Multiparametric magnetic resonance imaging for prostate cancer - a comparative study including radical prostatectomy specimens.

Abstract

PURPOSE: To evaluate the diagnostic and staging ability of multiparametric MRI (mpMRI) compared to radical prostatectomy (RP) specimens after dissemination of this technology to several centres. mpMRI is an evolving technique aiming to improve upon the diagnostic sensitivity of prostate biopsy for the diagnosis of prostate cancer. Differences in interpretation, expertise and application of mpMRI are responsible for the range of reported results. METHODS: This retrospective clinical study was conducted with consecutive patients through an electronic database of tertiary hospitals and adjacent private urology practices in Australia. Patients having undergone RP were assessed for the presence of a pre-operative mpMRI performed between 2013 and 2015 which was evaluated against the reference standard of the RP whole-mount specimen. MRI reports were evaluated using the Prostate Imaging Reporting and Data System (PI-RADS). RESULTS: In our cohort of 152 patients, the sensitivity and specificity of mpMRI (PI-RADS ≥ 4) for prostate cancer (Gleason ≥ 4 + 3) detection were 83 and 47%, respectively. For the identification of extraprostatic disease, the sensitivity and specificity were 29 and 94%, respectively. CONCLUSION: These results represent a 'real-world' approach to mpMRI and appear comparable to other single-centre studies. MRI staging information should be interpreted in context with other risk factors for extraprostatic disease. mpMRI has a useful role as an adjunct for prostate cancer diagnosis and directing management towards improving patient outcomes.