| 王 贝,钱 瑶,徐 琪.结合超声特征的腋窝淋巴结阳性乳腺癌新辅助化疗后腋窝病理完全缓解的预测模型[J].肿瘤学杂志,2021,27(7):536-541. |
| 结合超声特征的腋窝淋巴结阳性乳腺癌新辅助化疗后腋窝病理完全缓解的预测模型 |
| Predictive Model of Axillary Pathological Complete Response After Neoadjuvant Chemotherapy for Axillary Lymph Node-positive Breast Cancer Combined with Ultrasound Features |
| 投稿时间：2020-12-20 |
| DOI：10.11735/j.issn.1671-170X.2021.07.B005 |
|
 |
| 中文关键词: 乳腺癌 新辅助化疗 腋窝淋巴结 预测模型 |
| 英文关键词:breast cancer neoadjuvant chemotherapy axillary lymph nodes prediction model |
| 基金项目: |
|
| 摘要点击次数: 913 |
| 全文下载次数: 265 |
| 中文摘要: |
| 摘 要：［目的］ 分析经空芯针穿刺活检证实腋窝淋巴结阳性乳腺癌患者新辅助化疗（neoadjuvant chemotherapy，NAC）后腋窝病理完全缓解(pathological complete response，pCR)率及其影响因素，并整合超声影像特征与已知的临床病理特征建立预测模型，为新辅助化疗后乳腺癌患者腋窝处理的降级提供信息。［方法］ 回顾性分析哈尔滨医科大学附属肿瘤医院2017年1月至2018年12月入院接受NAC的481例乳腺癌患者的临床病理资料及超声影像特征，使用Logistic回归模型对临床病理特征及超声特征与NAC后腋窝淋巴结pCR的关系进行单因素及多因素分析，采用多因素分析中具有独立预测作用的指标构建新辅助化疗后腋窝pCR的预测列线图，并采用受试者工作特征（receiver operating characteristic，ROC）曲线及Bootstrapping法对此模型进行验证与校准。［结果］ 在481例患者中有147例(30.6%)实现了腋窝pCR。 单因素分析显示分子分型、乳腺原发灶临床疗效、淋巴结皮髓质分界是否清晰、彩色多谱勒血流图是否存在血流信号、淋巴结长径、淋巴结短径与腋窝pCR相关。多因素分析显示分子分型、乳腺原发灶临床疗效、CDFI血流信号、淋巴结短径是腋窝pCR的独立预测因素。与单独使用临床病理特征的预测模型相比，该模型具有良好的识别性能（ROC曲线下面积，0.784 vs 0.694，P<0.001）。［结论］ 结合超声特征的腋窝淋巴结阳性乳腺癌新辅助化疗后腋窝pCR的预测模型提高了仅应用临床病理特征的模型的预测能力，为NAC后选择合适的患者进行侵入性较小的腋窝手术方式提供了参考依据。 |
| 英文摘要: |
| Abstract： ［Objective］ To determine the complete axillary pathological complete response(pCR) rate and its influencing factors of breast cancer patients with axillary lymph node positive confirmed by hollow needle biopsy after neoadjuvant chemotherapy(NAC)， and to establish a prediction model by integrating ultrasound imaging features with known clinicopathological features， so as to provide information for the degradation of axillary treatment of breast cancer patients after neoadjuvant chemotherapy. ［Methods］ The clinicopathological data and ultrasound imaging features of 481 breast cancer patients admitted to the Cancer Hospital Affiliated to Harbin Medical University from January 2017 to December 2018 were analyzed retrospectively. A binary Logistic regression model was used to analyze the relationship between clinicopathological characteristics and ultrasound characteristics and the pCR of axillary lymph nodes after NAC for univariate and multivariate analysis. The indicators with independent predictive effect in multivariate analysis were used to construct the predictive nomogram of axillary pCR after neoadjuvant chemotherapy， and the receiver operating characteristic(ROC) curve and bootstrapping method were used to verify and calibrate the model. ［Results］ Among 481 patients， 147(30.6%) achieved axillary pCR. Univariate analysis showed that molecular typing， clinical efficacy of primary breast lesions， whether the boundary between skin and medulla of lymph nodes was clear(P<0.001)， whether there was blood flow signal in color Doppler flow imaging(CDFI)， long diameter and short diameter of lymph nodes were related to axillary pCR. Multivariate analysis showed that molecular typing， clinical efficacy of primary breast lesions， CDFI blood flow signal and short diameter of lymph node were independent predictors of axillary pCR. Compared with the prediction model using clinicopathological features alone， this model has better recognition performance(area under ROC curve， 0.784 vs 0.694， P<0.001). ［Conclusion］ The prediction model of axillary pCR for axillary lymph node positive breast cancer after neoadjuvant chemotherapy combined with ultrasound features significantly improves the prediction ability of the model only using clinical and pathological features， and provides reference for selecting suitable patients for less invasive axillary surgery after neoadjuvant chemotherapy. |
|
在线阅读
查看全文 查看/发表评论 下载PDF阅读器 |