影响子宫内膜癌淋巴结转移的临床及病理指标分析

Analysis of clinical and pathological indexes affecting lymph node metastasis in endometrial cancer

目的结合临床病理指标分析影响子宫内膜癌淋巴结转移的危险因素,构建模型并进行评估。方法回顾性分析收集2019年1月—2022年12月遵义医科大学附属医院及第二附属医院收治的子宫内膜癌患者,共160例。按照7∶3的比例分为训练集及验证集,运用二元Logistic回归分析确定子宫内膜癌淋巴结转移的独立危险因素,利用R软件构建列线图模型,并对预测模型进行评估。结果单因素分析显示术前影像学淋巴结转移、宫旁转移、附件累及、肿瘤直径、肌层浸润深度、病理类型、组织学分级、脉管间隙浸润、错配修复蛋白表达、中性粒细胞-淋巴细胞比值、术前CA125值、临床分期与子宫内膜癌淋巴结转移相关(P<0.05),其中术前影像学淋巴结转移阳性、肌层浸润深度≥1/2、错配修复缺陷、中性粒细胞-淋巴细胞比值≥2.19是子宫内膜癌患者淋巴结转移的独立危险因素(P<0.05)。据此构建的列线图预测模型,内部验证结果显示ROC曲线下面积为0.929,C-index为0.917(95%CI:0.915~0.919),校正曲线显示平均绝对误差为0.046,该模型具有较好的H-L拟合优度(χ2=4.507,P=0.808)。该模型通过验证组进行外部验证,结果显示ROC曲线下面积为0.870,该模型的准确度较好。结论术前影像学淋巴结转移阳性、肌层浸润深度≥1/2、错配修复蛋白表达、中性粒细胞-淋巴细胞比值≥2.19是子宫内膜癌患者淋巴结转移的独立危险因素,以此建立的预测模型具有较高的准确度及区分度,可据此采取个体化治疗方式及后续辅助治疗。

Objective Combining with clinical and pathological indicators,the risk factors affecting lymph node metastasis of endometrial cancer were analyzed,and the model was constructed and evaluated.Methods A retrospective analysis collected 160 patients with endometrial cancer admitted to Affiliated Hospital of Zunyi Medical University and The Second Affiliated Hospital of Zunyi Medical University from January 2019 to December 2022.They were divided into training set and validation set at a ratio of 7∶3,and binary logistic regression analysis was used to determine the independent risk factors for lymph node metastasis of endometrial cancer.The nomogram model was constructed by R software,and then the prediction model was evaluated.Results Univariate analysis showed that preoperative imaging lymph node metastasis,parauterine metastasis,fallopian tubes and ovaries involvement,tumor diameter,depth of muscular invasion,pathological type,histological grade,vascular space infiltration,MMR,NLR,preoperative CA125 value,and clinical stage were associated with lymph node metastasis of endometrial cancer(P<0.05).Among which preoperative imaging lymph node metastasis was positive,muscular invasion depth was≥1/2,dMMR,NLR≥2.19 was an independent risk factor for lymph node metastasis in patients with endometrial cancer(P<0.05).Based on the nomogram prediction model,the internal verification results show that the area under the ROC curve is 0.929,the C-index is 0.917(95%CI:0.915-0.919),and the correction curve shows that the average absolute error is 0.046,and the model has a good H-L goodness fit(χ2=4.507,P=0.808).The model is externally verified by the validation group,and the results show that the area under the ROC curve is 0.870,indicating that the accuracy of the model is good.Conclusion Positive lymph node metastasis in preoperative imaging,depth of muscular invasion≥1/2,dMMR and NLR≥2.19 are independent risk factors for lymph node metastasis in patients with endometrial cancer,and the prediction model established has high accuracy,thus individualized treatment and subsequent adjuvant therapy can be adopted.