Evaluation of extracapsular extension in prostate cancer using qualitative and quantitative multiparametric MRI

Abstract

Purpose: To investigate the value of multiparametric magnetic resonance imaging (mpMRI) for extracapsular extension (ECE) in prostate cancer (PCa). Materials and Methods: In all, 292 patients who received radical prostatectomy and underwent preoperative mpMRI at 3T were enrolled retrospectively. For determining the associations with ECE, the likelihood of ECE was assessed qualitatively on T2-weighted imaging (T2WI) and combined T2WI and diffusion-weighted imaging (DWI) or dynamic contrast-enhanced imaging (DCEI). Quantitative MRI parameters were measured in PCa based on histopathological findings. Two models for detecting ECE including imaging and clinical parameters were developed using multivariate analysis: Model 1 excluding combined T2WI and DWI and DCEI and Model 2 excluding combined T2WI and DWI, and combined T2WI and DCEI. Diagnostic performance of imaging parameters and models was evaluated using the area under the receiver operating characteristics curve (Az). Results: For detecting ECE, the specificity, accuracy, and Az of combined T2WI and DWI or DCEI were statistically better than those of T2WI (P < 0.05), and all quantitative MRI parameters showed a statistical difference between the patients with and without ECE (P < 0.05). On multivariate analysis, significant independent markers in Model 1 were combined T2WI and DWI, combined T2WI and DCEI, and Ktrans (P < 0.05). In Model 2, significant markers were combined T2WI and DWI and DCEI, Ktrans, Kep, and Ve (P < 0.05). The Az values of models 1 and 2 were 0.944 and 0.957, respectively. Conclusion: mpMRI may be useful to improve diagnostic accuracy of the models for determining the associations with ECE in PCa. Level of Evidence: 4. Technical Efficacy: Stage 2. J. MAGN. RESON. IMAGING 2017;45:1760–1770.