors (TNFi).
Methods: A comparative cohort study was conducted in the SNDS including all French patients aged 18 years or older with RA initiating baricitinib or TNFi between 2017 and 2019. A cohort was created for each event of interest: VTE, MACE, and SI, excluding patients with a history of the considered event. For each cohort, one propensity score model was generated, relying on history of relevant comorbidities, previous use of conventional RA treatments, and advanced treatment lines. New users of baricitinib were then 1:1 matched on propensity score with patients starting a new TNFi line. The incidence rate and 95% confidence interval (95%CI) of each outcome were estimated in the baricitinib- and the TNFi-matched cohorts. The incidence rate ratios (IRRs) and 95%CI were estimated using a modified Poisson regression model.
Results: Of 3 242 eligible baricitinib-treated patients, 2 859 were matched with TNFi-treated patients, contributing respectively to 1 855 and 1 923 person-years (PY) of baricitinib and TNFi exposure (VTE cohort, the distribution was similar for MACE and SI cohorts). During follow-up, 33 patients experienced VTE (20 with baricitinib), 36 experienced MACE (25 baricitinib), and 72 experienced SI (36 baricitinib). Incidence rates in baricitinib and TNFi-treated patients were, respectively: 1.1 (95%CI [0.7 to 1.7]) and 0.7 [0.4 to 1.2] VTE per 100 PY, 1.4 [0.9 to 2.0] and 0.6 [0.3 to 1.0] MACE per 100 PY, and 1.9 [1.3 to 2.6] and 1.8 [1.3 to 2.5] SI per 100 PY. Overall IRRs comparing baricitinib vs TNFi treatment were 1.59 [0.79 to 3.21] for VTE, 2.33 [1.15 to 4.74] for MACE, and 1.04 [0.65 to 1.65] for SI.
Conclusions: French results suggested a significative increased risk of MACE related to baricitinib use in patients with RA compared to patients with TNFi. Baricitinib has been available in France since 2017, thus, these analyses are only based on the first years of marketing where treated patients are often refractory to previous lines of treatment. These results must be interpreted in the light of those from the other participating countries to build a robust baricitinib safety profile.