Synthetic MRI Parameters and Breast Cancer Heterogeneity: A Quantitative Study

To explore the associations between synthetic MRI (SyMRI)-derived quantitative parameters and histopathological and imaging features in breast cancer, and to identify potential confounding factors affecting parameter interpretation.

In this retrospective study, 582 patients with pathologically confirmed breast cancer who underwent preoperative SyMRI (MAGiC sequence) on a 3.0 T scanner were included. Quantitative T1, T2, PD, and post-contrast T1 (T1C) values were extracted from tumor regions. Clinicopathological data (ER, PR, HER2, Ki-67, histological grade) and imaging features (tumor size, background parenchymal enhancement [BPE], fibroglandular tissue [FGT]) were collected and analyzed. Two-sample t-tests or Mann-Whitney U tests were used for intergroup comparisons. Effect sizes (η²) were calculated to quantitatively evaluate the clinical relevance and strength of associations.

Tumors of high histological grade and those in the ER-negative, PR-negative, and HER2-positive subgroups demonstrated significantly higher T1, T2, PD, and T1C values (P < 0.05). High Ki-67 expression was associated with increased T1, T2, and T1C values, but not with PD. Larger tumor size (>2 cm) was associated with higher T1, T2, and T1C values. PD values were elevated in patients with dense FGT and high BPE, whereas T2 values were slightly higher in those with low FGT. Effect size analysis revealed that histological grade had a moderate effect on T1C (η² = 0.132) and T2 (η² = 0.075), while the other features exhibited only small or negligible effects (η² < 0.06).

SyMRI-derived quantitative metrics are associated with both pathological and imaging features. Effect size analysis highlights histological grade as the most influential confounding factor in the interpretation of quantitative SyMRI.