Login to Access Video or Poster Abstract: MP03-12
Sources of Funding: none

Introduction

To evaluate mpMRI-based assessments of neurovascular bundle (NVB) infiltration and to determine the value of PI-RADS V2 scores for the prediction of NVB-infiltration before prostatectomy.

Methods

Our institutional review board approved the study. 198 patients underwent standardized mpMRI at 3T prior to surgery, including high resolution T2w-TSE-imaging in 3 planes, T1-w-TSE, DWI with ADC map, PD-TSE and Gd-DCE with post-processing of images. Assessment for NVB-infiltration was made for each side of each prostate (n=396). Maximum PI-RADS-V2 scores were determined for the posterolateral areas adjacent to the NVB (n=396). MRI-findings were correlated to pathologic analysis as reference standard, where NVB-infiltration was defined as tumor invasion into the NVB or extraprostatic expansion (EPE) in the posterolateral area adjacent to the NVB.

Results

Overall T-staging accuracy, sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of mpMRI were 78.3%, 644 %, 892 %, 82.4% and 76.2%, respectively. In 396 cases infiltration of the NVB was predicted with 89.4%, 75.2%, 94.0%, 80.2% and 92.1% overall accuracy, sensitivity, specificity, PPV and NPV, respectively. By correlating 365 maximum PIRADS-V2 scores to the pathology of adjacent NVBs, infiltration was demonstrated in 13 NVBs despite low likelihood of cancer presence (PI-RADS 1 or 2 scores), amounting to 14% false negative predictions.

Conclusions

mpMRI-based assessment of NVB-infiltration should be acknowledged when nerve sparing surgery is considered. However, areas without tumor suspicion (PI-RADS 1 or 2) might demonstrate NVB-infiltration in pathology causing false negative predictions.

Funding

none