Vagus Nerve Stimulation in Refractory Rheumatoid Arthritis: RESET-RA Trial

In RESET-RA, implantable vagus nerve stimulation improved the primary ACR20 clinical response at three months versus sham, introducing a device-based therapeutic option for refractory rheumatoid arthritis.

The trial enrolled 242 patients with moderate-to-severe RA who had inadequate response or intolerance to one or more biological or targeted synthetic DMARDs. The prespecified primary end point—ACR20 at 3 months—favored active stimulation: 35.2% versus 24.2% with sham, a stratification-adjusted difference of 11.8% (P = 0.0209). The trial used blinded implantation and sham stimulation to minimize bias and preserve clinical-grade rigor, confirming a statistically significant short-term clinical benefit for the device-based approach.

Benefit persisted through the open-label extension and deepened over time. By all-completers analysis, ACR20 rose to 50.0% at 6 months and 52.8% at 12 months. Secondary composite measures and patient-reported function trended higher through 12 months, and most participants remained off additional b/tsDMARD therapy during follow-up—indicating durable clinical responses beyond the initial blinded period.

In a prespecified high-erosive phenotype subgroup assessed by OMERACT RAMRIS, new or worsening bone erosion (>0.5 score increase) from baseline to 3 months occurred in 18.9% of stimulated patients versus 37.8% of sham patients (P = 0.016). The trial thus demonstrated a short-term, statistically significant reduction in radiographic erosion progression among patients at elevated structural risk, supporting potential structural preservation in this phenotype.

Safety data for vagus nerve stimulation were consistent with an acceptable tolerability profile through 12 months. Most device-related adverse events were mild-to-moderate and implantation-related. Nonserious implantation-related events occurred in 15.6% of patients (~38 of 242; 52 events), most commonly hoarseness—vocal cord paresis in 4.5% and dysphonia in 2.9%—with a small number of surgical-site infections or stitch complications; six patients elected explantation before 12 months. Related serious adverse events were perioperative (~1.6%) and resolved without sequelae; no unexpected long-term safety signals were identified over the study period.