What Do We Know About the Safety of JAK Inhibitors for Psoriatic Arthritis?

Janus kinase inhibitors (JAKis) form a class of small-molecule therapies for autoimmune disorders, such as psoriatic arthritis (PsA) and rheumatoid arthritis (RA). JAKis inhibit an array of intracellular processes related to immunity that depend on the JAK-signal transducer and activator of transcription (JAK-STAT) pathway. JAKis differ from biologics in their relatively broad mechanism of action and oral administration. Clinicians may consider prescribing JAKis for patients with an inadequate response to 1 or more biologics.¹

The JAKi tofacitinib was approved in 2012 for the treatment of RA, and in 2017 for PsA that was nonresponsive to methotrexate or other disease-modifying antirheumatic drugs (DMARDs).¹ Following the approval of JAKis for RA, retrospective studies have suggested that tofacitinib may increase serum lipid levels.² In addition, uncertainty remains regarding estimates of malignancy incidence with tofacitinib use in patients with RA.³ Therefore, the US Food and Drug Administration (FDA) has mandated a phase 3b/4 safety trial to clarify whether risks for cardiovascular adverse events and malignancy were increased in this patient population.

In 2021, the authors of the ORAL Surveillance study⁴ reported that tofacitinib failed to show noninferiority compared with tumor necrosis factor inhibitor (TNFi) adalimumab for major adverse cardiovascular events (MACE) and cancer. Venous thromboembolic events (VTE) were found to be numerically but not statistically more common in patients receiving 5 mg tofacitinib twice-daily than in those receiving adalimumab.

The study results prompted the FDA’s decision to expand a boxed warning to include these risks. However, results of the ORAL Surveillance trial included limitations in that the findings were not generalizable to other cohorts with RA⁵; MACE risk was not higher in subsequent analyses of large claims and registry datasets; and VTE risk was higher in only some real-world cohorts.

Because limited evidence is available for JAKi use among patients with PsA, outstanding questions remain.

• How should clinicians interpret the available evidence regarding adverse events?
• Are risks for MACE, VTE, malignancy, and death in patients with PsA who receive JAKi differ from those in patients with RA?
• Do the risks differ across the different JAKis?
• What patient-specific factors should clinicians consider, and how should clinicians and patients with PsA discuss issues specific to JAKis?

Interpreting the Current Evidence on JAKis

Insight into the mechanisms for elevated risk for MACE and VTE is lacking.⁶

Brittany Weber, MD, PhD, a cardiologist at the Brigham and Women’s Hospital, Boston, and director of the Cardio-Rheumatology Clinic, emphasized that “we will need further research to understand the mechanism underlying any increased [risks for] MACE and VTE. In the meantime, we have to work with the available evidence.”

She also noted, “From the currently available evidence, the signal to harm has mostly been observed in those already at elevated cardiovascular risk⁷; so a key point is to identify and manage this risk in particular.”

Steven Vlad, MD, PhD, a rheumatologist at Tufts Medical Center, assistant professor at Tufts University School of Medicine, Boston, and principal investigator of the Tufts Spondyloarthritis Registry, drew further attention to the features of the ORAL Surveillance study. He added to Dr Weber’s comment by saying, “We have to remember that the trial was enriched for people already at risk for MACE. Those with no known cardiac or stroke risk factors are probably at very low risk — though we don’t truly know — of seeing a side effect that could be attributed to a JAKi, such as tofacitinib. However, there was no enrichment for cancer outcomes. That suggests there is true risk even in those without cancer currently.”

Dr Vlad noted the need to balance these risks with the benefits of the medications. With regard to the ORAL Surveillance trial, he pointed to the fact that tofacitinib was used at a dose not approved for RA in the US.”⁸

Dr Vlad also noted, “Another thing that makes these results hard to interpret is that the comparator was adalimumab. We don’t definitively understand the risks for MACE [with] adalimumab. Maybe adalimumab is actually protective for those outcomes while the effect of tofacitinib is neutral; the results would be the same.”

Are Risks Similar Across Disease Types?

The ORAL Surveillance study was conducted among patients with PsA; however, it remains unclear whether adverse event risks from JAKis in patients with PsA differ from those in patients with RA.

It is conceivable that these risks could differ across diseases, given possible differences in the underlying pathologies. Patients with PsA vs RA were found to be younger, have fewer preexisting comorbidities, and use corticosteroids less commonly.⁹ Furthermore, following positive efficacy results in the pivotal phase 3 trials of tofacitinib in patients with PsA,^10,11 the long-term extension study Oral Psoriatic Arthritis (OPAL) Balance trial,¹² as well as the long-term follow-up,¹³ did not report new safety signals with regard to cardiovascular, thromboembolic, or cancerous outcomes.

Dr Weber observed, “We do not have direct comparisons of PsA [with] RA in the context of these drugs, so I prefer to err on the side of caution: I counsel patients similarly regardless of whether they are on a JAKi for RA or PsA.”

Are Risks Similar Across Different JAKis?

Janus kinase inhibitors differ in the terms of receptor selectivity. For example, tofacitinib inhibits activity at JAK-1 and JAK-3 receptors preferentially, with lesser activity at JAK-2; upadacitinib primarily acts at JAK-1.¹⁴ These differences prompt the question of whether adverse event risk profiles differ among the various JAKis.

The FDA approved upadacitinib for the treatment of PsA,¹⁵ based on the results of the SELECT-PsA 1¹⁶ and SELECT-PsA 2¹⁷ trials, as second-line therapy following failure with TNFis.

An integrated analysis of adverse events in both the SELECT-PsA trials,¹⁸ confirmed that infections, but not cardiovascular or thromboembolic events or malignancies, were higher with upadacitinib than with adalimumab.

Furthermore, Dr Vlad advised, “We have to be very careful making conclusions from the initial registration trials. There are small numbers of people in these trials who go on to develop conditions such as infections and cancers.”

Patient Care and Management

Dr Weber explained that a patient receiving treatment with JAKis inherently has a complex history of PsA. “These are not the first-line agents for rheumatic disease. This means that a patient with a prescription for a JAKi generally has [had an inadequate response to] other agents, and will require individualized care.”

She recalled one of her cases: a patient who had recently discontinued JAKi therapy and underwent cardiac risk stratification. Findings revealed that the patient had severe atherosclerosis with obstructive disease, which prompted aggressive preventive therapies, in addition to switching to a different disease-modifying agent.

“It’s important to look at each patient’s risk profile and discuss the signals of potential concern with the patient,” Dr Weber advised clinicians. “All patients, whether [receiving] a JAKi or a different agent, should receive screening that includes a lipid panel assay, including lipoprotein(a), body mass index (BMI) calculation, A1c level, and blood pressure measurement. We might also consider obtaining a calcium score to identify atherosclerosis or noninvasive cardiovascular stress testing, depending on the patient’s symptoms. And these screenings should be repeated periodically.”

Pharmacovigilance and real-world, long-term outcome studies can help refine risk estimates for rare adverse events, such as thromboembolic events or MACE.

“We want to be able to differentiate who is most at risk — given patient age, pertinent measurables, other medications they are taking, etc — vs who is at very low risk of an adverse event,” said Dr Vlad. “We probably won’t get another large, randomized safety trial like ORAL Surveillance, so we are going to need to do high-quality observational studies to better understand the risks.”