Differences in morphological features and minimum apparent diffusion coefficient values among breast cancer subtypes using 3-tesla MRI

Purpose: To compare the morphology and minimum apparent diffusion coefficient (ADC) values among breast cancer subtypes. Methods: Ninety-three patients, who underwent breast MRI and collectively had 98 pathologically proven invasive carcinomas, were enrolled. Morphology was evaluated according to BIRADS-MRI. Minimum ADC was measured. Morphology and minimum ADC were compared among subtypes. Multivariate logistic regression analyses were used to identify the characteristics associated with different subtypes. Results: Oval/round shape was significantly associated with triple-negative (TN) cancer (TN vs. non-TN: 90.9% vs. 45.2%; p = 0.0123). Rim enhancement was significantly less frequent in Luminal A (Luminal A vs. non-Luminal A: 34.2% vs. 76.1%; p = 0.0003). The minimum ADC of Luminal A was significantly higher than that of Luminal B (HER2-negative) (834 vs. 748 x 10^-6 mm2/s; p <0.025). The minimum ADC of the TN-special type was significantly higher than that of TN-ductal (997 vs. 702 x 10^-6 mm2/s; p <0.025). On the multivariate analysis comparing the characteristics associated with Luminal A vs. Luminal B (HER2-negative), the internal enhancement characteristics of the mass and minimum ADC were significant factors. Conclusion: Morphology and minimum ADC would be useful in distinguishing breast cancer subtypes.