JAK Inhibitors in RA: 2025 EULAR Safety Update Redraws Risk Guidance

Rheumatoid arthritis management has never had a shortage of therapeutic options, but it has had a persistent shortage of clarity on long-term safety trade-offs. JAK inhibitors, approved across multiple indications since tofacitinib's initial licence in 2012, accumulated substantial real-world use before the ORAL Surveillance trial forced a sector-wide reassessment of their cardiovascular and oncological risk profile. That reassessment is now embedded formally in European guidance.³

The 2025 EULAR recommendations for the management of RA with synthetic and biologic DMARDs represent the first full update to incorporate the post-ORAL Surveillance evidence landscape.³ Alongside the guideline document, Laskou and colleagues conducted the systematic literature review that informed the safety sections of those recommendations, covering evidence from trials, registries, and observational datasets across the full DMARD spectrum.¹ Cohen and Winthrop provide a parallel narrative synthesis of advances in JAK inhibitor use, contextualising where the evidence has moved since earlier safety alerts.²

What the evidence covers

The systematic literature review by Laskou et al. examined safety data across conventional synthetic DMARDs, biological DMARDs, and targeted synthetic DMARDs including JAK inhibitors, covering outcomes including serious infections, major adverse cardiovascular events (MACE), venous thromboembolism (VTE), and malignancy.¹ The review was conducted to the EULAR standard methodology for SLRs informing guideline updates, providing the evidentiary base from which the Smolen et al. recommendations were drafted.^1,3

Cohen and Winthrop note that differential selectivity across JAK inhibitor agents, tofacitinib, baricitinib, upadacitinib, and filgotinib, translates into clinically meaningful differences in safety signal distribution, though head-to-head comparative data between agents remains limited in scope and duration.² The ORAL Surveillance findings, which demonstrated elevated MACE and malignancy risk with tofacitinib versus TNF inhibitors in patients aged 50 or older with at least one cardiovascular risk factor, continue to anchor the regulatory and clinical framework, but the 2025 update attempts to move the conversation toward agent-specific and patient-specific calibration rather than a class-wide restriction.^2,3

The updated EULAR recommendations retain the positioning of JAK inhibitors after failure or intolerance of at least one bDMARD in patients with relevant risk factors, while acknowledging that in patients without elevated cardiovascular, thrombotic, or malignancy risk, the benefit-risk calculation is more permissive.³ The SLR identifies older age, smoking history, prior malignancy, known cardiovascular disease, and risk factors for VTE as the principal variables requiring prospective assessment before JAK inhibitor initiation.¹ Upadacitinib's higher selectivity for JAK1 was noted by Cohen and Winthrop as a pharmacologically plausible differentiator, though whether this translates into a superior long-term safety profile compared to less selective agents requires confirmatory long-term data.²

On infection risk, the SLR confirms that herpes zoster reactivation remains a consistent class effect across JAK inhibitors, with rates exceeding those observed with bDMARDs in most comparative datasets.¹ Vaccination against herpes zoster prior to JAK inhibitor initiation is incorporated into the updated recommendations as a standard pre-treatment step.³ Serious bacterial infection rates, while elevated relative to conventional DMARDs, appear broadly comparable to those seen with bDMARDs in most datasets reviewed, though the evidence quality varies considerably by agent and study design.¹