Correlation between Multiparametric Magnetic Resonance Imaging (MRI) Descriptors and Breast Cancer Molecular Subtypes.

Background and Objectives: Breast cancer is a variable group of diseases with different genotypic and phenotypic subtypes. Identifying the breast cancer molecular subtype is important in directing the type of treatment. The aim of this work: Our study aimed to evaluate the role of multiparametric MRI in prediction of molecular subtypes of breast cancer. Patients and Methods: A retrospective study included 135 female patients with pathologically proven invasive ductal carcinoma (IDC). Clinicopathological findings, morphological features, and dynamic contrast enhanced MRI (DCE-MRI) features were evaluated and correlated with the breast cancer molecular subtypes based on Immunohistochemistry classification. Results: We found that the most common histological grade (61.5%) was grade I &II (low grade) of invasive ductal carcinoma (IDC). There were 110 (81.5%) masses and 25 (18.5%) non-mass lesions. The luminal A-like (Lum A-like) was the most common subtype seen in 57 (42.2%), then luminal B-like (Lum B-like) in 46 (34.1%), human epidermal growth factor receptor2 (HER2) positive in 14 (10.4%), and triple-negative (TN) was the least common subtype seen in 18 (13.3%) of the lesions. Irregular or Spiculated mass margins and irregular shape were significantly associated with Luminal A-like subtype (87.7%&80.7%) and Luminal B-like (85%&100%), respectively (P=0.002). Larger mass size, multifocality, and non-mass enhancement were reliably associated with HER2-enriched subtype (78.6%,50% & 50%) respectively. The mass with oval shape, circumscribed margin, intra-tumoral high signal intensity on T2WI, rim enhancement, and higher histological grade was highly significant detected in TN breast cancer (P<0.001). The edema pattern was not significantly related to any molecular subtype. Conclusions: Morphological features of breast cancer, intra-tumoral T2WI signal intensity, the pattern of enhancement determined by multiparametric MRI, as well as the histological grade of the tumor could be helpful to predict and differentiate breast cancer molecular subtypes and hormone receptors. [ABSTRACT FROM AUTHOR]