Fibroblast TGFβ Signaling in Vascular Niches May Drive Treatment-Refractory RA, With Kevin Wei, MD, PhD

Treatment-refractory rheumatoid arthritis (RA) may be driven in part by a fibroproliferative process within synovial vascular niches that persists despite effective immune cell depletion — a finding that challenges the assumption that intensifying immunosuppression is the appropriate response to inadequate treatment response, according to research published in Nature Immunology.

The research used spatial transcriptomic profiling — a technique that captures gene expression while preserving tissue architecture — to analyze synovial biopsy samples from patients with RA taken 6 months before and after treatment. Patients who did not achieve remission were identified as having a distinct molecular signature at baseline: increased fibrogenic signaling within vascular tissue niches, marked by high fibroblast expression of COMP, a gene encoding cartilage oligomeric matrix protein and a component of the extracellular matrix.

RheumatologyLive spoke with investigator Kevin Wei, MD, PhD, assistant professor of medicine at Harvard Medical School and rheumatologist at Brigham and Women's Hospital in Boston, about what the new research means for clinicians.

The study further identified endothelial-derived Notch signaling as an upstream regulator of fibroblast TGFβ activity. Notch signaling was found to exert opposing effects on TGFβ isoform expression and TGFβ receptor expression within the joint, generating a proximal-to-distal gradient of TGFβ sensitivity across the synovial tissue. Disruption of steady-state Notch signaling was able to alter this gradient. In post-treatment biopsy samples from non-remitting patients, significant immune cell depletion was observed — consistent with the immunosuppressive drugs having achieved their intended immunologic effect — but this depletion was accompanied by expansion of fibrogenic niches rather than resolution of pathological tissue architecture. In patient-derived synovial organoids, inhibition of both Notch and TGFβ signaling was sufficient to reverse this fibrogenic expansion.

Wei described this pattern as having direct implications for how clinicians think about treatment failure in RA. The current clinical response to inadequate treatment response — whether switching to a different biologic class or adding immunosuppression such as corticosteroids — is premised on the assumption that residual inflammation is the primary driver of refractory disease. These data raise an alternative: in some patients, the immune cell compartment may already be effectively suppressed, and the ongoing disease is instead perpetuated by a fibroproliferative mesenchymal process that current immunosuppressive therapies do not target. Adding further immunosuppression in this setting may not alter outcomes. He drew an analogy to the role of adjuvant therapy in oncology, where a primary treatment — surgery — is augmented by chemotherapy or radiation to address pathological processes the primary intervention cannot independently resolve. In refractory RA, the concept would be analogous: ongoing immunosuppression as the primary treatment, supplemented by an adjuvant targeting fibroblast or mesenchymal stromal cell pathways to prevent exuberant synovial fibrosis and restore sensitivity to the primary agent.

Translating this mechanistic framework into clinical practice will require identification of accessible biomarkers that reflect what is occurring in the synovial tissue without requiring joint biopsy — which, while feasible in research settings, is not standard practice in the United States. Wei identified blood-based biomarkers — whether circulating proteins or specific cell populations — that correlate with the fibrogenic joint tissue signature as the immediate next research priority. Establishing such a surrogate would enable patient stratification in future clinical trials testing anti-fibrotic adjuvant strategies in refractory RA.

“With the idea that if you can treat inflammation, really dampen the inflammatory process in the joint, then a patient not only will feel better, but you prevent joint damages, long term consequences, from arthritis,” Wei said.

Wei’s reported disclosures include Merck Pharmaceuticals, BMS, Anaptys Bio and 10x Genomics.

Reference

Bhamidipati K, McIntyre ABR, Kazerounian S, et al; Accelerating Medicines Partnership RA/SLE Network. Spatial patterning of fibroblast TGFβ signaling underlies treatment resistance in rheumatoid arthritis. Nat Immunol. Published online January 15, 2026. doi:10.1038/s41590-025-02386-2