基于MRI影像特征联合病理和生物学指标预测乳腺浸润性导管癌脉管浸润状态

Prediction of lymphovascular invasion state in breast invasive ductal carcinoma based on MRI features combined with pathological and biological indicators

目的:探讨磁共振(MRI)影像特征、临床病理及生物学指标在预测乳腺浸润性导管癌脉管浸润(LVI)中的应用价值。方法:回顾性纳入经病理证实的325例乳腺浸润性导管癌患者,按照脉管内有无癌栓分为LVI阳性组(161例)和LVI阴性组(164例)。观察乳腺浸润性导管癌的MRI形态学特征和血流动力学参数。采用独立样本χ^(2)检验、t检验或Fisher确切概率法及Mann-Whitney U检验统计学方法,分析两组患者MRI影像特征、临床病理及生物学指标组间的差异,采用Logistic多因素回归分析脉管浸润的独立危险因素。结果:临床病理及生物学指标单因素分析结果显示:有无脉管浸润在病理肿瘤大小、组织学分级、前哨淋巴结转移、Ki-67指标、分子分型方面有显著差异,具有统计学意义(P值分别为0.000、0.005、0.000、0.000、0.000),而在年龄、ER状态、PR状态、Her-2状态方面差异均无统计学意义(P均>0.05)。MRI影像特征单因素分析结果显示:有无脉管浸润在MRI肿瘤大小、MRI腋窝淋巴结转移方面有差异,具有统计学意义(P值分别为0.000、0.002),而在乳腺密度、病灶类型、肿块形态、肿块边界、毛刺征、TIC类型方面比较,差异均无统计学意义(P均>0.05)。将单因素分析中有统计意义的指标,进一步纳入多因素Logistic回归分析,结果显示前哨淋巴结转移(OR=0.081,95%CI:0.031~0.213,P=0.000)、Ki-67高表达(OR=1.375E-8,95%CI:2.436E-9~7.767E-8,P=0.000)、分子分型Her-2过表达型(OR=12.148,95%CI:1.843~80.065,P=0.009)和磁共振肿瘤大小≥2 cm(OR=0.341,95%CI:0.126~0.922,P=0.034)是乳腺浸润性导管癌脉管浸润的独立危险因素。结论:乳腺浸润性导管癌脉管浸润与病理肿瘤大小、组织学分级、前哨淋巴结转移、Ki-67指标、分子分型、MRI肿瘤大小及MRI腋窝淋巴结转移有关。前哨淋巴结转移、Ki-67高表达、分子分型Her-2过表达型和磁共振肿瘤大小≥2 cm是乳腺浸润性导管癌脉管浸润的独立危险因素。基于MRI影像特征、临床病理及生物学指标有助于预测乳腺浸润性导管癌脉管浸润状态。

Objective:To investigate the application value of magnetic resonance imaging(MRI)features,clinicopathological and biological indicators in the prediction of lymphovascular invasion(LVI)in breast invasive ductal carcinoma.Methods:A total of 325 patients with breast invasive ductal carcinoma confirmed by pathology were retrospectively included.They were divided into LVI positive group(161 cases)and LVI negative group(164 cases)according to the presence or absence of cancer embolus in the vasculature.The MRI morphological features and hemodynamic parameters of breast invasive ductal carcinoma were observed.Independent sample χ^(2) test,t test,Fisher exact probability method and Mann-Whitney U test were used to analyze the differences between the two groups in MRI features,clinicopathological and biological indicators,and the independent risk factors of LVI were analyzed by Logistic multivariate regression.Results:The results of single factor analysis of clinicopathological and biological indicators showed there were significant differences in pathological tumor size,histological grade,sentinel lymph node metastasis,Ki-67 index,and molecular classification between LVI positive group and LVI negative group(P=0.000,0.005,0.000,0.000,0.000).There were no significant differences in age,ER status,PR status and Her-2 status(all P>0.05).The results of single factor analysis of MRI characteristics showed there were statistically significant differences in MRI tumor size and MRI axillary lymph node metastasis between LVI positive group and LVI negative group(P=0.000,0.002),while there were no statistically significant differences in breast density,lesion type,tumor morphology,tumor boundary,burr sign and TIC type(all P>0.05).Multivariate Logistic regression analysis further included the statistically significant indicators in the univariate analysis,and the results showed that sentinel lymph node metastasis(OR=0.081,95%CI:0.031~0.213,P=0.000),high expression of Ki-67(OR=1.375E-8,95%CI:2.436E-9~7.767E-8,P=0.000),molecular classification Her-2 overexpression type(OR=12.148,95%CI:1.843~80.065,P=0.009),MRI tumor size≥2 cm(OR=0.341,95%CI:0.126~0.922,P=0.034)were independent risk factors for LVI of breast invasive ductal carcinoma.Conclusion:The LVI of breast invasive ductal carcinoma is related to pathological tumor size,histological grade,sentinel lymph node metastasis,Ki-67 index,molecular classification,MRI tumor size and MRI axillary lymph node metastasis.Sentinel lymph node metastasis,high expression of Ki-67,molecular typing Her-2 overexpression and MRI tumor size≥2 cm were independent risk factors for LVI of breast invasive ductal carcinoma.MRI features,clinicopathological and biological indicators are helpful to predict the state of LVI in breast invasive ductal carcinoma.