Risk of Cardiovascular Events, Malignancy, and Serious Infection Associated With JAK inhibitors and TNF inhibitors in Rheumatoid Arthritis or Ulcerative Colitis: A Nationwide Cohort Study

Background: The ORAL surveillance trial, a prospective post-market surveillance study, unveiled the risks of cardiovascular (CV) events, malignancies, and serious infections in patients receiving Janus kinase inhibitors (JAKi) for rheumatoid arthritis (RA). However, there is also a need to establish the safety of JAKis in patients with ulcerative colitis (UC) that may be eligible for treatment with JAKis in the real-world setting.

Objectives: To assess the risks of CV events, malignancy and serious infection associated with JAKis, compared with tumor necrosis factor inhibitors (TNFis) in patients with either RA or UC.

Methods: Using the National Health Information Database (NHID) of South Korea (2013 – 2023), we constructed two separate cohorts comprising patients with RA and UC, respectively, and compared the safety of JAKi and TNFi within each cohort. Patients were followed up from the date of treatment initiation until outcome occurrence, treatment discontinuation, treatment class switch, end of study period, or death. Outcomes of interest were 3-point major adverse CV events (3P-MACE: CV death, nonfatal myocardial infarction, nonfatal stroke), malignancy and serious infection. For malignancy, a 1-year lag period was applied to account for the delayed onset of cancer in nature. Serious infections were defined as any infection that resulted in hospitalization. We performed propensity score overlap weighting to balance on pre-defined covariates. Cox proportional hazards models were used to estimate the HR and the corresponding 95% CI for the outcomes. Subgroup analyses were performed according to sex, eligibility for the ORAL Surveillance trial, history of cardiovascular disease, duration (≥ 18, < 18 months), and each ingredient of JAKi.

Results: The RA cohort consisted of 14,972 patients (mean age 54.7, 79.0% female), with 4,759 receiving JAKi, while the UC cohort included 2,085 patients (mean age 41.7, 28.4% female), with 347 receiving JAKi. In the RA cohort, the weighted HR was 0.92 (0.59 - 1.42) for 3P-MACE, 1.61 (1.08 – 2.41) for malignancy, and 1.08 (0.94 – 1.23) for serious infection. When restricting the cohort to those eligible for the ORAL Surveillance trial, the HR for 3P-MACE was 0.92 (0.52 - 1.64).
In the UC cohort, the weighted HR was, 0.98 (0.11 – 8.42) for malignancy, and 0.45 (0.26 – 0.78) for serious infection; the risk of 3P-MACE was not computed due to insufficient events.

Conclusions: While JAKis were found to be associated with an elevated risk of malignancy, no significant difference in the risk of CV events and serious infection was observed in patients with RA. Further, we were unable to identify any safety risks associated with JAKi use in patients with UC.