Pos1085 initiation of a janus kinase inhibitor before and after the safety warnings: changes in characteristics of patients with rheumatoid arthritis

Background In 2019, 2020 and 2021, the European and US-American regulatory agencies issued warnings about venous thromboembolism, major cardiovascular events and malignancy risks associated with the Janus kinase inhibitor (JAKi) tofacitinib and required changes in labelling. Objectives To investigate whether characteristics of patients with rheumatoid arthritis receiving a JAKi versus biologic therapy differed before and after the safety warnings. Methods Data from patients who started with any JAKi or biologics treatment in the German biologics register RABBIT between 01/2017 and 04/2022 were included. Multivariate logistic regression analyses were used to understand differences in characteristics of patients starting a JAKi treatment versus a biologic treatment in three annual cohorts: 2017 the year JAKis became available in Germany, 2019 before the EMA safety warnings and 2021. In each year, only the first treatment episodes of each JAKi, TNF inhibitor, interleukin-6 inhibitor or B/T-cell targeted therapy were considered. Prior treatment episodes were possible. The logistic regressions were corrected for clustering at the patient level. In 2017, we included 549 JAKi versus 2510 bDMARD treatment episodes, in 2019 674 versus 2233 and in 2021 700 versus 1296 episodes. Results Patient characteristics at treatment start have changed over time. In 2017, compared to patients receiving a biologic, patients starting a JAKi had been treated with a higher number of therapies, had a higher (worse) physician reported heath and were more likely to have comorbidities such as hypertension, coronary heart disease, diabetes, hyperlipoproteinaemia, thrombosis, malignancy or lymphoma. In 2019, patients initiating a JAKi therapy compared to a biologic were less likely to be women, had a worse physician reported health and were more likely to receive a dose of less than 10mg glucocorticoids than none. In 2021, after the safety warnings, compared to patients receiving a biologic, those who started a JAKi were older, had a worse physician reported health, had received a higher number of previous therapies, had poorer self-reported health and were less likely to receive a high dose of glucocorticoids. Although not significant, patients with comorbidities were less likely to receive a JAKi. Conclusion The analyses show that after the launch of JAKi treatment in 2017, comorbidities increased the likelihood to receive JAKis as a new treatment option. In 2021, patients with a high disease burden and with many other previous therapies were more likely to receive JAKis, but not those with comorbidities. This development shows that rheumatologists in Germany follow the safety recommendations and consider the patients’ disease burden and risk factors when prescribing JAKis. Table 1. Odds Ratios of the logistic regressions for treatment start of a JAKi compared to a bDMARD in 2017, 2019, 2021 Year of treatment start 2017 2019 2021 OR LogReg (95%CI) intercept 0.10 0.25 0.24 Ref: age <50 50-65 1.00 (0.77-1.31) 1.27 (0.99-1.64) 1.50 (1.14-1.97) 65+ 0.81 (0.60-1.10) 1.15 (0.86-1.53) 1.51 (1.11-2.05) sex (m vs. f) 0.96 (0.76-1.20) 0.74 (0.60-0.91) 1.03 (0.83-1.26) Relevant comorbidities* 1.37 (1.10-1.70) 1.16 (0.96-1.40) 0.86 (0.70-1.06) Seropositivity 0.95 (0.76-1.19) 0.98 (0.13-1.20) 1.16 (0.93-1.45) disease duration 0.99 (0.98-1.00) 0.99 (0.98-1.00) 1.00 (0.99-1.01) #previous b/tsDMARDs 1.37 (1.30-1.44) 1.02 (0.98-1.07) 1.18 (1.12-1.25) physician global health (0-10) 1.16 (1.10-1.22) 1.09 (1.04-1.14) 1.08 (1.03-1.13) patient global health (0-10) 0.95 (0.89-1.01) 1.02 (0.97-1.08) 0.95 (0.90-1.00) % of full physical function 1.00 (0.99-1.00) 1.00 (0.99-1.00) 1.00 (0.99-1.00) Ref: 0 mg/d glucocorticoids >0-<10 mg/d 0.89 (0.70-1.12) 1.27 (1.05-1.54) 1.01 (0.83-1.24) >=10 mg/d 0.75 (0.54-1.03) 1.01 (0.75-1.37) 0.70 (0.50-0.98) *≥ 1 comorbidity mentioned in the safety warnings (hypertension, coronary heart disease, diabetes, hyperlipoproteinaemia, thrombosis, malignancy, lymphoma) Acknowledgements RABBIT is currently supported by a joint, unconditional grant from AbbVie, Amgen, BMS, Celltrion, Fresenius Kabi, Galapagos, Hexal, Lilly, MSD, Pfizer, Samsung Bioepis, Sanofi, VIATRIS SANTE and UCB and previously by Roche. Disclosure of Interests Doreen Huschek Grant/research support from: Non-personal, joint grant from a consortium of 14 pharmaceutical companies for the biologics register RABBIT to my institute., Peter Herzer Speakers bureau: ABBVIE, NOVARTIS, JANSSEN-CILAG, Angela Zink Grant/research support from: Previously, but not during last three years, Martin Feuchtenberger Speakers bureau: Martin Feuchtenberger reports fees from AbbVie, personal fees from Novartis, personal fees from Roche, and personal fees from UCB outside of the submitted work., Consultant of: Martin Feuchtenberger reports fees from AbbVie outside of the submitted work., Anja Strangfeld Speakers bureau: AbbVie, Amgen, BMS, Celltrion, Janssen, Lilly, Pfizer, Roche, Sanofi, UCB. Non-personal, joint grant from a consortium of 14 pharmaceutical companies for the biologics register RABBIT to my institute., Grant/research support from: Non-personal, joint grant from a consortium of 14 pharmaceutical companies for the biologics register RABBIT to my institute.