摘要
目的分析乳腺癌患者术后局部复发的危险因素,建立个体化预测风险模型,并验证模型的预测准确性。方法选取乳腺癌患者,应用整体随机划分法将数据分为建模组(n=300)和验证组(n=100),并统计一般临床资料,单因素和多因素Logistic回归分析乳腺癌患者术后局部复发的危险因素,并建立相关预测模型,行Bootstrap法内部验证模型。结果肿瘤大小(OR=3.796,95%CI:1.373~10.499)、腋淋巴结转移数目多(OR=8.142,95%CI:2.373~27.934)、肿瘤分期(OR=2.517,95%CI:1.224~6.360)、原发灶分期晚(OR=2.995,95%CI:1.338~6.703)、术后放疗(OR=5.349,95%CI:2.004~14.281)是乳腺癌患者术后局部复发的独立危险因素(P<0.05),基于此5项危险因素构建预测风险模型,R方程检验表明建模集和验证集的C-index指数分别为0.831(95%CI:0.793~0.854)和0.813(95%CI:0.782~0.845);校正曲线和标准曲线结果基本相似,表明该风险预测模型精准度和区分度较好。结论肿瘤大小、腋淋巴结转移数目多、肿瘤分期、原发灶分期晚、术后放疗是乳腺癌患者术后局部复发的独立危险因素,建立的风险预测模型可以详细、精准的预测乳腺癌患者术后局部复发的风险大小,临床应用价值较高。
Objective To analyze the risk factors of local recurrence in patients with breast cancer after operation,and to establish an individualized risk prediction model,and verify the accuracy of the model.Methods Breast cancer patients were selected,and the data were divided into modeling set(n=300)and validation set(n=100)by the method of global random division.The general clinical data were statistically analyzed.The risk factors of local recurrence of breast cancer patients were analyzed by univariate and multivariate logistic regression,and the related prediction model was established,and the bootstrap internal validation model was performed.Results Tumor size(OR=3.796,95%CI:1.373~10.499),large number of axillary lymph node metastasis(OR=8.142,95%CI:2.373~27.934),tumor stage(OR=2.517,95%CI:1.224~6.360),late primary tumor stage(OR=2.995,95%CI:1.338~6.703),postoperative radiotherapy(OR=5.349,95%CI:2.004~14.281)were independent risk factors for local recurrence of breast cancer patients(P<0.05).Based on these 5 risk factors,a predictive risk model was constructed.R-equation test showed that the c-index index of modeling set and validation set were 0.831(95%CI:0.793~0.854)and 0.813(95%CI:0.782~0.845)respectively.The results of calibration curve and standard curve are basically similar,indicating that the accuracy and differentiation of the risk prediction model are good.Conclusion Tumor size,the number of axillary lymph node metastasis,tumor stage,primary tumor stage,postoperative radiotherapy are independent risk factors for local recurrence of breast cancer patients.The established risk prediction model can predict the risk of local recurrence of breast cancer patients in detail and accurately,and has high clinical application value.
作者
宋竹清
全舒萍
朱宁
王玲
SONG Zhuqing;QUAN Shuping;ZHU Ning(Peking University Shenzhen Hospital,Shenzhen,518000)
出处
《实用癌症杂志》
2021年第8期1347-1351,共5页
The Practical Journal of Cancer
基金
广东省深圳市三名工程项目(编号:SZSM201612010)。
关键词
乳腺癌
局部复发
危险因素
风险模型
Breast cancer
Local recurrence
Risk factors
Risk model
