前哨淋巴结阳性乳腺癌患者非前哨淋巴结转移的影响因素及其预测模型建立

Influencing factors of non-sentinel lymph node metastasis in sentinel lymph node-positive breast cancer patients and establishment of predictive models

目的分析前哨淋巴结(sentinel lymph node, SLN)阳性的乳腺癌患者腋窝非前哨淋巴结(non-sentinel lymph node, NSLN)转移的危险因素,建立SLN阳性乳腺癌患者NSLN转移的列线图预测模型,为腋窝淋巴结清扫方案提供参考。方法 346例SLN活检阳性且行腋窝淋巴结清扫手术的女性乳腺癌患者,根据术后组织病理结果分为NSLN阳性组250例和NSLN阴性组96例,比较2组临床和病理资料;采用多因素logistic回归分析SLN阳性的乳腺癌患者NSLN转移的影响因素。根据影响因素,应用RStudio 3.4软件建立列线图预测模型,绘制ROC曲线,评估列线图预测模型预测SLN阳性乳腺癌患者NSLN转移风险的效能;采用Bootstrap法绘制校准预测曲线,进行内部验证以评价模型的校准度。结果 346例SLN活检阳性患者NSLN阳性率为72.3%,NSLN阴性率为27.7%。NSLN阳性组患者肿瘤最大直径>2~5 cm(53.6%)、有脉管癌栓(66.0%)、SLN阳性数目>2个(14.8%)、人类表皮生长因子受体-2(human epidermal growth factor receptor-2, HER-2)阳性(85.6%)比率高于NSLN阴性组(38.5%、32.3%、6.2%、66.7%)(P<0.05)。肿瘤最大直径>2~5 cm(OR=1.910,95%CI:1.126~3.281,P=0.017)、有脉管癌栓(OR=4.306,95%CI:2.560~7.396,P=0.001)、SLN阳性数目>2个(OR=2.933,95%CI:1.207~8.356,P=0.027)、HER-2阳性(OR=3.445,95%CI:1.874~6.398,P=0.001)是SLN阳性的乳腺癌患者NSLN转移的独立危险因素。以肿瘤最大直径、脉管癌栓、SLN阳性数目、HER-2构建列线图预测模型,列线图预测模型预测SLN阳性的乳腺癌患者NSLN转移的AUC为0.752(95%CI:0.696~0.808,P<0.001),预测效能良好。校准预测曲线与理想曲线贴合好。结论肿瘤最大直径>2~5 cm、有脉管癌栓、SLN阳性数目>2个、HER-2阳性为SLN阳性的乳腺癌患者发生NSLN转移的独立危险因素;以肿瘤最大直径、脉管癌栓、SLN阳性数目、HER-2建立的列线图模型对预测NSLN转移和制定腋窝淋巴结清扫方案有指导意义。

Objective To analyze the risk factors of axillary non-sentinel lymph node(NSLN) metastasis in patients with sentinel lymph node(SLN)-positive breast cancer patients, and establish a nomogram prediction model for NSLN metastasis in SLN-positive breast cancer patients to provide a reference for the axillary lymph node dissection program. Methods Totally 346 female breast cancer patients with positive SLN biopsy and axillary lymph node dissection were divided into NSLN positive group(n=250) and NSLN negative group(n=96) based on postoperative histopathological results. The clinical and pathological data were compared between two groups. Multivariate logistic regression was used to analyze the influencing factors of NSLN metastasis in SLN-positive breast cancer patients. Based on the influencing factors, RStudio 3.4 software was used to establish a nomogram prediction model. ROC was drawn to evaluate the efficacy of the nomogram prediction model on the prediction of the risk of NSLN metastasis. Bootstrap method was used to draw calibration prediction curve, and internal verification was performed to evaluate the calibration of the model. Results The positive rate of NSLN in 346 patients with positive SLN biopsy was 72.3%, and the negative rate of NSLN was 27.7%. The percentages of the patients with tumor maximum diameter of >2-5 cm(53.6%), vascular tumor thrombus(66.0%), the number of positive SLN>2(14.8%), and positive human epidermal growth factor receptor-2(HER-2)(85.6%) in NSLN-positive group were higher than those in NSLN-negative group(38.5%, 32.3%, 6.2%, 66.7%)(P<0.05). Tumor maximum diameter of >2-5 cm(OR=1.910, 95%CI: 1.126-3.281, P=0.017), vascular tumor thrombus(OR=4.306, 95%CI: 2.560-7.396, P=0.001), the number of positive SLN >2(OR=2.933, 95%CI: 1.207-8.356, P=0.027), and positive HER-2(OR=3.445, 95%CI: 1.874-6.398, P=0.001) were the independent risk factors of NSLN metastasis.The nomogram prediction model was constructed based on the maximum diameter of the primary tumor,vascular tumor thrombus,the number of positive SLN and HER-2.The AUCvalue of nomogram prediction model for predicting NSLN metastasis risk was 0.752(95%CI:0.696-0.808,P<0.001),with a good predictive performance.The calibration prediction curve fitted well with the ideal curve.Conclusion Tumor maximum diameter of >2-5 cm,vascular tumor thrombus,SLN positive number >2,and positive HER-2 are the independent risk factors of NSLN metastasis in patients with SLN-positive breast cancer.The nomogram model established by tumor maximum diameter,vascular tumor thrombus,number of positive SLN and HER-2 has a guiding significance for predicting NSLN metastasis and formulating axillary lymph node dissection plan.