浸润性乳腺癌改良根治术后化疗患者复发转移的影响因素及预测模型建立
The influencing factors of recurrence and metastasisin invasive breast cancer patients undergoing chemo-therapy after modified radical surgery and the establishment of prediction model
狄淑群 1潘华 1狄挺松 2王水3
作者信息
- 1. 213300 溧阳,溧阳市人民医院甲乳外科
- 2. 213399 溧阳,溧阳市中医医院甲乳外科
- 3. 210029 南京,江苏省人民医院乳腺科
- 折叠
摘要
目的 研究浸润性乳腺癌(IBC)改良根治术后化疗患者复发转移的影响因素并建立预测模型.方法 选取行改良根治术后并进行化疗的复发转移的女性IBC患者115例作为复发组,于术后进行化疗且未发生复发转移的260例女性IBC患者中以随访时间为匹配因素按1∶1选取115例作为对照组,采集两组人口学特征、肿瘤信息、治疗过程及随访结果等资料并进行比较,通过配对Logistic回归分析术后复发转移的影响因素并建立预测模型.另随机选取行改良根治术的女性IBC患者102例作为验证组,采用Hosmer-Lemeshow拟合度检验和受试者工作特征(ROC)曲线分析模型预测术后复发转移的校准度和区分度.结果 配对Logistic回归分析显示,肿瘤直径＞5 cm、组织学分级高、腋窝淋巴结转移≥4枚、ER/PR均阴性以及未完成既定周期化疗为术后复发转移的独立危险因素(P＜0.05),术后靶向治疗为保护因素(P＜0.05);预测模型为logit(P)=0.723×(肿瘤直径＞5 cm)+0.515×(组织学分级)+0.202×(腋窝淋巴结转移)+1.565×(ER/PR均阴性)-0.791×(术后靶向治疗)+0.980×(未完成既定周期化疗),Hosmer-Lemeshow拟合度检验显示具有良好校准度(P＞0.05);Logistic回归模型预测术后复发转移的AUC为0.789(95％CI:0.729～0.849),敏感度为80.00％,特异度为73.90％.结论 肿瘤直径＞5 cm、组织学分级高、腋窝淋巴结转移数量≥4枚、ER/PR均阴性、未完成既定周期化疗为IBC患者改良根治术后复发转移的独立危险因素,术后靶向治疗为保护因素,据此建立的Logistic回归模型可为预测患者术后复发转移风险提供有效参考信息.
Abstract
Objective To study the influencing factors of recurrence and metastasis in patients with invasive breast cancer(IBC)undergoing chemotherapyafter modified radical surgery and establish a prediction model.Methods A total of 115 female IBC patients with recurrence and metastasis who underwent chemotherapy after modified radical surgery were selected as the recur-rence group,and another 115 patientscollected from 260 female IBC patients without recurrence and metastasis who underwent chemo-therapy after surgerywere enrolled as the control group according to the matching factor of follow-up timein 1∶1 ratio.The data such as de-mographic characteristics,tumor information,treatment process and follow-up results were collected and compared between the two groups.The influencing factors of postoperative recurrence and metas-tasis were analyzed by paired Logistic regression and a prediction model was established.In addition,102 female IBC patients who un-derwent modified radical surgery were randomly selected as the verifi-cation group.Hosmer-Lemeshow fit test and receiver operating charac-teristic(ROC)curve analysis model were used to predict the calibration degree and differentiation of postoperative recurrence and me-tastasis.Results Paired Logistic regression analysis showed that tumor diameter＞5 cm,high histological grade,axillary lymph node metastases≥4,negative ER/PR and failure to complete the prescribed cycle of chemotherapy were independent risk factors for postoper-ative recurrence and metastasis(P＜0.05),and postoperative targeted therapy was protective factor(P＜0.05).The prediction model was logit(P)=0.723×(tumor diameter＞5 cm)+0.515×(histological grade)+0.202×(axillary lymph node metastasis)+1.565×(ER/PR negative)-0.791×(postoperative targeted therapy)+0.980×(chemotherapy did not complete the prescribed cycle).Hosmer-Leme-show fit test showed good calibration(P＞0.05).The AUC of postoperative recurrence and metastasis predicted by Logistic regression model was 0.789(95％CI:0.729-0.849),sensitivity was 80.00％,and specificity was 73.90％.Conclusion Tumor diameter＞5 cm,high histological grade,number of axillary lymph node metastasis≥4,negative ER/PR,and failure to complete the prescribed cycle of chemotherapy are independent risk factors for recurrence and metastasis of IBC patients after modified radical surgery,and postopera-tive targeted therapy is a protective factor.The established Logistic regression model can provide effective reference information for pre-dicting the risk of postoperative recurrence and metastasis.
关键词
浸润性乳腺癌/改良根治术/复发/影响因素/预测/Logistic回归模型Key words
invasive breast cancer/modified radical surgery/recurrence/influencing factors/predict/Logistic regression model引用本文复制引用
出版年
2024
NETL
NSTL
万方数据