摘要
Immunoglobulin G4-related disease (IgG4-RD) is a complex systemic fibroinflammatory disorder characterized by tumefactive lesions, storiform fibrosis, and a dense lymphoplasmacytic infiltrate rich in IgG4-positive plasma cells. Despite its recognition, the precise pathogenic mechanisms driving the aberrant immune response, chronic inflammation, and resultant tissue fibrosis remain incompletely understood, hindering the development of targeted therapies. Conventional analytical methods, such as flow cytometry and bulk transcriptomics, fail to preserve the critical spatial context of cellular interactions within the tissue microenvironment. This study aimed to overcome this limitation by employing Imaging Mass Cytometry (IMC), a high-plex, high-resolution technology at 1 μm resolution, to generate a comprehensive spatial atlas of IgG4-RD lesions. Our primary objectives were to dissect the detailed cellular composition and architecture, map intercellular communication networks, and correlate specific spatial-cellular features with clinical phenotypes, particularly long-term disease stability, to identify novel diagnostic and therapeutic targets.
We analyzed formalin-fixed, paraffin-embedded (FFPE) tissue specimens obtained from 15 treatment-naïve IgG4-RD patients. The samples represented major involved organs, including salivary glands, pancreas, lymph nodes, and the ureter, capturing disease heterogeneity. A highly multiplexed antibody panel of 37 metal-tagged markers was meticulously designed and validated. This panel encompassed markers for: (1) major immune lineages (T cells, B cells, IgG4+plasma cells); (2) myeloid cell subsets (macrophages, monocytes, dendritic cells, neutrophils); (3) stromal and structural components (fibroblasts, epithelial cells, smooth muscle cells, endothelium); and (4) functional states (proliferation, apoptosis, immune checkpoint expression, fibrotic signaling). For each sample, three histopathologically defined Regions of Interest (ROIs)—Germinal Center (GC), Inflammatory Region (IFM), and Fibrotic Region (FIB)—were selected for analysis. IMC imaging was performed following standardized protocols. Subsequent computational analysis involved image pre-processing and registration using the Steinbock pipeline, noise reduction with IMC-Denoise, and high-accuracy single-cell segmentation coupled with cellular phenotyping using DeepCell. From a total of 403,446 segmented cells, unsupervised clustering and expert-guided annotation were employed to systematically define 22 distinct, biologically relevant cell subsets. Advanced spatial analysis included cellular abundance quantification, assessment of regional enrichment, cellular neighborhood analysis, and construction of k-nearest neighbor (kNN) interaction graphs. Statistical correlations were performed to link cellular signatures with clinical data, focusing on patients exhibiting long-term stability either with or without prior immunosuppressive therapy.
Our high-dimensional spatial analysis unveiled a structured yet heterogeneous cellular microenvironment within IgG4-RD lesions. The global cellular landscape was dominated by six major subsets: IgG4+ plasma cells, B cells, fibroblasts, CD4+ T cells, CD8+ T cells, and T follicular helper (Tfh) cells. Significant compositional heterogeneity was observed across different anatomical sites, underscoring the organ-specific manifestation of the disease. Spatial mapping revealed highly compartmentalized cellular distributions: GCs were predominantly lymphoid structures enriched in B cells and Tfh cells; IFMs constituted the core inflammatory zones with the highest density of IgG4+ plasma cells; and FIB regions were primarily occupied by fibroblasts, smooth muscle cells, and specific macrophage subsets (CD163hi, TGF-β1+), indicative of an active fibrogenic niche. Cellular neighborhood analysis further resolved the tissue into 8 recurrent functional microdomains. Notably, a Tfh-B cell-dominant neighborhood representing a GC dark zone was spatially adjacent to an IgG4+ plasma cell-rich neighborhood, providing architectural evidence for the local differentiation and maintenance of pathogenic plasma cells. Critically, comparative analysis between clinical subgroups identified key cellular features associated with distinct modes of disease stability. Lesions from treatment-naïve patients with long-term stability showed a significant elevation in CD163hi macrophages, CD38hi macrophages, and proliferating B cells compared to lesions from patients who achieved stability following treatment. This signature suggests a potentially distinct, macrophage-modulated immunologic set-point in untreated stable disease.
This study establishes the first high-resolution spatial cellular atlas of human IgG4-RD lesions, providing an unprecedented view of the disease microenvironment. We delineate a pathogenic "immune-stromal unit" centered on IgG4+ plasma cells, orchestrated by specific activated macrophage subsets and fibroblasts, which spatially coordinates B-cell hyperactivity, innate immune activation, and fibrotic remodeling. The identification of unique cellular signatures—specifically enriched CD163hi macrophages, CD38hi macrophages, and proliferating B cells—in treatment-naïve stable patients offers novel, spatially-defined biomarkers that may help stratify disease endotypes and predict clinical course. These findings move beyond cellular inventories to reveal the functional spatial logic of IgG4-RD, offering crucial insights for the development of niche-targeted therapeutic strategies and refined patient management.
