Magnetic Resonance Elastography in Breast Cancer: A Prospective Study on Optimal Imaging Frequency Selection and Differentiation of Molecular Subtypes.
Abstract
Background
MR elastography (MRE) has applications in breast cancer, but its optimal frequency and value in discriminating molecular subtypes remain unclear.
Purpose
To identify the preferable breast MRE frequency and evaluate MRE parameters in the precision diagnosis of breast cancer. STUDY TYPE Prospective. POPULATION One hundred fifty-four female patients (mean age 50 ± 10.67 years) were enrolled, including five with bilateral cancer, yielding 159 lesions. FIELD STRENGTH/SEQUENCE 3.0 T, fast gradient-echo for DCE-MRI, single-shot EPI for DWI/ADC, SE-EPI for multifrequency MRE. ASSESSMENT The image quality of 10 frequency MRE sets was assessed (including overall image quality, diagnostic confidence, artifact impact, and signal-to-noise ratio). Stiffness (c) and viscosity (φ) were measured at the optimal frequency and compared across clinicopathological subgroups. STATISTICAL TESTS Independent-sample t-tests or one-way analysis of variance, Mann-Whitney U or Kruskal-Wallis H tests, Friedman test, χ2 tests. Spearman's rank correlation. Univariate and multivariate analyses. Receiver operating characteristic (ROC) curves, the area under the curve (AUC). p < 0.05 was considered statistically significant.
Results
The 40-70 Hz multifrequency band demonstrated significantly superior image quality compared with all other bands. Subgroup analyses showed that c and φ differed significantly between ER-positive and ER-negative groups (c values: 2.54 (2.17-2.86) vs. 2.80 (2.37-3.30) m/s; φ values: 1.17 ± 0.21 vs. 1.24 ± 0.22 radians) and across groups with different Ki-67 expression levels (c values: 2.54 ± 0.54 vs. 2.77 ± 0.57 m/s), T stage, molecular subtypes (c values: 2.55 (2.20-2.82) vs. 2.69 (2.20-2.96) vs. 2.94 (2.52-3.33) m/s), and enhancement patterns (c values: 2.34 ± 0.41 vs. 2.73 ± 0.58 m/s). Multivariable analysis identified c [OR: 5.37 (2.13-15.54)] and age group [OR: 0.27 (0.09-0.82)] as significant independent predictors of triple-negative breast cancer, and a predictive model incorporating these variables demonstrated an AUC of 0.739 in the training set and 0.700 in the validation set. DATA
Conclusion
The 40-70 Hz multifrequency protocol may offer favorable image quality for breast MRE. MRE parameters, particularly stiffness, show potential for characterizing tissue biomechanical properties and could assist in distinguishing molecular subtypes. EVIDENCE LEVEL 1. STAGE OF TECHNICAL EFFICACY Stage 2.