列线图预测乳腺癌腋窝淋巴结转移的建立与验证。

Development and Validation of a Nomogram for Predicting Axillary Lymph Node Metastasis in Breast Cancer.

机构信息

College of Biomedical Engineering, Taiyuan University of Technology, Jinzhong, Shanxi, China.

College of Information and Computer, Taiyuan University of Technology, Jinzhong, Shanxi, China.

出版信息

Clin Breast Cancer. 2023 Jul;23(5):538-545. doi: 10.1016/j.clbc.2023.04.002. Epub 2023 Apr 12.

DOI:10.1016/j.clbc.2023.04.002

PMID:37137800

Abstract

BACKGROUND

Axillary lymph node (ALN) status is a key prognosis indicator for breast cancer patients. To develop an effective tool for predicting axillary lymph node metastasis in breast cancer, a nomogram was established based on mRNA expression data and clinicopathological characteristics.

MATERIALS AND METHODS

A 1062 breast cancer patients with mRNA data and clinical information were obtained from The Cancer Genome Atlas (TCGA). We first analyzed the differentially expression genes (DEGs) between ALN positive and ALN negative patients. Then, logistic regression, least absolute shrinkage and selection operator (Lasso) regression, and backward stepwise regression were performed to select candidate mRNA biomarkers. The mRNA signature was constructed by the mRNA biomarkers and corresponding Lasso coefficients. The key clinical factors were obtained by Wilcoxon-Mann-Whitney U test or Pearson's χ test. Finally, the nomogram for predicting axillary lymph node metastasis was developed and evaluated by concordance index (C-index), calibration curve, decision curve analysis (DCA), and receptor operating characteristic (ROC) curve. Furthermore, the nomogram was externally validated using Gene Expression Omnibus (GEO) dataset.

RESULTS

The nomogram for predicting ALN metastasis yielded a C-index of 0.728 (95% CI: 0.698-0.758) and an AUC of 0.728 (95% CI: 0.697-0.758) in the TCGA cohort. In the independent validation cohort, the C-index and AUC of the nomogram were up to 0.825 (95% CI: 0.695-0.955) and 0.810 (95% CI: 0.666-0.953), respectively.

CONCLUSION

This nomogram could predict the risk of axillary lymph node metastasis in breast cancer and may provide a reference for clinicians to design individualized axillary lymph node management strategies.

摘要

背景

腋窝淋巴结（ALN）状态是乳腺癌患者的关键预后指标。为了开发一种有效的工具来预测乳腺癌的腋窝淋巴结转移，我们基于 mRNA 表达数据和临床病理特征建立了一个列线图。

材料与方法

从癌症基因组图谱（TCGA）中获取了 1062 名具有 mRNA 数据和临床信息的乳腺癌患者。我们首先分析了 ALN 阳性和 ALN 阴性患者之间的差异表达基因（DEGs）。然后，进行逻辑回归、最小绝对值收缩和选择算子（Lasso）回归以及向后逐步回归以选择候选 mRNA 生物标志物。通过 mRNA 标志物和相应的 Lasso 系数构建 mRNA 特征。通过 Wilcoxon-Mann-Whitney U 检验或 Pearson χ 检验获得关键临床因素。最后，通过一致性指数（C-index）、校准曲线、决策曲线分析（DCA）和受体操作特征（ROC）曲线开发和评估用于预测腋窝淋巴结转移的列线图。此外，使用基因表达综合数据库（GEO）数据集对列线图进行外部验证。

结果

列线图预测 ALN 转移在 TCGA 队列中的 C-index 为 0.728（95%CI：0.698-0.758），AUC 为 0.728（95%CI：0.697-0.758）。在独立验证队列中，列线图的 C-index 和 AUC 高达 0.825（95%CI：0.695-0.955）和 0.810（95%CI：0.666-0.953）。