Application value of different model of dynamic contrast-enhanced magnetic resonance imaging in pathological grading of breast invasive ductal carcinoma

Abstract

Objective To explore the application value of different dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) models in pathological grading of breast invasive ductal carcinoma. Methods Forty-five cases of breast invasive ductal carcinoma confirmed by clinical pathology from June 2016 to October 2017 in Shanxi Provincial People's Hospital were enrolled in this study. Grade 1 and 2 invasive ductal carcinomas were classified into the low-grade group, grade 3 invasive ductal carcinomas was classified as the high-grade group. The whole patients underwent DCE-MRI scans. Volume transport constant (Ktrans) was obtained by using the Extended Tofts Linear model with population arterial input function (AIF). Then, the Reference Region model was used to obtain the reference region model volume transport constant (RRKtrans). The performances of histogram analysis of these two quantitative parameters in pathological grading of breast invasive ductal carcinoma were compared. Results The mean, 25 %, 50 %, 75 %, 90 % percentiles, kurtosis and skewness of RRKtransin in high grade group were (0.793±0.258)/min, (0.484±0.209)/min, (0.773± 0.277)/min, (1.066±0.351)/min, (1.322±0.406)/min, 2.647 (1.426, 3.679), 0.398 (0.297, 0.514) respectively, and the corresponding parameters in low grade group were (0.506±0.203)/min, (0.301±0.142)/min, (0.487±0.211)/min, (0.692±0.281)/min, (0.861±0.323)/min, 1.725 (0.779, 2.316), 0.258 (0.133, 0.302) respectively. There were significant differences between the two groups (all P < 0.05). The mean, 50 %, 75 %, 90 % percentiles of Ktrans in high grade groups were (0.099±0.034)/min, (0.110±0.033)/min, (0.132±0.045)/min, (0.140±0.047)/min respectively, and the corresponding parameters in low grade group were (0.067±0.030)/min, (0.082±0.067)/min, (0.096±0.059)/min, (0.113±0.074)/min respectively. There were significant differences between the two groups (all P < 0.05). RRKtrans was superior to Ktrans in distinguishing area under the curve (AUC) of receiver operating characteristic curve (ROC) of high and low grading of breast invasive ductal carcinoma. Conclusion RRKtrans obtained by Reference Region model and Ktrans obtained by Population AIF DCE-MRI have some values in pathological grading of breast invasive ductal carcinoma, but the performance of RRKtrans is superior to Ktrans. Key words: Breast neoplasms; Dynamic contrast-enhanced magnetic resonance imaging; Reference region model; Arterial input function