SAT0057 PREDICTING INADEQUATE RESPONSE TO JAK INHIBITORS BY CLUSTER ANALYSIS IN PATIENTS WITH RHEUMATOID ARTHRITIS

Background: Staphylococcus aureus is a frequent cause of bacteremia (SAB) associated with high mortality and morbidity. Patients with rheumatoid arthritis (RA) are at increased risk of septic arthritis and prosthetic joint infection with S. aureus. Objectives: To assess the incidence rate (IR) of first-time SAB in patients with RA and to estimate the incidence rate ratio (IRR) of SAB with a general population cohort without RA serving as the reference. Methods: Individuals with no prior history of SAB or RA were included consecutively from 31 December 1996, their 18th birthday or date of immigration, whichever came latest, and followed until first-time SAB, death, emigration or 31 December 2017, whichever came first. Information on RA diagnosis, vital status, age, sex, place of residence, comorbidities, medication and first-time SAB were achieved on an individual level through cross-linkage between five virtually complete Danish nationwide registries (Civil Registration System, National Patient Registry, Register of Medicinal Product Statistics, DANBIO rheumatology registry and the SAB database). We used Poisson regression to estimate adjusted IRRs overall and stratified by age and sex. Results: In total, 6,127,150 individuals were included of whom 34,627 individuals developed RA. In the RA cohort, 228 first-time SAB events occurred during 283,186 person years (PY) of follow-up (IR 80.5/100,000 PY) compared with 25,268 events during 87,521,120 PY of follow-up in the general population cohort (IR 28.9/100,000 PY). Median follow-up was 7.2 years (IQR 3.5-12.3) after RA diagnosis and 18.7 years (IQR 6.8-21) in the general population cohort. Individuals with RA who developed SAB were more often women, had an orthopaedic implant and had recent use of glucocorticoids compared with individuals with SAB without RA. (Table 1) IRs of SAB were higher among patients with RA compared with the general population in all age categories. The IRs increased with age and were higher in men, both in patients with RA and in the general population cohort. After adjustment, the IRR remained higher for individuals younger than 70 years with RA compared with the general population but was similar for older individuals. (Figure 1) Conclusion: In this nationwide cohort with more than 25,000 observed first-time SAB events, patients with RA younger than 70 years old had a 1.5-2 times higher incidence rate compared with the general population. The significance of anti-rheumatic treatments on risk and the prognosis of SAB in patients with RA remain to be explored. Acknowledgments: We wish to thank patient representative Pia Luchau Pedersen Disclosure of Interests: Sabine Dieperink: None declared, Bente Glintborg Grant/research support from: Grants from Pfizer, Biogen and Abbvie, Louise Bruun Oestergaard: None declared, Mette Norgaard: None declared, Thomas Benfield Grant/research support from: Pfizer, Novo Nordisk and GSK, Frank Mehnert: None declared, Andreas Petersen: None declared, Merete L. Hetland Grant/research support from: BMS, MSD, AbbVie, Roche, Novartis, Biogen and Pfizer, Consultant of: Eli Lilly, Speakers bureau: Orion Pharma, Biogen, Pfizer, CellTrion, Merck and Samsung Bioepis