Background/Purpose: Recent concern has been raised for a risk for venous thromboembolism (VTE) associated with janus kinase inhibitors among patients with RA who are already known to be at increased risk for VTE based on their underlying conditions. We evaluated the risk of VTE, defined as the composite of pulmonary embolism (PE) or deep vein thrombosis (DVT) among RA patients initiating tofacitinib (Tofa) compared to patients initiating adalimumab (ADA).

Methods: Using 2010-2015 US Marketscan claims data, we conducted a retrospective cohort study among RA patients (identified based on ≥ 1 ICD-9 RA diagnosis codes initiating Tofa or ADA). Patients were not allowed to have used Tofa or ADA prior to the date of initiation (index date) using all available data.  Eligible patients must have had ≥ 12 months (baseline) prior health plan enrollment. Patients with history of other autoimmune disease, advanced kidney or liver disease, malignancy, HIV, hepatitis B/C, or pregnancy during baseline were excluded. Follow up started from the index date and ended at the earliest date of: VTE, death, loss of medical/pharmacy coverage, switch to other biologic, or drug discontinuation (with a 90 day extension). PE was identified using inpatient diagnosis codes plus outpatient anti-coagulant treatment within 0-60 days; DVT was identified using inpatient/outpatient diagnosis codes and medication use as above. We calculated incidence rates per 100 patient-years of VTE for each drug and compared VTE risks during follow-up using Cox regression, adjusting for potential confounders including age, gender, history of VTE, comorbidities and concurrent medications. Several sensitivity analyses were conducted using different exclusion criteria and outcome definitions (such as using only inpatient diagnosis and anti-coagulant use within 60 days to define DVT occurrence).

Results: We identified 6,022 ADA initiators (4,798 PY) and 2,155 for Tofa (1,523 PY). During median (IQR) follow up of 0.5 (0.3, 1.0) years, we identified 20 VTE events among tofa users with an IR of 1.31 (95% CI: 0.80-2.03) and 40 VTE events among ADA users with an IR of 0.83 (95% CI: 0.60-1.14). After adjustment for potential cofounders, the hazard ratio of VTE for Tofa users was 1.07 (95% CI: 0.54-2.14) referent to ADA users. Sensitivity analyses using different exclusion criteria and outcome definitions had a large impact on the number of events and absolute incidence rates, but minimal effect on the adjusted hazard ratios comparing the two therapies. For an example, using an outcome definition of inpatient VTE events with anti-coagulation within 60 days, the IR rate associated with tofacitinib was 0.65/100PY (95% CI: 0.31, 1.20) with similar adjusted HR to the main analysis (1.02, 95% CI 0.40-2.62).

Conclusion: In this comparison of tofacitinib vs. adalimumab based on 60 VTE events, we observed comparable risk between these two RA treatments.