Real-world data back CV safety of JAK inhibitors

Two real-world studies presented at EULAR 2023 reflected the cardiovascular (CV) safety of Janus kinase inhibitors (JAKi*) in patients with rheumatoid arthritis (RA).

“Concerns have been raised about the CV safety of JAKi,” said Romain Aymon from the Geneva University Hospital, Switzerland, who presented findings from the JAK-POT study.

“The ORAL Surveillance study showed an increased risk of CV events in patients at risk. The results thus prompted regulatory authorities to issue warnings and recommendations about the use of JAKi. However, we do not have robust real-world evidence to support the findings,” Aymon continued.

JAK-POT study

JAK-POT used data from an international collaboration of RA treatment registries to evaluate the incidence of MACE** (primary outcome) in RA patients on JAKi, TNFi***, or bDMARD-OMA^#. A total of 51K treatment courses were recorded, a quarter of which were JAKi. Secondary outcomes included the individual components of MACE plus deep venous thrombosis and pulmonary embolism (PE). [EULAR 2023, abstract OP0219]

A total of 186 MACE was recorded, translating to an incidence rate (IR) of 2.06 events/1,000 person-years (PY). Most of these events were from the TNFi group (n=84, 65, and 37 for TNFi, bDMARD-OMA, and JAKi, respectively).

IR/1,000 PY was lower with JAKi than TNFi (1.73 vs 1.94). Compared with TNFi, crude incidence rate ratio [IRR] with JAKi was 0.89, which persisted in the adjusted analysis (adjusted IRR [aIRR], 0.74). This trend was similarly seen in a cohort of patients aged ≥50 years with ≥1 CV risk factor (aIRR, 0.71).

Looking at the secondary outcomes, apart from PE, there was no difference across all other outcomes (aIRRs ranging between 0.68 and 1.06). “There appeared to be a higher incidence of PE with JAKi than with TNFi (aIRR, 1.5),” said Aymon.

“[Overall, although] registers tend to underreport the number of MACE, we did not find a significantly higher risk of MACE and other CV outcomes in patients treated with JAKi compared with TNFi,” said Aymon.

Dutch cohort

A retrospective inception study also evaluated the CV safety of JAKi in a large Dutch RA cohort (n=15,191; median age 62 years, 72 percent female). The investigators used a nationally representative prescription database covering about 2/3 of all outpatient prescriptions in the Netherlands. [EULAR 2023, abstract OP0221]

“We defined outcomes as a new prescription of a platelet aggregation inhibitor^## and used this as a proxy for a new CV event,” said Dr Merel Opdam from Sint Maartenskliniek, Ubbergen, Netherlands, at EULAR 2023.

After a median follow-up of 1.8 years, >20K treatment episodes were reported, mostly with bDMARD (>18K).

In the overall cohort, events/100 PY between the JAKi and bDMARD arms were similar (1.52 vs 1.47; aIRR, 1.01). A similar trend was seen when comparing tofacitinib and baricitinib against bDMARD (aIRRs, 1.02 and 1.04 for tofacitinib and baricitinib, respectively).

“[However,] there is a risk of over and underestimation of CV events due to the proxy we used. For instance, there may be some overestimation as platelet aggregation inhibitors are also used for peripheral artery disease, which is usually not included in the MACE definition,” Opdam noted.

“[Nonetheless, our findings show that] in the general RA population, JAKi do not seem to be associated with an increased CV risk, but caution is required in certain subgroups such as older patients or those with CV risk factors or pre-existing CV disease,” she continued. [www.ema.europa.eu/en/news/ema-recommends-measures-minimise-risk-serious-side-effects-janus-kinase-inhibitors-chronic, accessed June 20, 2023]