Role of DCE-MR in predicting breast cancer subtypes.

OBJECTIVE: The purpose of this retrospective study is to find a correlation between dynamic contrast-enhanced MR features with histological, immunohistochemical and loco-regional characteristics of breast cancer. MATERIALS AND METHODS: A total of 149 patients with histopathologically confirmed invasive breast carcinoma underwent MR imaging. Histological analysis included: histological features (histological type, necrosis, vascular invasion and Mib-1), immunohistochemical characterization (immunophenotype, receptor status, HER2-neu and grading) and loco-regional characteristics (T and N). The kinetic MR features analyzed were: curve type, maximum enhancement, time to peak, wash-in and wash-out rate, brevity of enhancement and area under curve. RESULTS: MRI kinetic parameters and immunohistological features were compared using chi square test, two-tailed student t test and Anova test, with p = 0.05 level of significance. Vascular invasion was shown to be significantly related to time to peak (p = 0.02). The immunohistotype was shown to be significantly related with maximum enhancement (p = 0.05), time to peak (p = 0.04) and wash-in rate (p = 0.01). ER status correlates with maximum and relative enhancement (p = 0.004 and p = 0.028), wash-in rate (p = 0.0018) and area under curve (p = 0.006). PR status was significantly related to time to peak (p = 0.048) and wash-in rate (p = 0.05). CONCLUSION: Maximum enhancement absolute and relative, time to peak, wash-in rate and area under the curve significantly correlate with several prognostic factors, like ER status, immune profile and tumoral vascular invasion, and may predict the aggressiveness of the tumor.

Evaluation of Multifocality and Multicentricity With Breast Magnetic Resonance Imaging in Each Breast Cancer Subtype.

INTRODUCTION: The purpose of this study was to evaluate whether diagnostic performance of breast magnetic resonance imaging (MRI) for detection of multifocality and multicentricity (MFMC) of breast cancer (BC) can be influenced by different histotypes or immunophenotypes in newly diagnosed patients with breast cancer. MATERIALS AND METHODS: In this institutional review board-approved retrospective study, 289 patients who underwent both preoperative breast MRI and radical or modified mastectomy in our institution because of primary BCs were selected. Patients were stratified based on the pathologic report in 2 main histotypes and 5 immunophenotypes. By matching the radiologic report with the corresponding pathologic report for each patient, breast MRI performance for detection of MFMC were obtained in each histotype and immunophenotype and subsequently compared. RESULTS: Overall breast MRI sensitivity for MFMC detection was 88.1%, specificity was 80.0%, positive predictive value 82.1%, negative predictive value 85.8%, diagnostic accuracy 83.7%, and area under the curve 0.835. Breast MRI sensitivity for MFMC detection in triple-negative BC was 84.6% (P = .88), specificity 70.8% (P = .63), positive predictive value 61.1% (P = .02), negative predictive value 89.5% (P = .20), diagnostic accuracy 75.7% (P = .65), and area under the curve 0.777 (P = .87). CONCLUSION: Performance of breast MRI for the detection of MFMC are not influenced by the BC histotypes, in accordance with published literature. Conversely, the triple-negative immunophenotypes demonstrated lower performance, statistically significant only for positive predictive value (P = .02), for the detection of MFMC.