摘要
目的:基于正电子发射断层成像(PET)相关代谢参数及临床病理特征建立前哨淋巴结转移的预测模型。方法:回顾性分析梅州市人民医院2016年1月—2023年3月经手术病理诊断为乳腺癌,且术前行全身PET/CT检查的患者211例,收集其临床、病理及PET相关代谢参数等特征。分析乳腺癌原发灶临床病理特征及PET代谢参数与前哨淋巴结转移的关联,构建Logistic回归预测模型。结果:乳腺癌病灶PET可疑腋窝淋巴结转移、脉管癌栓、雌激素受体(ER)及孕激素受体(PR)表达在前哨淋巴结阳性组与阴性组间差异有统计学意义(P<0.05),而肿瘤最大径、肿瘤位置、最大标准化摄取值(SUVmax)、代谢肿瘤体积(MTV)、糖醇解总量(TLG)、病灶数量、病理类型、组织分级、神经侵犯、人表皮生长因子受体-2(HER-2)及细胞核相关抗原(Ki-67)等在两组间差异无统计学意义(P>0.05)。将单因素分析中差异有统计学意义的变量进行多因素Logistic回归分析,使用Logistic回归建立回归模型为Logit(P)=-0.437×脉管癌栓+4.685×可疑腋窝淋巴结转移。预测模型AUC为0.738(P<0.001,95%CI:0.664~0.812),敏感度及特异度分别为63.4%和74.4%。结论:乳腺癌原发灶PET是否有可疑腋窝淋巴结转移、脉管癌栓、ER及PR与前哨淋巴结转移有关,基于乳腺癌原发灶PET可疑腋窝淋巴结转移及脉管癌栓建立的预测模型对前哨淋巴结转移的预测有一定的价值,可能为临床提供一种无创的检查模式。
Objective:A model for predicting sentinel lymph node metastasis was established based on positron emission tomography(PET)-related metabolic parameters and clinicopathological characteristics.Methods:A retrospective analysis was conducted on 211 patients diagnosed with breast cancer through surgical pathology from January 2016 to March 2023 in Meizhou People's Hospital,who underwent whole-body PET/CT examinations prior to surgery.Clinical,pathological,and PET-related metabolic parameters were collected.The study analyzed the association between clinicopathological characteristics of the primary breast cancer lesion,PET metabolic parameters,and sentinel lymph node metastasis.A logistic regression predictive model was constructed using Broussonetia papyrifera.Results:Breast cancer primary lesion PET-suspicious axillary lymph nodes,vascular tumor thrombus,estrogen receptor(ER),and progesterone receptor(PR)showed statistically significant differences between the two groups(all P<0.05),while maxi-mum tumor diameter,SUVmax tumor location,maximum standardized uptake value(SUVmax),metabolic tumor volume(MTV),total lesion glycolysis(TLG),tumor location,number of lesions,pathological type,histological grade,neural inva-sion,Homo sapiens epidermal growth factor receptor 2(HER-2),and nuclear-associated antigen Ki-67(Ki-67)showed no statistically significant differences between the two groups(all P>0.05).Parameters with P<0.05 in the univariate analysis were subjected to multivariate logistic regression analysis,and the logistic regression model was established as Logit(P)=-0.437×vascular tumor thrombus+4.685×suspicious axillary lymph node metastasis.The predictive model AUC was 0.738(P<0.001,95%CI:0.664~0.812),with sensitivity and specificity of 63.4%and 74.4%,respectively.Con⁃clusion:PET findings of the primary breast cancer lesion regarding suspicious axillary lymph node metastasis,vascular tumor thrombus,ER and PR are associated with sentinel lymph node metastasis.The predictive model established based on PET-detected suspicious axillary lymph node metastasis and vascular tumor thrombus in primary breast cancer lesions has certain value in predicting sentinel lymph node metastasis,potentially providing a non-invasive examination modality for clinical practice.
作者
陈丹丹
楼云龙
林政
CHEN Dan-dan;LOU Yun-long;LIN Zheng(Department of Nuclear Medicine,Meizhou People's Hospital,Meizhou 514003,China)
出处
《中国现代普通外科进展》
2025年第9期697-701,共5页
Chinese Journal of Current Advances in General Surgery
关键词
乳腺癌
前哨淋巴结
PET/CT显像
预测
Breast cancer
PET/CT imaging
Sentinel lymph node metastasis
Prediction
