Clinical Study Of Multiparametric MRI Imaging Features For Predicting Molecular Subtypes Of Breast Cancer

Background and Objective Worldwide,breast cancer is the leading type of cancer in women,the first disease is high,and the death rate is second.Currently,breast cancer is regarded as a highly heterogeneous disease with multiple biological subtypes.Invasive breast cancer has been classified into four molecular subtypes using gene expression analysi: Luminal A,Luminal B,HER-2 overexpression,and triple negative.Each molecular subtype correlates with different biological behavior and clinical outcome,response to hormonal treatment and/or chemotherapy.However,to date the information on molecular subtypes has to be obtained by invasive tissue sampling.But it also has some limitations,such as disagreements in receptor status between biopsy and surgical specimen.Magnetic resonance imaging（MRI）of the breast is increasingly used as an adjunct to mammography and ultrasonography to improve the detection and characterization of primary and recurrent breast cancers.Dynamic contrast-enhanced magnetic resonance imaging（DCE-MRI）technique can reflect the characteristics of blood flow and microvascular permeability.Diffusion weighted imaging（DWI）can identify the function and metabolic state of tumor tissues and quantify the dispersion of water molecules through apparent diffusion coefficient（ADC）values.Recent studies showed that MRI is likely to predict the clinical,histological and molecular features of tumoral lesion.Analysis of breast MR images could potentially help the decision-making process of treatment at the early stage of diagnosis.This study aimed to investigate the correlation between multiparametric MRI imaging features and molecular subtypes of breast cancer and to predict the molecular staging of breast cancer by MRI features and some quantitative parameters.Materials and Methods Collection and retrospective analysis of 160 patients with confirmed breast cancer in anqing hospital affiliated to anhui medical university from January 2019 to April 2021.The patients divided into four groups according to the results of molecular typing,Luminal A（44 cases）,Luminal B（66 cases）,HER-2overexpression（26 cases）,and triple negative（24 cases）.According to the 2013 edition of breast imaging report and data system（BI-RADS）,the MRI features,ADC and early enhancement rate（EER）of different molecular subtypes of breast cancer were compared.The receive operating characteristic（ROC）curve was used to evaluate the efficiency of ADC and EER in differentiating HER-2 overexpression subtypes.Results There were statistically significant differences in the size,type,the shape,edge,and internal enhancement of mass-like lesion,peritumoral edema and peritumoral vascular among different molecular subtypes of breast cancer（P < 0.05）.HER-2overexpression subtype accounted for a higher proportion（8/26,30.78%）of non-mass type breast cancer compared with the other subtypes.All 24 triple-negative breast cancers patients showed mass-like lessions,which were mostly round,with clear boundary and ring enhancement.HER-2 overexpression subtype and triple negative subtype are usually accompanied by peritumoral edema and peritumoral vascular.There was no significant correlation between type of time-signal intensity cueve,background parenchymal enhancement and molecular subtypes（P>0.05）.The ADC value and EER value of HER-2 overexpression type are higher than other subtypes,the differences were statistically significant（P<0.05）.There was a positive correlation between ADC value,EER value,and HER-2 overexpressed breast cancer,which was a significant independent predictor of HER-2 overexpression type（P<0.05）.Area under the curve（AUC）of ADC,EER and ADC+EER in the diagnosis of HER-2 overexpression breast cancer were 0.785,0.726,0.835 respectively.The comparison results showed that the AUC of ADC+EER was higher than that of EER,and the difference was statistically significant（P<0.05）.There were no significant differences in AUC between ADC and EER,ADC and ADC+EER（P>0.05）.Conclusion Different molecular subtypes of breast cancer have certain characteristics on MRI.The type,the shape,edge,and internal enhancement pattern of mass-like lesion on MRI might reflect specific gene expression profiles of breast cancers.According to the ADC and EER,the molecular classification can be preliminarily identified.ADC and EER were significant independent predictors of HER-2 overexpression type.ADC joint EER helps to predict HER-2 overexpression breast cancer with the best diagnostic performance.