Scholarlink

Objectives <p>This study sought to identify key independent determinants of axillary lymph node metastasis (ALNM) in breast cancer, focusing on tumor spatial distribution and clinicopathological characteristics. A preoperative model was established to predict ALNM risk and guide individualized axillary treatment strategies. To assess the model’s clinical practicality, we validated it further via decision curve analysis (DCA) and clinical impact curves (CIC), with results highlighting its relevance for daily clinical work.</p> Methods <p>We retrospectively included 446 breast cancer patients who received care at the Affiliated Hospital of Qingdao University from December 2022 to December 2024. The subjects were randomly assigned to a training set (358 cases) or a validation set (88 cases). Spatial measurements—shortest distances from the tumor edge to the skin, nipple, and sentinel lymph node —were obtained via ultrasonography and binarized. These parameters were analyzed with pathological features. Multivariate logistic regression was used to identify key predictive factors and develop a visualized nomogram. Receiver operating characteristic (ROC) curves assessed predictive accuracy; separately, DCA and CIC estimated clinical benefit and supported decision-making.</p> Results <p>Among the 446 enrolled patients, 124 (27.8%) had pathologically confirmed ALNM. Multivariate logistic regression identified nuclear grade, lymphovascular invasion (LVI), tumor location, and three spatial indicators as independent ALNM predictors. A nomogram incorporating these six variables was constructed, showing strong discriminative ability: the area under the ROC curve (AUC) was 0.83 (95% confidence interval [95% CI]: 0.79–0.87) in the training cohort and 0.89 (95% CI: 0.81–0.95) in the validation cohort. DCA across a threshold probability range of ~ 0.1–0.7 showed the nomogram outperformed both "treat-all" and "treat-none" strategies. CIC further confirmed effective patient risk stratification by the model, which helps reduce unnecessary treatments while maintaining optimal patient care.</p> Conclusion <p>This study developed a clinically relevant model to predict preoperative axillary lymph node metastasis in breast cancer. Commonly available clinicopathological and spatial factors were used as key inputs for model development. The model showed favorable discrimination and clinical utility in both training and validation cohorts. Its application may aid individualized axillary risk assessment and refine surgical decision-making, especially for de-escalating axillary intervention.</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

A clinicopathological and spatial feature-based nomogram for individualized prediction of axillary lymph node involvement in breast cancer

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年1月27日

查看全文
  • Xin-Yi Fang,
  • Chang-Gen Liu,
  • Teng Ma,
  • Xin-Yi Sun,
  • Xiang-Yu Shen,
  • Yan Mao,
  • Hai-Bo Wang

摘要

Objectives

This study sought to identify key independent determinants of axillary lymph node metastasis (ALNM) in breast cancer, focusing on tumor spatial distribution and clinicopathological characteristics. A preoperative model was established to predict ALNM risk and guide individualized axillary treatment strategies. To assess the model’s clinical practicality, we validated it further via decision curve analysis (DCA) and clinical impact curves (CIC), with results highlighting its relevance for daily clinical work.

Methods

We retrospectively included 446 breast cancer patients who received care at the Affiliated Hospital of Qingdao University from December 2022 to December 2024. The subjects were randomly assigned to a training set (358 cases) or a validation set (88 cases). Spatial measurements—shortest distances from the tumor edge to the skin, nipple, and sentinel lymph node —were obtained via ultrasonography and binarized. These parameters were analyzed with pathological features. Multivariate logistic regression was used to identify key predictive factors and develop a visualized nomogram. Receiver operating characteristic (ROC) curves assessed predictive accuracy; separately, DCA and CIC estimated clinical benefit and supported decision-making.

Results

Among the 446 enrolled patients, 124 (27.8%) had pathologically confirmed ALNM. Multivariate logistic regression identified nuclear grade, lymphovascular invasion (LVI), tumor location, and three spatial indicators as independent ALNM predictors. A nomogram incorporating these six variables was constructed, showing strong discriminative ability: the area under the ROC curve (AUC) was 0.83 (95% confidence interval [95% CI]: 0.79–0.87) in the training cohort and 0.89 (95% CI: 0.81–0.95) in the validation cohort. DCA across a threshold probability range of ~ 0.1–0.7 showed the nomogram outperformed both "treat-all" and "treat-none" strategies. CIC further confirmed effective patient risk stratification by the model, which helps reduce unnecessary treatments while maintaining optimal patient care.

Conclusion

This study developed a clinically relevant model to predict preoperative axillary lymph node metastasis in breast cancer. Commonly available clinicopathological and spatial factors were used as key inputs for model development. The model showed favorable discrimination and clinical utility in both training and validation cohorts. Its application may aid individualized axillary risk assessment and refine surgical decision-making, especially for de-escalating axillary intervention.