摘要
目的探讨术前外周血中性粒细胞与淋巴细胞的比值(NLR)和纤维蛋白原与白蛋白的比值(FAR)对乳腺癌患者预后的影响。方法回顾性分析2013年1月1日至2015年12月31日承德医学院收治的995例乳腺癌患者的临床资料,分为复发组(n=121)和非复发组(n=874),比较2组的临床病理特征及术前外周血检验指标,评价术前NLR和FAR对患者预后的影响。患者的年龄及术前外周血检验指标为定量资料,符合偏态分布,用M(P 25~P 75)表示,用秩和检验进行2组间比较。用字2检验比较2组患者的其他临床病理特征。绘制受试者操作特征(ROC)曲线,评价NLR、FAR的诊断效能,确定最佳诊断界值。采用Kaplan-Meier方法绘制高、低NLR、FAR患者的无复发生存曲线,并用Log-rank法进行比较。采用Cox比例风险逐步回归模型分析影响患者预后的因素。结果FAR的最佳诊断界值为0.079,ROC曲线下面积(AUC)为0.778,95%CI为0.733~0.823,敏感度为66.9%(81/121),特异度为78.1%(683/874),阳性预测值为0.298,阴性预测值为0.945,阳性似然比为3.597,阴性似然比为0.407。NLR的最佳诊断界值为2.042,其AUC为0.863,95%CI为0.733~0.823,敏感度为79.3%(96/121),特异度为80.3%(702/874),阳性预测值为0.346,阴性预测值为0.959,阳性似然比为4.241,阴性似然比为0.225。NLR≥2.042与NLR<2.042患者和FAR≥0.079与FAR<0.079的患者无复发生存比较,差异均具有统计学意义(x^(2)=19.512,P<0.001;x^(2)=28.125,P<0.001)。单因素分析结果显示:复发组与非复发组患者的体质量指数、原发肿瘤直径、腋窝淋巴结转移、HER-2表达、脉管侵犯、组织学分级、术前中性粒细胞、淋巴细胞、纤维蛋白原、白蛋白水平及术前NLR、FAR比较,差异均有统计学意义(x^(2)=15.255、32.096、134.504、8.596、14.065、8.970,Z=-14.600、-2.760、-3.055、-11.668、-12.942,-9.927,P均<0.050)。多因素分析结果显示:原发肿瘤直径>2 cm、腋窝淋巴结转移、有脉管侵犯、组织学2~3级、HER-2阳性、FAR≥0.079、NLR≥2.042均为影响乳腺癌患者预后的独立危险因素(HR=2.347,95%CI:1.449~3.801,P=0.001;HR=4.667,95%CI:3.061~7.112,P<0.001;HR=1.994,95%CI:1.354~2.955,P=0.001;HR=1.676,95%CI:1.139~2.467,P=0.009;HR=1.586,95%CI:1.090~2.307,P=0.016;HR=2.214,95%CI:1.459~3.359,P<0.001;HR=6.491,95%CI:4.167~10.113,P<0.001)。结论术前NLR、FAR可作为预测乳腺癌预后的标志物。
Objective To investigate the effect of neutrophil-to-lymphocyte ratio(NLR)and fibrinogen-to-albumin ratio(FAR)of preoperative peripheral blood on the prognosis of breast cancer patients.Methods A total of 995 breast cancer patients admitted to Chengde Medical College from January 1,2013 to December 31,2015 were enrolled in a retrospective study.They were divided into recurrence group(n=121)and non-recurrence group(n=874).The clinicopathological characteristics and preoperative parameters in peripheral blood test were compared between two groups to evaluate the effect of preoperative NLR and FAR on the prognosis of patients.The patients’age and preoperative parameters in peripheral blood test were quantitative data of the skew distribution,so they were expressed as M(P25-P75)and compared between two groups using the rank sum test.The other clinicopathological characteristics were compared between two groups usingx^(2) test.The receiver operating characteristic(ROC)curve was drawn to evaluate the diagnostic efficacy of NLR and FAR and determine the optimal cut-off value.The Kaplan-Meier method was used to draw the recurrence-free survival curves of patients with high and low NLR/FAR,and the log-rank method was used for comparison.The Cox proportional hazards stepwise regression model was used to analyze the factors affecting the prognosis of patients.Results The cut-off value of FAR was 0.079,area under the ROC curve(AUC)0.778,95%CI 0.733-0.823,sensitivity 66.9%(81/121),specificity 78.1%(683/874),positive predictive value 0.298,negative predictive value 0.945,positive likelihood ratio 3.597 and negative likelihood ratio 0.407.The cut-off value of NLR was 2.042,AUC 0.863,95%CI 0.733-0.823,sensitivity 79.3%(96/121),specificity 80.3%(702/874),positive predictive value 0.346,negative predictive value 0.959,positive likelihood ratio 4.241 and negative likelihood ratio 0.225.The recurrence-free survival showed a significant difference between patients with NLR≥2.042 and NLR<2.042,between patients with FAR≥0.079 and FAR<0.079(x^(2)=19.512,P<0.001;x^(2)=28.125,P<0.001).The results of univariate analysis showed that body mass index,primary tumor diameter,axillary lymph node metastasis,HER-2 expression,vascular invasion,histological grade,preoperative levels of neutrophils,lymphocytes,fibrinogen and albumin and preoperative NLR and FAR all presented a significant difference(x^(2)=15.255,32.096,134.504,8.596,14.065,8.970,Z=-14.600,-2.760,-3.055,-11.668,-12.942,-9.927,all P<0.050).The results of multivariate analysis showed that primary tumor diameter>2 cm,axillary lymph node metastasis,vascular invasion,histological grade 2 to 3,HER-2 positive,FAR≥0.079 and NLR≥2.042 are all independent risk factors affecting the prognosis of breast cancer patients(HR=2.347,95%CI:1.449-3.801,P=0.001;HR=4.667,95%CI:3.061-7.112,P<0.001;HR=1.994,95%CI:1.354-2.955,P=0.001;HR=1.676,95%CI:1.139-2.467,P=0.009;HR=1.586,95%CI:1.090-2.307,P=0.016;HR=2.214,95%CI:1.459-3.359,P<0.001;HR=6.491,95%CI:4.167-10.113,P<0.001).Conclusion Preoperative NLR and FAR can be used as a marker to predict the diagnosis of breast cancer prognosis.
作者
张宏旭
牛梦晔
张炳洲
郑骞
牛旭鹏
吴振宇
胡大为
Zhang Hongxu;Niu Mengye;Zhang Bingzhou;Zheng Qian;Niu Xupeng;Wu Zhenyu;Hu Dawei(Department of Breast Surgery,Affiliated Hospital of Chengde Medical University,Chengde 067000,China;Department of General Surgery,First Hospital of Hebei Medical University,Shijiazhuang 050000,China)
出处
《中华乳腺病杂志（电子版）》
CAS
CSCD
2021年第4期206-213,共8页
Chinese Journal of Breast Disease(Electronic Edition)
