Opposing roles of the stromal architecture surrounding and occupying tumor mass in hepatocellular carcinoma
Source: CNGBdb Project (ID CNP0004214)
Source: CNGBdb Project (ID CNP0004214)
0 0
Project name: Opposing roles of the stromal architecture surrounding and occupying tumor mass in hepatocellular carcinoma
Description: We performed an integrated spatial characterization of the stromal architecture surrounding and occupying tumor mass in hepatocellular carcinoma (HCC) using tumor margin samples paired with tumor core non-tumor liver tissues from 58 patients. By reanalyzing prior molecular and morphological datasets, we discovered opposing stromal architecture and clinical outcomes between SUR- and SUR+ HCC. Spatial proteomics revealed major patterns of cancer-associated stroma in HCC characterized by ECM deposition, metabolic preferences, and immunological features. Using Stereo-seq in combination with scRNA-seq, snRNA-seq, and snATAC-seq revealed 35 cell subsets (across fibroblast, immune and malignant cells) and their spatial coordinates with respect to tumor stroma. Especially, we defined 5 fibroblast subtypes with distinct ECM profiles and spatial positioning. CAF1_FAP were enriched in the occupying stroma of SUR- tumors, with widespread RUNX1 chromatin activity, and co-localized with PDCD1+CD8T cells forming immune-regulatory hubs. By contrast, CAF3_C7 were gathered in the surrounding stroma of SUR+ tumors, with increased chromatin activity of USF2, and were often accompanied by SPP1+Macrophages to form wound-healing hubs. The clinical relevance of these 2 fibroblast subsets was further validated as transcriptional signatures and protein markers. This study suggested that spatially heterogeneous fibroblast subpopulations may play a role in driving disease progression and immunologic processes in human cancer by shaping stromal architecture and we provided a reference for designing future spatial multi-omics profiling. A more comprehensive understanding of stromal heterogeneity across human tumors in the spatial context may generate novel biomarkers and therapeutic opportunities.
Data type: Raw sequence reads; Proteome; Transcriptome or Gene expression; Variation
Sample scope: Multiisolate
Relevance: Medical
Submitter: 芳 小 陈(Xiaofang Chen); BGI
Literatures
- PMID: 39870619
Release date: 2025-02-21
Last updated: 2025-02-21
DOI: 10.26036/CNP0004214
Statistics: 4 single cells