Risk of non-bacterial opportunistic infections associated with janus kinase inhibitors versus tumor necrosis factor inhibitors among rheumatoid arthritis patients: A cohort study

Objective: To compare the risk of serious opportunistic infections between janus kinase inhibitors (JAKis) versus tumor necrosis factor inhibitors (TNFis) among rheumatoid arthritis (RA) patients. Methods: Using 2009–2020 Korea National Health Insurance Service database, we conducted a cohort study on RA patients initiating a JAKi or TNFi. The primary outcome was a composite of hospitalized viral, fungal, and tuberculous infections. Secondary outcomes were individual components of the primary outcome. Propensity-score fine stratification (PSS) and weighting were applied to control confounding. Hazard ratios (HRs) and 95% confidence intervals (CIs) were estimated using Cox proportional hazard models. Subgroup analyses by age, cumulative corticosteroid dose, and comorbidity score were done. Results: During a mean follow-up of 479 days, PSS-weighted 4252 JAKi initiators and 6653 TNFi initiators generated 173 cases of serious opportunistic infections, of which incidence rate was 1.66 and 0.87 per 100 person-years in JAKi and TNFi users, respectively, with the PSS-weighted HR (95% CI) of 1.94 (1.44–2.64). The PSS-weighted HR (95% CI) for secondary outcomes was 2.89 (1.98–4.20) for viral, 2.07 (0.59–7.26) for fungal, 0.55 (0.27–1.15) for tuberculosis. Results were consistent across subgroups. Conclusions: This population-based cohort study on RA patients found that the overall risk of serious opportunistic infections was higher with JAKi than TNFi. However, individual types of infections showed different patterns in that the risk of viral infections (and fungal infections only numerically) was higher among JAKi initiators, while that of tuberculosis tended to be lower in TNFi initiators.