Related Experiment Video
Updated: Apr 12, 2026
Optimization of a Multiplex RNA-based Expression Assay Using Breast Cancer Archival Material
Published on: August 1, 2018
Breast cancer molecular subtype as a predictor of the utility of preoperative MRI
Richard Ha1, Brian Jin, Victoria Mango
11 Department of Radiology, Columbia University Medical Center, Herbert Irving Pavilion, 161 Fort Washington Ave, 10th Fl, New York, NY 10032.
Objective:
The purpose of this study was to discern whether breast cancer molecular subtype, a known prognostic indicator, can be used to select patients with the highest likelihood of having clinically significant additional findings on breast MRI.
Materials And Methods:
A database review from January 2010 through December 2013 identified 299 patients who underwent preoperative breast MRI with tumors classifiable into molecular subtypes. Subtypes were classified on the basis of immunohistochemical staining surrogates as luminal A (hormone receptor [ER or PR] positive, ERBB2 [formerly HER2 or HER2/neu] negative, luminal B (hormone receptor positive, ERBB2 positive), ERBB2 (hormone receptor negative, ERBB2 positive), or basal (hormone receptor and ERBB2 negative). Univariate and multivariate logistic regression analyses were used to determine the association between subtype and additional breast MRI findings, including multicentric or multifocal disease, contralateral disease, chest wall involvement, skin and nipple involvement, and internal mammary and axillary lymphadenopathy.
Results:
The subtype distribution was luminal A, 70.6% (211/299); luminal B, 14.1% (42/299); ERBB2, 5.4% (16/299); and basal, 10.0% (30/299). ERBB2 and luminal B sub-types were more often associated with multicentric disease (25.0% and 26.2%), multifocal disease (37.5% and 35.7%), and axillary disease (50.0% and 45.2%) than were luminal A cancers (multicentric disease, 10.9%; multifocal disease 20.4%; axillary disease, 22.7%) (p < 0.001). In multivariate analysis, after control for patient age, tumor size, and nuclear grade, patients with ERBB2-overexpressing tumors were 2.4 times as likely as patients with luminal A tumors to have multicentric disease (p = 0.016), 2.0 times as likely to have multifocal disease (p = 0.024), 1.7 times as likely to have skin and nipple involvement (p = 0.013), and 1.9 times as likely to have axillary disease (p = 0.011).
Conclusion:
Preoperative MRI may most benefit patients with tumors with ERBB2 overexpression because of the increased likelihood of the presence of additional disease.
