Evaluation of the accuracy of multiparametric MRI for predicting prostate cancer pathology and tumour staging in the real world: an multicentre study.

To assess the accuracy of multiparametric magnetic resonance imaging (mpMRI) for the detection of significant prostate cancer in men undergoing radical prostatectomy (RP) in an Australian multicentre setting, and to assess concordance between mpMRI and RP for local tumour staging and index lesion locations. Men who underwent mpMRI within 12 months of RP between January 2013 and August 2016 at three Australian sites were included (Central Coast, NSW, St Vincents Hospital, Melbourne, Vic., and Bendigo Hospital, Vic.). The results of mpMRI were compared with the final RP specimen to analyse the performance of mpMRI for significant prostate cancer detection, index lesion localization, prediction of T3 disease and lymph node metastasis. A comparison between mpMRI cases performed using the technical and reporting specifications of Prostate Imaging Reporting and Data System (PI-RADS) version 1 and version 2 was also performed. Data analysis was performed using spss 24.0. A total of 235 cases were included for analysis. mpMRI PI-RADS score ≥3 had a 91% sensitivity and 95% positive predictive value (PPV) for significant prostate cancer at RP. The overall concordance between index lesion location on mpMRI and RP specimen was 75%. The sensitivity for predication of significant prostate cancer was higher in the PI-RADS version 2 cases compared with PI-RADS version 1 (87-99%; P = 0.005). Index lesion concordance was higher in the PI-RADS version 2 group (68% vs 91%; P = 0.002). mpMRI had a 38% sensitivity, 95% specificity, 90% PPV and 57% negative predictive value for extraprostatic disease. Sensitivity for prediction of T3 disease improved from 30% to 62% (P = 0.008) with PI-RADS version 2. In patients undergoing RP, an abnormal mpMRI is highly predictive (95% PPV) of significant prostate cancer, with an index lesion concordance of 75%. There has been a significant improvement in accuracy after the adoption of PI-RADS version 2 technical specifications and reporting criteria; however; further study is required to determine if this is attributable to improved experience with mpMRI or changes in the PI-RADS system.

Informations de copyright

© 2019 The Authors BJU International © 2019 BJU International Published by John Wiley & Sons Ltd.