Scholarlink

<p>The existing TNM staging system provides insufficient prognostic information in gastric cancer (GC) patients. This study aims to establish a pathomics signature of GC (PS<sub>GC</sub>) that uses deep learning (DL) to directly analyze H&amp;E slides for predicting GC outcomes. We propose a multi-scale graph neural network with gated attention mechanism for multi-instance learning (MS-GMIL) for the construction of PS<sub>GC</sub>. Moreover, transcriptomic data investigated the possible pathophysiological mechanisms of the PS<sub>GC</sub>. The PS<sub>GC</sub> was identified as an independent prognostic factor in all cohorts. Patients with stage II and III GC, along with a high PS<sub>GC</sub>, showed considerable benefits from chemotherapy and an effective response to immunotherapy. The primary histological features underlying the PS<sub>GC</sub> were tumor cell anaplasia, intraepithelial neoplasia, tumor-stroma fibrosis, and intestinal epithelial metaplasia. Moreover, the PS<sub>GC</sub> was associated with cell cycle regulation, drug resistance pathways, and mechanisms of cancer progression. The PS<sub>GC</sub> functions as a valuable tool in clinical decision-making for the management of GC, providing insights into the underlying pathogenic mechanisms.</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

Deep learning-based pathomics signature predicts prognosis and treatment response in gastric cancer: a multicenter retrospective study

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年4月10日

查看全文
  • Hao Wang,
  • Hao Li,
  • Keru Ma,
  • Genshen Mo,
  • Meihong Yan,
  • Xinyue Zhang,
  • Haonan Xie,
  • Yuze Huang,
  • Huiying Li,
  • Yingwei Xue,
  • Peng Han,
  • Shenghan Lou

摘要

The existing TNM staging system provides insufficient prognostic information in gastric cancer (GC) patients. This study aims to establish a pathomics signature of GC (PSGC) that uses deep learning (DL) to directly analyze H&E slides for predicting GC outcomes. We propose a multi-scale graph neural network with gated attention mechanism for multi-instance learning (MS-GMIL) for the construction of PSGC. Moreover, transcriptomic data investigated the possible pathophysiological mechanisms of the PSGC. The PSGC was identified as an independent prognostic factor in all cohorts. Patients with stage II and III GC, along with a high PSGC, showed considerable benefits from chemotherapy and an effective response to immunotherapy. The primary histological features underlying the PSGC were tumor cell anaplasia, intraepithelial neoplasia, tumor-stroma fibrosis, and intestinal epithelial metaplasia. Moreover, the PSGC was associated with cell cycle regulation, drug resistance pathways, and mechanisms of cancer progression. The PSGC functions as a valuable tool in clinical decision-making for the management of GC, providing insights into the underlying pathogenic mechanisms.