早期宫颈癌患者非前哨淋巴结转移的临床病理因素分析

Analysis between non-sentinel lymph node metastasis and clinicopathological features of early-stage cervical cancer.

目的:分析早期宫颈癌非前哨淋巴结转移的临床病理因素。方法:收集2017年1月至2021年11月在广西医科大学附属肿瘤医院诊断为Ia1期伴有淋巴脉管间隙浸润至Ib2期宫颈癌患者的临床病理资料。患者均行前哨淋巴结示踪术+系统的淋巴结清扫术。单因素分析临床病理特征及其与非前哨淋巴结转移的关系,多因素分析非前哨淋巴结转移的独立危险因素。采用受试者工作特征(ROC)曲线计算AUC及前哨淋巴结阳性率的最佳诊断阈值。构建列线图模型,通过ROC曲线和内部校准曲线评价模型性能及拟合优度。结果:肿瘤大小、宫颈浸润深度、淋巴脉管间隙浸润、阳性前哨淋巴结数、前哨淋巴结阳性率与非前哨淋巴结转移相关。SLN-PR(P=0.049)及淋巴脉管间隙浸润(P=0.042)是非前哨淋巴结转移的独立危险因素。SLN-PR>45%对非前哨淋巴结转移的诊断效率最高,AUC为0.723。ROC曲线和校准曲线结果均显示列线图模型具有较好预测性能和拟合优度。结论:SLN-PR、阳性前哨淋巴结数、肿瘤大小、宫颈浸润深度及淋巴脉管间隙浸润与非前哨淋巴结转移相关,SLN-PR及淋巴脉管间隙浸润是非前哨淋巴结转移的独立危险因素,SLN-PR>45%的临界值可预测非前哨淋巴结转移,基于这些因素建立的预测非前哨淋巴结转移风险的列线图模型有较高的准确性,有助于识别临床淋巴结阴性但非前哨淋巴结阳性的早期宫颈癌患者。

Objective:To evaluate the clinicopathological features of early-stage cervical cancer and its relationship with non-sentinel lymph node metastasis.Method:We retrospectively collected clinicopathological datas of cervical cancer patients clinically considered as Ia1 with lymphovascular space invasion to Ib2.All patients underwent sentinel lymph node mapping+systematic pelvic lymphadenectomy in department of Gynecologic of the Affiliated Tumor Hospital of Guang Xi Medical University from January 2017 to November 2021.Statistically analyzed the clinicopathological features of early cervical cancer and its relationship with non-sentinel lymph node metastasis.The independent risk factors of non-sentinel lymph node metastasis were analyzed by multivariate analysis.The ROC curve was used to count AUC and Optimal diagnostic threshold for sentinel node positive rate.Build nomograph model,and the model performance and goodness of fit were evaluated by the ROC curve and the internal calibration curve.Result:Univariate analysis showed a significant association between non-sentinel lymph node metastasis and size of primary tumor,the number of positive sentinel lymph nodes,depth of cervical invasion,lymphovascular space invasion,the ratio of positive sentinel lymph node(SLN-PR,number of metastatic sentinel lymph nodes/total number of resected sentinel lymph nodes×100%).SLN-PR(P=0.049)and lymphovascular space invasion(P=0.042)were independent risk factors for non-sentinel lymph node metastasis.The SLN-PR>45%had the highest diagnostic efficiency for non-sentinel lymph node metastasis,with the area under the ROC curve was 0.723.Both the ROC curve and the calibration curve results showed that the nomogram model had good prediction performance and goodness of fit.Conclusion:SLN-PR,number of positive sentinel lymph nodes,tumor size,cervical invasion depth,and lymphovascular space invasion correlates to the risk of non-sentinel lymph node metastasis.SLN-PR and lymphovascular space invasion are independent risk factors for non-sentinel lymph node metastasis.The cut-off value of SLN-PR>45%can predict non-sentinel lymph node metastasis.The nomogram model based on these factors for predicting the risk of non-sentinel lymph node metastasis has high accuracy,which may help identify the clinically node-negative but non-sentinel node-positive early cervical cancer patients.