Correlation Between Basal-like Breast Carcinoma And The Features Of BI-RADS-US

Objective:To investigate the correlation between basal-like breast carcinoma(BLBC)and the features of Breast Imaging Reporting and Data System of Ultrasonographic(BI-RADS-US)which can explore the biological characteristics of BLBC and provide abasis for BLBC clinical diagnosis and treatment.Methods:75 cases(76 masses)of BLBC,17 cases(18 masses)of non-basal like triple negative breast cancer(non-BLBC TNBC),64 cases(72 masses)of the non-triple negative breast cancer(non-TNBC)were reviewed for the features of BI-RADS-US and pathological data which were confirmed by pathology,including 32 BLBCs,10non-BLBC TNBCs and 72 non-TNBCs for ultrasound strain elastography.Chi-square Test and Fisher's exact test was applied to comparing BLBC respectively with nonBLBC TNBC,non-TNBC.Subsequently,model and equation for predicting BLBC was established by using binary logistic regression analysis and evaluate the model in terms of sensitivity,specificity,positive predictive value and negative predictive value.Results:The results derived from Chi-square Test or Fisher's Exact Test show that BLBC is statistically significant(P<0.05)in terms of mass shape(P=0.044),orientation(P=0.000),tumor necrosis or mucoid degeneration(P=0.044),histology grade(P=0.040),and Ki67 expression(P=0.008)compared with non-BLBC TNBC,and BLBC is statistically significant(P < 0.05)in terms of mass size(P=0.000),margin(P=0.000),interior echo(P=0.016),posterior echo(P=0.009),lymph node metastasis(P=0.042),BI-RADS-US classification(P=0.033),ultrasound elasticity score(P=0.020),tumor necrosis or mucoid degeneration(P=0.000),histology grade(P=0.000),and Ki67 expression(P=0.002).The binary logistic regression model: Logit(P)=-3.516+2.309×no spiculated margin+1.041×no angular margin+1.521×interior anechoic+0.931×the mass larger than 2cm.The sensitivity,specificity,positive predictive value and negative predictive value,BLBC were 59.2%,75.6%,67.2%,68.9%.Conclusion:1.There exist a correlation between BLBC and the features of BI-RADS-US in terms of mass size,shape,orientation,margin,interior echo,posterior echo,BI-RADSUS classification,ultrasound elasticity score.2.The binary logistic regression analysis model of BLBC shows that there are more closely relation between BLBC and the features of BI-RADS-US as no spiculated margin,no angular margin,interior anechoic,the mass larger than 2cm,which can hint BLBC ultrasonic imaging in some levels and to enhance the practicality for diagnosing breast cancer.3.The features of BI-RADS-US can speculate the pathology of BLBC histology grade,tumor necrosis or mucoid degeneration,Ki67 expression and so on.