Subtype differentiation of renal cell carcinoma using diffusion-weighted and blood oxygenation level-dependent MRI.

OBJECTIVE: The purpose of this article is to evaluate the utility of diffusion-weighted imaging (DWI) and blood oxygenation level-dependent (BOLD) MRI for characterizing renal cell carcinoma (RCC) subtypes at 3 T.

MATERIALS AND METHODS: Seventy-seven patients underwent 3-T DWI and BOLD MRI. Apparent diffusion coefficient (ADC; × 10(-3) mm(2)/s) and rate of spin dephasing (R2*, which equals 1 / T2* relaxation time, or 1/s) values were measured in the three RCC subtypes and normal renal parenchyma, and the results were compared. Statistical analyses were performed using analysis of variance, Student t test, and ROC curve analysis.

RESULTS: Clear cell RCCs showed statistically significantly greater ADC values (1.81 × 10(-3) mm(2)/s) than did papillary (1.29 × 10(-3) mm(2)/s) and chromophobe (1.55 × 10(-3) mm(2)/s) RCCs (p < 0.01); however, no statistically significant differences between papillary and chromophobe RCCs were observed (p = 0.26). Chromophobe RCCs showed the greatest mean R2* (33.6 1/s) of the three subtypes (p < 0.01); however, no statistically significant differences between clear cell RCCs and papillary RCCs were seen (p = 0.48). Low-grade clear cell RCCs showed statistically significantly higher ADC value (1.97 × 10(-3) mm(2)/s) than did high-grade clear cell RCCs (1.66 × 10(-3) mm(2)/s; p = 0.021). For differentiating clear cell RCCs from non-clear cell RCCs, the AUCs of ADC and R2* values were 0.756 × 10(-3) mm(2)/s and 0.607 (1/s), respectively (p = 0.047): cutoff values of ADC (1.4 × 10(-3) mm(2)/s) and R2* (26.3 1/s) resulted in sensitivities and specificities of 85% and 73%, and 86% and 47%, respectively.

CONCLUSION: For characterizing RCC subtypes, DWI and BOLD MRI at 3 T may be useful, but the current technique of BOLD MRI seems to have a limited diagnostic accuracy.