DCE‐MRI of the prostate using shutter‐speed vs. Tofts model for tumor characterization and assessment of aggressiveness

摘要

Purpose To quantify Tofts model (TM) and shutter‐speed model (SSM) perfusion parameters in prostate cancer (PCa) and noncancerous peripheral zone (PZ) and to compare the diagnostic performance of dynamic contrast‐enhanced magnetic resonance imaging (DCE‐MRI) to Prostate Imaging and Reporting and Data System (PI‐RADS) classification for the assessment of PCa aggressiveness. Materials and Methods Fifty PCa patients (mean age 60 years old) who underwent MRI at 3.0T followed by prostatectomy were included in this Institutional Review Board‐approved retrospective study. DCE‐MRI parameters (K trans , v e , k ep [TM&SSM] and intracellular water molecule lifetime τ i [SSM]) were determined in PCa and PZ. Differences in DCE‐MRI parameters between PCa and PZ, and between models were assessed using Wilcoxon signed‐rank tests. Receiver operating characteristic (ROC) analysis for differentiation between PCa and PZ was performed for individual and combined DCE‐MRI parameters. Diagnostic performance of DCE‐MRI parameters for identification of aggressive PCa (Gleason ≥8, grade group [GG] ≥3 or pathology stage pT3) was assessed using ROC analysis and compared with PI‐RADSv2 scores. Results DCE‐MRI parameters were significantly different between TM and SSM and between PZ and PCa ( P 0.037). Diagnostic performances of TM and SSM for differentiation of PCa from PZ were similar (highest AUC TM: K trans +k ep 0.76, SSM: τ i +k ep 0.80). PI‐RADS outperformed TM and SSM DCE‐MRI for identification of Gleason ≥8 lesions (AUC PI‐RADS: 0.91, highest AUC DCE‐MRI: K trans +τ i SSM 0.61, P ?=?0.002). The diagnostic performance of PI‐RADS and DCE‐MRI for identification of GG ≥3 and pT3 PCa was not significantly different ( P 0.213). Conclusion SSM DCE‐MRI did not increase the diagnostic performance of DCE‐MRI for PCa characterization. PI‐RADS outperformed both TM and SSM DCE‐MRI for identification of aggressive cancer. Level of Evidence: 3 Technical Efficacy: Stage 2 J. MAGN. RESON. IMAGING 2017;46:837–849