Clinical utility of endorectal MRI in determining PSA outcome for patients with biopsy Gleason score 7, PSA <or=10, and clinically localized prostate cancer

PURPOSE: Although the optimal management for patients with high-grade clinically localized prostate cancer is undefined, radical prostatectomy (RP) or external beam radiotherapy (EBRT) is performed. The clinical utility of the pretreatment prostrate-specific antigen (PSA) level (10 ng/mL) and endorectal MRI (erMRI) stage (T3 vs. T2) to stratify PSA outcome after RP in these patients was evaluated. METHODS AND MATERIALS: erMRI was performed in 147 men with biopsy Gleason score >or=7 and 1992 AJCC clinical Stage T1c or T2a disease before RP. Enumerations of the biopsy and prostatectomy Gleason scores, pathologic stage, and margin status were performed for each pretreatment group on the basis of erMRI findings and PSA level. Comparisons were made using a chi-square metric. The median follow-up was 4.5 years (range 1-10 years). Comparisons of the actuarial freedom from PSA failure (bNED) were made using the log-rank test. RESULTS: erMRI Stage T2 and T3 disease was found in 132 and 15 patients, respectively. On stratification by PSA level, patients with erMRI T3 disease had similar bNED outcomes (p = 0.46), regardless of the PSA level. The 3-year bNED rate was 82%, 64%, and 25% (p <0.0001) for Group 1 (erMRI T2 and PSA 10 ng/mL), and Group 3 (erMRI T3 with any PSA level), respectively. The rates of prostatectomy T3 disease, biopsy and prostatectomy Gleason score 8-10, and positive surgical margins were significantly higher (p or=7, PSA