探讨乳腺癌非前哨淋巴结(NSLN)转移的危险因素并构建非前哨淋巴结转移的预测模型。

回顾性分析2016年1月至2019年6月接受前哨淋巴结活检(SLNB)且确诊前哨淋巴结(SLN)阳性，行腋窝淋巴清扫术(ALND)的95例乳腺癌患者的临床资料。应用SPSS 20.0软件对数据进行处理，计数资料用[例(%)]描述，其中连续变量使用秩和检验，分类变量使用χ²检验。对与NSLN转移相关的临床病理因素进行多因素Logistic回归分析，根据Logistic回归分析各变量的回归系数建立非前哨淋巴结转移风险预测模型计算每例患者NSLN转移的预测概率，通过描绘受试者工作特征曲线(ROC)并计算曲线下面积(AUC)从而来评估模型的预测能力。

通过单因素分析显示乳腺癌患者NSLN转移与肿瘤大小、肿瘤位置、淋巴血管是否受侵犯、SLN转移灶大小及SLN阳性率相关(P<0.05)；将相关因素纳入多因素Logistic回归分析中，结果显示，肿瘤大小、淋巴血管是否受侵犯、SLN阳性率为乳腺癌患者NSLN转移的独立危险因素。根据Logistic回归分析建立NSLN转移风险预测模型，绘制研究对象的ROC曲线，计算AUC为0.792(95%CI为0.651～0.934)；Hosmer-Lemeshow拟合优度检验P＝0.603。

肿瘤大小、淋巴血管是否受侵犯、SLN阳性率为乳腺癌患者非前哨淋巴结转移的独立危险因素，NSLN转移风险预测模型对NSLN转移患者具有较高的预测价值，可辅助临床医师术前判断，选择合理的术式。

To explore the risk factors of non-sentinel lymph node (NSLN) metastasis in breast cancer and to construct a predictive model of non-sentinel lymph node metastasis.

The clinical data of 95 patients were analyzed retrospectively, who were with breast cancer and underwent sentinel lymph node biopsy (SLNB) and were diagnosed with positive sentinel lymph node (SLN) and underwent axillary lymph node dissection (ALND). Statistical analysis were performed by using SPSS 22.0 software. Count data were expressed as % and the rank sum test was used for continuous variables and the χ² test was used for categorical variables. Multi-factor logistic regression analysis were performed on the clinicopathological factors related to NSLN metastasis. According to the regression coefficient of each variable of logistic regression analysis, a non sentinel lymph node metastasis risk prediction model was established to calculate the prediction probability of each patient's NSLN metastasis. The prediction ability of the model was evaluated by depicting the receiver operating characteristic curve (ROC) and calculating the area under the curve (AUC).

Univariate analysis showed that NSLN metastasis in breast cancer patients was related to tumor size, tumor location, whether lymph vessels were invaded, size of SLN metastasis and SLN positive rate (P<0.05); Relevant factors were included in the multivariate logistic regression analysis. The results showed that tumor size, whether lymphatic vessels were invaded, and SLN positive rate were independent risk factors for NSLN metastasis in breast cancer patients. According to Logistic regression analysis, establish NSLN transfer risk prediction model. The ROC curve showed the AUC of 0.792 (95%CI was 0.651~0.934); the Hosmer-Leme show goodness-of-fit test P=0.603.

Tumor size, whether lymph vessels are invaded, and SLN positive rate are independent risk factors for non-sentinel lymph node metastasis in breast cancer patients. The NSLN metastasis risk prediction model has a higher predictive value for patients with NSLN metastasis. It could help clinician fpr preoperative evaluation so as to choose the reasonable operation method.