Background/Purpose: Rheumatoid arthritis (RA) is a chronic autoimmune disease driven by persistent synovial inflammation and progressive joint damage. Selecting an effective biologic therapy remains a costly, time-consuming trial-and-error process, as blood-based biomarkers poorly predict therapeutic response. In this study, we analyzed synovial tissue from RA patients by integrating single-cell and spatial transcriptomics to identify tissue-resident immune populations, particularly macrophage subsets, associated with response or resistance to therapy

Methods: Samples from minimally invasive, ultrasound‐guided synovial biopsies were collected at baseline (n=15) and 12‐week follow‐up (n=6) from RA patients prior to start of a new DMARD (responders n=8; non-responders n=6; dropped out n=1) under IRB-approved protocols with informed consent. Samples were dissociated, enriched for CD45⁺ or CD45⁺CD11b⁺ cells, and processed on the 10X Genomics Chromium platform to generate libraries for single-cell RNA sequencing with cell-surface protein detection (CITE-seq). Adjacent sections were analyzed with 10X Xenium (n=3, 196-gene panel) and NanoString CosMx (n=1, 1,000-gene panel) platforms to profile spatially resolved gene expression of synovial tissue at single-cell resolution.

Results: From the synovial CITE-seq analysis, five major immune populations were identified: myeloid cells, T cells, B cells, plasma cells, and mast cells. Clinical response stratification of patients revealed that baseline samples from non-responders exhibited increased T cells proportions compared to responders. However, the T cell to myeloid ratio tended to decrease in these patients following therapy. In the myeloid compartment, the FCN1+S100A12 macrophage subpopulation was enriched in non-responders, but decreased after treatment. On the other hand, the MERTK⁺LYVE1⁺ and TIMD4 subpopulations were depleted in responders but both increased after treatment. Spatial molecular imaging by CosMx revealed layered tissue organization, with transcriptionally distinct macrophages (including MERTK+LYVE1+) and fibroblasts at the synovial lining, and sub-lining macrophages, stromal, vascular, and immune cells occupying interstitial and perivascular zones. Xenium spatial profiling confirmed a compartmentalized immune–stromal landscape with lymphocytes forming discrete follicle-like clusters, supporting spatially organized immune–stromal niches in RA.

Conclusion: Our results demonstrating that TIMD4⁺ and SPP1⁺ macrophages in non-responders reflects persistent synovial inflammation despite therapy, whilesynovial lining MERTK⁺LYVE1⁺ and TIMD4+ expansion indicates a broader tissue-adaptive programmacrophages are enriched in responders support prior observations that these cells are necessary for synovial health. The opposing trends in the myeloid compartment suggest that deciphering the interactions between cells is critical to understanding treatment response. Our integrated single-cell and spatial analysis highlights macrophage subset dynamics and immune–stromal organization linked to treatment outcomes, offering new insights into RA pathogenesis and therapeutic resistance.

« Back to ACR Convergence 2025

ACR Meeting Abstracts - https://acrabstracts.org/abstract/multimodal-spatial-and-single-cell-profiling-of-synovial-tissue-reveals-macrophage-programs-associated-with-biologic-therapytreatment-response-in-rheumatoid-arthritis/