Multiparametric MRI Features of Breast Cancer Molecular Subtypes

Area of Science:

• Radiology
• Oncology
• Medical Imaging

Background:

• Breast cancer (BC) molecular subtypes exhibit distinct clinical behaviors and treatment responses.
• Multiparametric breast MRI, incorporating T2-weighted imaging (T2WI), apparent diffusion coefficient (ADC) values, and dynamic contrast-enhanced (DCE) sequences, offers detailed tumor characteristics.
• Breast density and background parenchymal enhancement (BPE) are also considered imaging biomarkers.

Purpose of the Study:

• To correlate multiparametric breast MRI features with specific breast cancer molecular subtypes.
• To investigate the utility of MRI findings, including ADC values, breast density, and BPE, in differentiating BC subtypes.
• To assess the potential of MRI in predicting Luminal B HER2-negative breast cancer.

Main Methods:

• Retrospective analysis of 344 patients with multiparametric breast MRI (T2WI, ADC, DCE).
• Extraction of imaging features based on the latest Breast Imaging Reporting and Data System (BIRADS) lexicon.
• Assessment of inter-reader agreement for ADC measurements.

Main Results:

• Luminal A tumors were associated with specific histology, smaller size, and persistent kinetic curves.
• Luminal B HER2-negative tumors showed the lowest ADC values (0.77 × 10−3 mm2/s2), achieving 0.583 accuracy for subtype prediction.
• Triple-negative breast cancer (TNBC) presented with asymmetric BPE, round/oval masses, circumscribed margins, and rim enhancement. HER2-enriched BC were associated with the largest tumor size.

Conclusions:

• Multiparametric MRI features, including T2WI, ADC, and DCE, are associated with distinct breast cancer molecular subtypes.
• ADC values demonstrate potential as a predictive biomarker for Luminal B HER2-negative breast cancer.
• MRI offers valuable insights into the heterogeneity of breast cancer at the molecular level.