Promising Breakthrough: MRI Accurately Identifies Breast Cancer Subtypes
A prospective single‑centre study of 102 patients suggests multiparametric 3T breast MRI can non‑invasively distinguish luminal from non‑luminal invasive ductal carcinoma (IDC) with high accuracy, offering a potential imaging adjunct to histopathology when planning personalized treatment. Combining specific MRI features produced an area under the curve (AUC) of 0.92 for differentiating the two groups, with 87% sensitivity and 86% specificity, though the authors caution that external validation is required before routine clinical use.
Patients with newly diagnosed IDC underwent a pre‑treatment MRI protocol that included T1‑ and T2‑weighted sequences, subtraction imaging, dynamic contrast‑enhanced MRI (DCE‑MRI), diffusion‑weighted imaging (DWI) and MR spectroscopy (MRS). Tumours were classified by immunohistochemistry as luminal (A and B) or non‑luminal (HER2‑enriched and triple‑negative). Investigators assessed morphology plus semiquantitative and quantitative DCE‑MRI parameters, apparent diffusion coefficient (ADC) values from DWI, and choline peaks on MRS.
Of the 102 cases, 48 (47.1%) were luminal and 54 (52.9%) were non‑luminal. Luminal tumours more often appeared irregular, with spiculated margins and heterogeneous contrast enhancement. Non‑luminal tumours were more likely to show circumscribed margins, rim enhancement, perilesional oedema and axillary lymphadenopathy.
Quantitatively, non‑luminal cancers demonstrated higher Ktrans (a measure of contrast transfer across vessels), higher signal enhancement ratio (SER) and a steeper maximum slope of enhancement, alongside lower extracellular extravascular volume fraction (Ve) and lower ADC values. A choline peak on MR spectroscopy-a marker sometimes associated with increased cellular turnover-was observed more frequently in non‑luminal tumours.
On multivariable analysis, three MRI features emerged as independent predictors of non‑luminal disease: presence of perilesional oedema, low Ve and high SER. When combined, these parameters yielded the reported AUC of 0.92, with 87% sensitivity and 86% specificity for identifying non‑luminal IDC.
The authors note the study’s limitations, principally its single‑centre design and modest sample size, and recommend external validation before adopting these imaging biomarkers into routine practice. If validated, multiparametric MRI could help refine risk stratification, guide biopsy targeting or supplement pathology in planning subtype‑specific therapy. (Mohakud S et al., Clin Radiol. 2026; DOI:10.1016/j.crad.2026.107426.)
Original Source: https://www.emjreviews.com/radiology/news/multiparametric-mri-shows-promise-in-breast-cancer-subtyping/
Category:
Tags:
Publish Date: 2026-07-19 13:31:00