Janus kinase (JAK) inhibitors are a class of medications that bind intracellularly to the adenosine triphosphate binding site on JAK and suppress downstream cytokine effects by preventing phosphorylation and activation of signal transducers and activators (STAT) and subsequent gene transcription. They have been shown to benefit patients with some hematologic and autoimmune diseases. Unlike when used for hematologic disorders, their use in patients with autoimmune diseases, especially rheumatoid arthritis, has come with concerns for increased risk of thromboembolism, particularly major adverse cardiac events (MACE) and venous thromboembolism (VTE). As a class, the JAK inhibitors are known to cause elevations in total cholesterol, triglycerides, and high- and low-density lipoprotein. Baricitinib had a black box warning in 2017 for increased risk of VTE, as did tofacitinib in 2019. Due in part to the concern about MACE and VTE, an additional long-term safety trial was mandated by the U.S. Food and Drug Administration (FDA).
Dr. Steven Ytterberg and colleagues from the ORAL Surveillance Study recently reported their findings in the New England Journal of Medicine. In this open-label, phase IIIb-IV noninferiority safety endpoint study of patients with active rheumatoid arthritis despite methotrexate treatment, patients were randomly assigned to tofacitinib 5 mg twice daily, tofacitinib 10 mg twice daily, or a tumor necrosis factor (TNF) inhibitor (either etanercept or tocilizumab). To be included in the study, patients needed to be older than 50 years and have one additional cardiovascular risk factor (smoking, hypertension, low HDL, diabetes mellitus, family history of coronary artery disease, personal history of coronary artery disease, or extra-articular rheumatoid arthritis). The study was powered for MACE (and cancer) outcomes with plans to enroll 4,000 patients, observing 1,500 patients for at least three years. The prespecified hazard ratio (HR) upper CI limit for noninferiority was less than 1.8.
Before study completion, the data safety monitoring board recommended termination of the arm with tofacitinib 10 mg twice daily due to an increased risk of VTE, more specifically pulmonary emboli (PE), and these patients were dose reduced to 5 mg twice daily. At trial completion, the group randomized to tofacitinib 10 mg twice daily at study onset were found to have a significantly higher risk of PE (HR, 8.26; 95% CI 2.49-27.43) and overall VTE (HR, 3.52; 95% CI 1.74-7.12) when compared to the TNF inhibitor group. An increased, but not statistically significant, HR was also seen with tofacitinib 5 mg dosing (HR for PE, 2.93; HR for VTE, 1.66) compared to the TNF inhibitor group. For the primary safety outcome of MACE, the incidence rates were 0.91 with the 5 mg dose, 1.05 with the 10 mg dose, and 0.73 (per 100 patient-years) with the TNF inhibitors. The HR for the 5 mg dose and 10 mg dose was 1.24 (95% CI, 0.81-1.91) and 1.43 (95% CI, 0.94-2.18), respectively, compared to the TNF inhibitors; neither dose meet the noninferiority margin (upper CI <1.8). The study also demonstrated a higher risk with tofacitinib at both doses for malignancies (especially non-melanoma skin cancer) and a higher risk for death from any cause with the 10 mg dose.
In Brief
This study has led to an updated and somewhat broad FDA black-boxed warning for tofacitinib, baricitinib, and upadacitinib for MACE, malignancy, and VTE, while simultaneously limiting the approved indication to those patients with rheumatoid arthritis, psoriatic arthritis, polyarticular juvenile idiopathic arthritis, and ankylosing spondyloarthritis or ulcerative colitis who have had inadequate response or intolerance to one or more TNF inhibitors. As it relates to the thromboembolic outcomes of this trial, it is important to interpret these findings as they relate to the comparator, TNF inhibitor therapy, rather than an inherent feature of these drugs. It is noteworthy that the overall incidence rates of MACE in all arms of the study were in fact somewhat lower than what has been observed in other studies of patients with rheumatoid arthritis. In a meta-analysis of JAK inhibitors comparing the drugs to placebo, there was a reduced, but nonsignificant, risk ratio (0.68; 95% CI, 0.36-1.29).1 Interestingly, baricitinib and tofacitinib have both now been studied in randomized trials of hospitalized patients with SARS-CoV-2, a population known to be at a very high risk of both venous and arterial thromboembolism, and neither has demonstrated concerning safety signals for VTE or MACE2–4 with short-term use. An equally valid conclusion and perhaps more accurate conclusion from this study, knowing that these diseases themselves contribute to increased risk of thromboembolism, would be that TNF inhibitors seem to reduce MACE and VTE better than JAK inhibitors in patients with rheumatoid arthritis. The increased risk for cancer remains a significant concern, but given the ease of administration (oral) with JAK inhibitors, this will certainly remain an active area of investigation, with some speculation as to whether more selective JAK1 inhibitors (upadacitinib and filgotinib) produce better results.
Competing Interests
Dr. Houghton indicated no relevant conflicts of interest.
