Advertisement

Prostate cancer loci with negative multiparametric MRI for prostate cancer: correlation with prostatectomy specimens

Login to Access Video or Poster Abstract: MP20-08
Sources of Funding: None

Introduction

Precise localization of tumor locus is critically important for successful focal therapy in prostate cancer. The accuracy of multiparametric MRI (mpMRI) for prostate cancer localization is still unknown.

Methods

We retrospectively analyzed 176 prostate cancer patients who underwent preoperative 3.0T multiparametric MRI (mpMRI) followed by radical prostatectomy (RP) without neoadjuvant androgen deprivation. Images of mpMRI was evaluated by a single radiologist based on PI-RADS version 2. PI-RADS score 4 or greater was considered positive. Tumor distribution was evaluated on radical prostatectomy specimen sliced at 5-mm thick and tumor volume was estimated based on planimetry.

Results

Of the 176 study patients, 79 (45%) had negative mpMRI. Patients with negative mpMRI had smaller index tumor compared with those with positive mpMRI with a marginal significance (1.79 ± 0.25 vs 20.5 ± 0.22 cc., p = 0.05, Figure and Table). They were also more likely to have clinically insignificant cancer (Gleason score ≤6 and tumor volume <0.5 cc.) compared with those with positive mpMRI (Tables). However, the sensitivity and specificity for clinically significant prostate cancer was only 60% and 57%, respectively, indicating that mpMRI missed 40% of clinically significant cancer.

Conclusions

Although negative mpMRI is associated with favorable pathological findings in prostate cancer patients who underwent radical prostatectomy, mpMRI could not safely exclude clinically significant index tumor. Further improvement in the accuracy for tumor localization is warranted for the possible application of mpMRI to plan focal therapy.

Funding

None

Authors
Katsuhiro Ito
Takashi Kobayashi
Akihiro Furuta
Yuki Teramoto
Naoki Terada
Shusuke Akamatsu
Toshinari Yamasaki
Takahiro Inoue
Osamu Ogawa
back to top