基于SEER数据库构建老年乳腺癌患者腋窝淋巴结转移的列线图

Nomogram for predicting axillary lymph node metastasis in elderly patients with breast cancer:an analysis based on SEER database

目的筛选老年乳腺癌患者腋窝淋巴结转移的独立预测因子并构建风险预测模型,评估老年乳腺癌患者腋窝淋巴结转移的风险。方法研究纳入SEER数据库40647例老年乳腺癌患者,并按3∶1随机分为训练集和测试集,武汉大学中南医院2018年6月至2019年9月确诊的70例老年乳腺癌作为验证集。由训练集进行单因素分析和多因素Logistic回归分析以选择变量并构建预测腋窝淋巴结转移的风险预测模型。ROC曲线下面积(AUC)和校准曲线分别用于评估列线图的准确性和判别能力,并在训练集和验证集中进行内部和外部验证。结果多因素分析表明性别,肿瘤分级,胸壁或皮肤浸润,分子亚型和肿瘤大小是老年乳腺癌患者腋窝淋巴结转移的重要预测指标。预测老年乳腺癌患者腋窝淋巴结转移列线图的AUC值为0.744,校准曲线显示模型预测和实际转移情况具有较高的一致性。验证了列线图的准确性和判别能力。结论该列线图有助于预测老年乳腺癌患者术前腋窝淋巴结的状况,进而指导手术方案的选择。

Objective To screen independent predictors of axillary lymph node(ALN)metastasis in elderly patients with breast cancer and construct a risk prediction model to assess the risk of ALN metastasis in elderly patients with breast cancer.Methods The study included 40,647 elderly patients with breast cancer from the Surveillance,Epidemiology and End Results(SEER)database and randomly divided them into training set and test set according to the ratio of 3∶1,and 70 elderly patients with breast cancer who were diagnosed between June 2018 and September 2019 at the Zhongnan Hospital of Wuhan University as validation set.Univariate analysis and multivariate logistic regression were performed in the training set to select factors predicting axillary lymph node metastasis and construct risk prediction model.The area under the curve(AUC)and the calibration curve are used to evaluate the accuracy and discriminative ability of the nomogram,respectively.Internal and external verifications was performed in the training set and validation set.Results Multivariate results showed that gender,grade,chest wall or skin infiltration,subtype and tumor size are important predictors of ALN metastasis in elderly patients with breast cancer.The AUC value of the nomogram for predicting ALN metastasis in elderly patients with breast cancer is 0.744,and the calibration curve shows that actual probabilities and predicted probabilities are consistent,which verifies the accuracy and discriminating ability of the nomogram.Conclusion This nomogram can help predict the condition of ALN in elderly patients with breast cancer before operation,and then guide the choice of surgery plan.