Preoperative assessment of prostate cancer using prebiopsy MRI.

Abstract

The purpose of this article is to retrospectively evaluate the impact of prebiopsy MRI on preoperative assessment of prostate cancer. Between January 2005 and June 2012, 89 patients (mean age, 64 years; range, 50-77 years) underwent prebiopsy MRI, transrectal ultrasound-guided biopsy, and prostatectomy for adenocarcinoma. On MR images, prostate cancer was diagnosed if T2-weighted images, diffusion-weighted images, and dynamic contrast-enhanced images were all positive. Preoperative cancers were classified as low, intermediate, high, and very high risk using prostate-specific antigen (PSA) level, biopsy Gleason score, or MRI tumor stage. Postoperative cancers were classified as low, intermediate, high, and very high risk using PSA level, prostatectomy Gleason score, and prostatectomy tumor stage. Interrater agreement using weighted kappa values was used to compare the paired data. Preoperative cancers assessed by PSA level alone and postoperative cancers were concordant in 24.7% (22/89) and discordant in 75.3% (67/89) of cases, with poor agreement (weighted κ, 0.13; 95% CI, 0.05-0.21). Preoperative cancers assessed by biopsy Gleason score alone and postoperative cancers were concordant in 40.4% (36/89) and discordant in 59.6% (53/89) of cases, with fair agreement (weighted κ, 0.22; 95% CI, 0.09-0.36). Preoperative cancers assessed by MRI tumor stage alone and postoperative cancers were concordant in 74.2% (66/89) and discordant in 25.8% (23/89) of cases, with moderate agreement (weighted κ, 0.54; 95% CI, 0.38-0.71). Prebiopsy MRI may provide better information on preoperative assessment of prostate cancer than PSA level and biopsy Gleason score.