Single-Cell Transcriptomics in Rheumatoid Arthritis: A Comprehensive Review of Cellular Heterogeneity and Therapeutic Targets

Abstract

Rheumatoid arthritis is a chronic inflammatory joint disease in which persistent synovitis can progress to cartilage destruction, bone erosion, pain, and disability despite expanding therapeutic options. Clinically similar disease presentations may arise from distinct cellular and molecular programs within synovial tissue, limiting the effectiveness of uniform immunomodulation. This review evaluates how single-cell RNA sequencing and single-nucleus RNA sequencing have reshaped understanding of rheumatoid arthritis by resolving stromal and immune heterogeneity, defining context-dependent cell states, and linking intercellular signaling to therapeutic opportunities.
We synthesize recurring cell states across synovial tissue, synovial fluid, and peripheral blood and assess how experimental and computational choices influence state identification and cross-cohort comparability. Across studies, synovial fibroblast-like synoviocytes consistently organize along a spatial lining versus sublining axis, with inflammation-induced activation layered onto positional identity. Perivascular niches, including endothelial-stromal signaling, appear to instruct fibroblast programs and help explain gradients in stromal phenotypes. Immune profiling repeatedly identifies peripheral helper T cell states that support lymphoid organization and B cell help, alongside clonally expanded cytotoxic T cell programs whose interpretation is complicated by overlap between activation and inhibitory receptor expression. Myeloid diversity is best captured as continuous transitions shaped by tissue niche and inflammatory modules rather than binary polarization, with evidence for protective barrier-like macrophage programs and recruited inflammatory macrophage states. B cell and plasma cell heterogeneity reflects antigen experience, class switching, and association with ectopic lymphoid structures, while emerging work suggests metabolic dependencies across stromal and antibody-secreting compartments. We conclude that a Cell State–Niche–Target framework can reconcile heterogeneous atlases by prioritizing program-level concordance and explicitly conditioning therapeutic hypotheses on niche context. This review is synovium-centric; synovial fluid and peripheral blood are discussed primarily as bridging compartments for biomarker hypotheses rather than as independently comprehensive atlases. Progress toward precision medicine will require standardized sampling and analysis, spatial validation, and longitudinal studies that connect state shifts to treatment response.