3.0T MRI in predicting triple negative breast cancer / 中国医学影像技术

ABSTRACT

Objective To evaluate the related factors in 3.0T MR imaging features for predicting triple negative breast cancer (TNBC).Methods Totally 191 patients with breast cancer confirmed by histopathology were enrolled,among them 25 were TNBC (TNBC group),while the other 166 were non-TNBC (non-TNBC group).All patients underwent 3.0T MR examination.Patient age,tumor size,shape,margin,enhancement pattern,signal intensity on T2WI,time-intensity curve (TIC),ADC value,and peritumoral vessel between the two groups were compared.Univariate and multivariate Logistic regression analyses were used to predict the related factors.Results Compared with non-TNBC group,smooth margin (P=0.023),round mass (P=0.001),rim enhancement (P＜0.001),high signal intensity on T2WI (P=0.003),higher ADC value (P=0.022) and increased peritumoral vessel (P=0.046) were significantly associated with TNBC.Further multivariate Logistic regression analysis identified four independent risk factors of TNBC (all P＜0.05),including enhancement pattern,signal intensity on T2WI,ADC value and increased peritumoral vessel.The area under ROC curve of combination of the 4 factors was 0.840.Conclusion Rim enhancement of mass,high signal intensity on T2WI,higher ADC value,and increased peritumoral vessels are independent risk factors in predicting TNBC.