Scholarlink

Background <p>The progression of hepatocellular carcinoma (HCC) is significantly influenced by the tumor microenvironment, in which cancer-associated fibroblasts (CAFs), collagen fibers (CFs), and tumor-infiltrating lymphocytes (TILs) interact to regulate tumor growth and immune infiltration. Understanding the spatial and functional coordination among these components is essential for prognostic assessment.</p> Methods <p>This retrospective study included 335 patients from two tertiary hospitals. Early-stage HCC tissue samples were analyzed using artificial intelligence-based algorithms to quantify TIL fraction and aggregation, while collagen architecture was characterized using multiphoton microscopy (MPM). Moreover, single-cell RNA sequencing was used for functional profiling of CAFs, and multiplex immunofluorescence was applied for spatial validation. The relationship between TILs and collagen features was associated with clinical outcomes. Prognostic models were developed using Cox regression and validated internally and externally.</p> Results <p>MPM demonstrated strong co-localization of CFs and TILs, with lymphocytes spatially aligned following collagen orientation within the tumor core. CAF subtypes exhibited heterogeneous expression of extracellular matrix (ECM)-related and lymphocyte-modulatory genes, with ECM-mediated signaling dominating CAF–T/B cell communication. TIL aggregation and CF orientation independently predicted disease-free and overall survival. The multifactorial Fiber-TIL-Clinical model (FTC model) achieved superior predictive performance (area under the curve = 0.902 for early recurrence) by integrating these features with clinical variables.</p> Conclusions <p>The FTC model integrates CF features and TIL characteristics, demonstrating strong performance in predicting HCC recurrence.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Cancer-associated fibroblasts shape features of lymphocyte aggregation and collagen deposition in hepatocellular carcinoma

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月30日

查看全文
  • Yuan’e Lian,
  • Zehong Zhang,
  • Dongsheng Xia,
  • Zhongchang Weng,
  • Jiaying Ye,
  • Siqin Zhang,
  • Xingxing Huang,
  • Lianhuang Li,
  • Jie Lin,
  • Songqiang Zhou,
  • Shi Chen,
  • Yannan Bai

摘要

Background

The progression of hepatocellular carcinoma (HCC) is significantly influenced by the tumor microenvironment, in which cancer-associated fibroblasts (CAFs), collagen fibers (CFs), and tumor-infiltrating lymphocytes (TILs) interact to regulate tumor growth and immune infiltration. Understanding the spatial and functional coordination among these components is essential for prognostic assessment.

Methods

This retrospective study included 335 patients from two tertiary hospitals. Early-stage HCC tissue samples were analyzed using artificial intelligence-based algorithms to quantify TIL fraction and aggregation, while collagen architecture was characterized using multiphoton microscopy (MPM). Moreover, single-cell RNA sequencing was used for functional profiling of CAFs, and multiplex immunofluorescence was applied for spatial validation. The relationship between TILs and collagen features was associated with clinical outcomes. Prognostic models were developed using Cox regression and validated internally and externally.

Results

MPM demonstrated strong co-localization of CFs and TILs, with lymphocytes spatially aligned following collagen orientation within the tumor core. CAF subtypes exhibited heterogeneous expression of extracellular matrix (ECM)-related and lymphocyte-modulatory genes, with ECM-mediated signaling dominating CAF–T/B cell communication. TIL aggregation and CF orientation independently predicted disease-free and overall survival. The multifactorial Fiber-TIL-Clinical model (FTC model) achieved superior predictive performance (area under the curve = 0.902 for early recurrence) by integrating these features with clinical variables.

Conclusions

The FTC model integrates CF features and TIL characteristics, demonstrating strong performance in predicting HCC recurrence.