Analysis of kinetic curve and model-based perfusion parameters on dynamic contrast enhanced MRI in breast cancer patients: Correlations with dominant stroma type

Abstract

OBJECTIVE: Our purpose was to evaluate imaging findings of breast cancers according to the dominant stroma type by using kinetic curve analysis and model-based perfusion parameters from dynamic contrast-enhanced magnetic resonance imaging. METHODS: From March 2011 to September 2011, 64 cancers in 64 patients were included for data analysis. Kinetic curve analysis and model based perfusion parameters (Ktrans, Kep and Ve) were obtained using dynamic contrast-enhanced magnetic resonance imaging and post-processing software. Imaging characteristics were analyzed according to the tumor-stroma ratio and dominant stroma type. RESULTS: Ve values were significantly lower in tumors with more than 50% cellularity (0.44 vs. 0.29, p=0.008). Histologic grade, estrogen receptor status and subtype of cancer (triple negative versus non-triple negative) were significantly different (p=0.009, p=0.019 and p=0.03, respectively). Median Kep values were different between collagen dominant, fibroblast dominant and lymphocyte dominant groups. By post hoc comparisons, mean Kep values were significantly higher in lymphocyte dominant group than collagen dominant group (p=0.003). Ktrans and Ve values were not significantly different according to dominant stroma type (p=0.351 and p=0.257, respectively). In multivariate regression analysis, nuclear grade (p=0.021) and dominant stroma type (collagen dominant, p=0.017) were independently correlated with Kep values. In terms of the dominant stroma type, the collagen dominant type showed a decrease of 0.247 in Kep values, compared with the fibroblast-dominant type (p=0.017). CONCLUSIONS: Ve values were significantly lower in tumors with high tumor-stroma ratio. Kep values were significantly lower in breast cancers with dominant collagen type and higher in cancers with high nuclear grade.