Combined Diffusion-Weighted, Blood Oxygen Level-Dependent, and Dynamic Contrast-Enhanced MRI for Characterization and Differentiation of Renal Cell Carcinoma

Purpose: To investigate a multiparametric magnetic resonance imaging (MRI) approach comprising diffusion-weighted imaging (DWI), blood oxygen-dependent (BOLD), and dynamic contrast-enhanced (DOE) MRI for characterization and differentiation of primary renal cell carcinoma (ROC). Material and Methods: Fourteen patients with clear-cell carcinoma and four patients with papillary RCC were examined with DWI, BOLD MRI, and DOE MRI at 1.5T. The apparent diffusion coefficient (ADC) was calculated with a monoexponential decay. The spin-dephasing rate R2(star) was derived from parametric R2(star) maps. DOE-MRI was analyzed using a two-compartment exchange model allowing separation of perfusion (plasma flow [F-P] and plasma volume [V-P]), permeability (permeability surface area product [PS]), and extravascular extracellular volume (V-E). Statistical analysis was performed with Wilcoxon signed-rank test, Pearson's correlation coefficient, and receiver operating characteristic curve analysis. Results: Clear-cell RCC showed higher ADC and lower R2(star) compared to papillary subtypes, but differences were not significant. F-P of clear-cell subtypes was significantly higher than in papillary ROC. Perfusion parameters showed moderate but significant inverse correlation with R2(star). V-E showed moderate inverse correlation with ADC. F-p and V-p showed best sensitivity for histological differentiation. Conclusion: Multiparametric MRI comprising DWI, BOLD, and DOE MRI is feasible for assessment of primary ROC. BOLD moderately correlates to DOE MRI-derived perfusion. ADC shows moderate correlation to the extracellular volume, but does not correlate to tumor oxygenation or perfusion. In this preliminary study DOE-MR1 appeared superior to BOLO and DWI for histological differentiation.