Serious infection risk of tofacitinib compared to biologics in patients with rheumatoid arthritis treated in routine clinical care

Abstract

Abstract Background: Comparative safety assessments are important for informed treatment decisions. Recently, serious infections related to tofacitinib (TOF) have raised considerable interest. This study aims to compare the risk for serious infections in patients with rheumatoid arthritis (RA) upon receiving TOF versus biologic disease-modifying antirheumatic drugs (bDMARDs) by age at treatment initiation. Methods: We identified adult RA patients exposed to TOF or bDMARDs using data collected by the Swiss registry for inflammatory rheumatic diseases (SCQM) from 2015 to 2018. The event of interest was the first non-fatal serious infection (SI) during drug exposure. Missing or incomplete event dates were imputed as either the lower (left) or upper (right) limit of the interval in which the event must have occurred. Separate analyses were done with the left and right imputed dates and interpreted together. The comparative risk (hazard ratio, HR) of TOF versus bDMARDs was estimated as a function of age using Cox proportional hazards regression adjusted for several possible confounders or predictors. Results: 1687 patients provided time at risk for a first SI during study participation and drug exposure for a total of 2238 different treatment courses, 345 for TOF and 1893 for bDMARDs. Exact dates were missing from 21% of the SIs, with 44 (left imputation) or 43 (right imputation), respectively, identified as first SIs (12/12 on TOF versus 32/31 on bDMARDs). Left and right imputation produced similar results. For ages >= 69 years, HRs were >1 with both imputations. Indicatively, for right imputation, at 69 and 76 years, the HRs (95% confidence intervals) were, respectively: 2.05 (1.04, 4.05; increased), and 2.87 (1.27, 6.52; clinically relevantly increased). For ages < 65 years, the data were insufficient to draw conclusions one way or another. Conclusions: Our results suggest that we should expect an increased risk for SIs upon treatment with TOF compared to bDMARDs in older patients (aged ≥ 69 years). Our study thus supports a cautious use of TOF in these patients. Further research is needed to assess whether the comparative risk is age dependent and, if so, whether it is relevantly increased in older as opposed to younger patients.