LAUSR

Background The biologic DMARD abatacept has been associated with respiratory adverse events in patients with RA who have chronic obstructive pulmonary disease (COPD) in the ASSURE trial (NCT00048932), based on only 54 patients with RA and COPD.1 A large observational study of patients with RA and COPD, involving over 1,800 patients using abatacept did not find an increased incidence of respiratory adverse events with abatacept compared with other biologic DMARDs.2 It remains uncertain, however, whether this potential respiratory risk affects all biologic DMARDs, including abatacept, when compared with non-biologic DMARDs. Objectives To assess in a real-world observational setting whether patients with RA and COPD treated with biologic DMARDs, including abatacept, have an increased risk of serious respiratory adverse events compared with similar patients treated with non-biologic DMARDs. Methods The Truven MarketScan® Commercial and Supplemental Medicare databases were used to identify patients diagnosed with RA and COPD, treated with a biologic or non-biologic DMARD between January 2007 and December 2015. A prevalent new-user cohort design was used to match each new user of a biologic DMARD with a new user of a non-biologic DMARD on time-conditional propensity scores.3Patients were followed up from new use until the end of enrolment or 31 December 2015. The Cox model was used to estimate the hazard ratios (HRs) of respiratory adverse events associated with biologic DMARDs compared with non-biologic DMARDs, further adjusted for confounders found to be unbalanced despite matching on propensity scores. Results The study cohort included 7,424 new users of biologic DMARDs matched to 7,424 new users of non-biologic DMARDs, followed for up to 9 years. The adjusted HR (95% CI) of the combined respiratory endpoint, including severe COPD exacerbation, bronchitis and severe pneumonia or influenza, with biologic DMARD use relative to non-biologic DMARDs was 1.06 (0.91-1.24). For severe COPD exacerbation it was 0.88 (0.64-1.21), 1.02 (0.82-1.27) for bronchitis, while for pneumonia or influenza it was 1.18 (0.90-1.54) if hospitalized and 1.01 (0.89-1.14) as outpatient. For users of abatacept relative to non-biologic DMARDs, the HR of the combined respiratory endpoint was 1.06 (0.80-1.42). Results remained unchanged with sensitivity analyses. Conclusion In this large real-world study of patients with RA and COPD, the risk of pre-specified serious respiratory adverse events was not significantly increased in patients using biologic DMARDs, and specifically abatacept, compared with those using non-biologic DMARDs. This study does not support the safety signal for abatacept from the ASSURE trial. References [1] Weinblatt E, et al. Arthritis Rheum 2006;54:2807–16. [2] Suissa S, et al. Ann Rheum Dis 2018;77:613-614. [3] Suissa A, et al. Pharmacoepidemiol Drug Saf 2017;26:459–68. Disclosure of Interests Samy Suissa Grant/research support from: Advisory board meetings, or as speaker, or received research grants from AstraZeneca, Bayer, Boehringer-Ingelheim, Bristol-Myers Squibb, Novartis, Speakers bureau: Advisory board meetings, or as speaker, or received research grants from AstraZeneca, Bayer, Boehringer-Ingelheim, Bristol-Myers Squibb, Novartis, Marie Hudson Grant/research support from: Unrestricted research funds from Bristol-Myers Squibb, Pierre Ernst: None declared, Sophie Shen Employee of: Bristol-Myers Squibb, Teresa Simon Employee of: Bristol-Myers Squibb