New biomarkers in RA: Who to treat, Who to watch, Who to taper? Save

Identifying biomarkers that can be effectively translated into routine clinical practice has long been one of the holy grails of rheumatoid arthritis research, and is increasingly moving toward precision medicine. Yet despite major therapeutic advances, we still largely treat patients using a "one-size-fits-all" approach, whether at disease onset, when assessing extra-articular complications, or when considering treatment de-escalation. The ability to identify patients who are likely to respond poorly to standard therapy, develop severe disease manifestations, or flare during tapering remains an important unmet need.

Here is my summary of what EULAR 2026 is presenting to us in the field of biomarker-driven personalized care in RA.

In abstract OP045, Mankia and colleagues present results from the TEEMS trial, a proof-of-concept study exploring whether naïve CD4+ T-cell frequency can be used to guide first-line treatment decisions in early RA. Previous work has shown that reduced naïve CD4+ T-cell frequencies are associated with poorer responses to MTX. In this study, 103 newly diagnosed ACPA-positive RA patients with subnormal naïve CD4+ T-cell frequencies were randomized to receive either MTX alone or MTX + etanercept. 45 patients with normal naïve T-cell frequencies received standard MTX therapy. Remission rates at week 12 and 24 were higher in patients receiving etanercept + MTX (48% vs. 11.5% week 12). These findings suggest that immune phenotyping at diagnosis may help identify patients who would benefit from earlier biologic intervention. The author suggest this could represent a potentially cost-effective strategy although this is yet to be demonstrated.

In abstract OP046, Venkat et al. focus on early detection RA-ILD. Using a comprehensive autoantibody profiling approach in 149 RA-ILD patients compared to 378 RA-noILD controls, 50 IPF cases, and 99 healthy controls, the authors evaluated both traditional RA autoantibodies and a range of novel immune biomarkers. Among all markers examined, anti-carbamylated protein (anti-CarP) antibodies emerged as one of the strongest associations with RA-ILD. Elevated anti-CarP IgG levels distinguished RA-ILD patients from RA patients without lung disease, healthy controls, and even patients with idiopathic pulmonary fibrosis. Calprotectin, RF, CCP, and dsDNA positivity were also associated with RA-ILD, although non-RA-related autoantibodies generally showed limited utility. These findings reinforce the concept that distinct autoantibody signatures may characterize pulmonary involvement in RA and suggest that anti-CarP could become a useful biomarker for earlier identification and phenotyping of RA-ILD if this study was replicated in an independent cohort.

In abstract OP047, Lourido and colleagues address another key challenge in modern RA management: identifying patients who can safely taper biologic therapy. The authors used proteomic analysis from the OPTIBIO RCT, conducted in five Spanish hospitals, which evaluated whether reduced-dose biologic therapy could sustain remission in RA patients. Baseline serum samples from 44 patients in sustained remission after biologic dose reduction were analyzed for 800 circulating proteins. Higher baseline levels of VSIG4 and anti-interferon gamma (anti-IFNγ) autoantibodies independently predicted future disease flare (OR 1.2 and OR 4.4 respectively). Combining these biomarkers with DAS28-CRP significantly achieved an AUC of 0.876. These results suggest that circulating proteomic and immunologic signatures may help identify patients at risk of relapse before tapering.

Together, these abstracts highlight new biomarker-driven approaches which can influence multiple stages of the RA journey for patients and clinicians. While further validation is needed, these studies represent important steps toward a more personalized approach to RA management, where treatment decisions are guided not only by disease activity but also by the underlying biology of each patient. The next challenge, however, will be to provide evidence that these approaches are cost-effective and add value beyond existing clinical tools.