肿块型乳腺癌CBBCT特征与Ki-67、HER-2表达的相关性研究

Cone-Beam Breast CT Features of Mass Breast Cancers:Correlation with Ki-67 and HER-2 Expression

目的探讨肿块型乳腺癌锥光束乳腺计算机断层扫描(CBBCT)特征与Ki-67、HER-2表达的相关性。方法搜集行CBBCT检查并经病理证实的141例肿块型乳腺癌患者的影像资料,利用单因素分析CBBCT特征与Ki-67、HER-2表达之间的关系,多因素Logistic回归分析评估联合特征区分Ki-67、HER-2表达的能力,采用受试者工作特征曲线评价预测效能。结果Ki-67表达与病灶数、肿块直径、毛刺、钙化、病灶周围血管增多、相关征象、强化程度显著相关(P<0.05),HER-2表达与钙化存在及钙化形态显著相关(P<0.05)。多因素Logistic回归分析显示,直径2~5 cm(OR=7.214)及直径>5 cm(OR=8.541)、强化程度为51~100 HU(OR=5.391)及101~150 HU(OR=13.657)与Ki-67表达水平有关(曲线下面积=0.801),粗糙不均质钙化(OR=0.370)、细小多形或细线样或细线分支状钙化(OR=0.145)、相关征象(OR=0.426)及毛刺(OR=2.036)与HER-2表达水平有关。结论CBBCT特征与Ki-67、HER-2的表达具有一定的相关性,联合CBBCT特征可能有助于预测乳腺癌Ki-67、HER-2表达,进而有助于预判患者的分子亚型和预后预测。

Objective To investigate the association of cone-beam breast CT(CBBCT)features,Ki-67 and HER-2 expression in mass breast cancer.Methods The imaging data of 141 patients with mass breast cancer who underwent CBBCT examination in our hospital and confirmed by pathology were collected.The association between the CBBCT features and the expression of Ki-67 and HER-2 was analyzed using univariate analysis.Multivariate Logistic regression analysis was performed to evaluate the ability of combined imaging features to discriminate the expression of Ki-67 and HER-2.ROC curve was used to evaluate prediction performance.Results The expression of Ki-67 in mass breast cancer was significantly correlated with the number of lesions,tumor diameter,spiculation,calcifications,increased peripheral vascularity of lesion,other findings and degree of lesion enhancement(all P<0.05).The expression of HER-2 was significantly correlated with calcifications and calcifications’morphology(all P<0.05).A multivariate logistic regression model showed that tumor diameter 2-5 cm(OR=7.214)and tumor diameter>5 cm(OR=8.541)、degree of lesion enhancement 51-100 HU(OR=5.391)and 101-150 HU(OR=13.657)were associated with the expression of Ki-67(AUC=0.801).Coarse heterogeneous calcification(OR=0.370)、fine pleomorphic or fine linear or linear branching calcification(OR=0.145),other findings(OR=0.426)and spiculation(OR=2.036)were associated with the expression of HER-2.Conclusion CBBCT features may have certain correlation with Ki-67 and HER-2 expression.Combined with CBBCT features may help to predict the expression of Ki-67 and HER-2 in breast cancer,which could in turn help to predict the molecular subtypes and prognosis of patients.