Scholarlink

Objective <p>To develop and validate a nomogram for predicting lymph node metastasis in early gastric cancer according to the characteristics of the tumor microenvironment to optimize treatment strategies.</p> Methods <p>Clinicopathological data of 882 early gastric cancer patients from three medical centers were retrospectively collected. Among them, 744 cases from Center 1 were assigned to the training set, and 138 cases from Centers 2 and 3 were the validation set. Tumor-infiltrating lymphocytes (TILs) and tumor-stroma ratio (TSR) were quantified in hematoxylin–eosin stained sections using QuPath software. The number and maturity of tertiary lymphoid structures (TLS) were evaluated through multiplex immunofluorescence. Nomogram prediction models were constructed using independent risk factors identified by univariate and multivariate logistic regression analyses. The performance of the model was evaluated by receiver operating characteristics (ROC) curve, calibration curve, and clinical decision curve analyses.</p> Results <p>Lymph node metastasis rates were 19.2% in the training set and 23.9% in the validation set; sex, lymphovascular invasion, TSR, and TLS were identified as independent risk factors. The model achieved an area under the ROC curve of 0.807 [95% confidence interval (CI) 0.766–0.843] in the training set and 0.822 (95% CI 0.721–0.907) in the validation set. The calibration curves indicated good agreement between predicted probabilities and actual incidence rates, and the clinical decision curve analysis revealed a positive clinical net benefit over a broad range of thresholds.</p> Conclusion <p>The nomogram developed in this study showed high accuracy, stability, and clinical utility, suggesting its value for guiding treatment decision-making in early gastric cancer.</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

Multicenter research: tumor microenvironment features combined with tumor cell characteristics predict lymph node metastasis in early gastric cancer

  • 期刊论文

  • 发表日期：

    2025年11月26日

查看全文
  • Xiaozhuo Gao,
  • Xiaoyan Zhao,
  • Huihui Xu,
  • Ning Zhang,
  • Fujing Sun,
  • Yong Zhang,
  • Jing Yang,
  • Yanmei Zhu

摘要

Objective

To develop and validate a nomogram for predicting lymph node metastasis in early gastric cancer according to the characteristics of the tumor microenvironment to optimize treatment strategies.

Methods

Clinicopathological data of 882 early gastric cancer patients from three medical centers were retrospectively collected. Among them, 744 cases from Center 1 were assigned to the training set, and 138 cases from Centers 2 and 3 were the validation set. Tumor-infiltrating lymphocytes (TILs) and tumor-stroma ratio (TSR) were quantified in hematoxylin–eosin stained sections using QuPath software. The number and maturity of tertiary lymphoid structures (TLS) were evaluated through multiplex immunofluorescence. Nomogram prediction models were constructed using independent risk factors identified by univariate and multivariate logistic regression analyses. The performance of the model was evaluated by receiver operating characteristics (ROC) curve, calibration curve, and clinical decision curve analyses.

Results

Lymph node metastasis rates were 19.2% in the training set and 23.9% in the validation set; sex, lymphovascular invasion, TSR, and TLS were identified as independent risk factors. The model achieved an area under the ROC curve of 0.807 [95% confidence interval (CI) 0.766–0.843] in the training set and 0.822 (95% CI 0.721–0.907) in the validation set. The calibration curves indicated good agreement between predicted probabilities and actual incidence rates, and the clinical decision curve analysis revealed a positive clinical net benefit over a broad range of thresholds.

Conclusion

The nomogram developed in this study showed high accuracy, stability, and clinical utility, suggesting its value for guiding treatment decision-making in early gastric cancer.