Prostate specific antigen adjusted for transition zone epithelial volume: the powerful predictor for the detection of prostate cancer on repeat biopsy

J Urol. 2005 May;173(5):1541-5. doi: 10.1097/01.ju.0000154636.24375.76.

Abstract

Purpose: The indications for repeat prostate biopsy for persistently increased prostate specific antigen (PSA) in men with prostate cancer never detected on previous biopsy are not clear. In this study we determined that PSA adjusted for transition zone (TZ) epithelial volume is the most powerful predictor for detecting prostate cancer on repeat biopsy.

Materials and methods: Repeat prostate biopsies including additional TZ cores were performed in 75 men with PSA between 4.0 and 10.0 ng/ml. TZ epithelial volume was calculated by multiplying TZ volume by the percent of epithelium, which was measured by morphometric analysis using image analysis computer software.

Results: Prostate cancer was detected on repeat biopsy in 19 of the 75 patients. Patients with prostate cancer had a significant smaller percent area of epithelium or glandular lumen than those without cancer. In patients without prostate cancer TZ epithelial volume significantly correlated with total PSA. According to ROC analysis PSA adjusted for TZ epithelial volume had the greatest AUC for cancer detection (0.879). This parameter was able to avoid more than 90% of unnecessary repeat biopsies with 90% sensitivity. Multiple logistic regression analysis showed that PSA complex adjusted for TZ epithelial volume was the significant independent predictor of cancer.

Conclusions: PSA adjusted for TZ epithelial volume is the most powerful predictor of cancer in men who have undergone previous negative prostate biopsies and in whom PSA remains between 4.0 and 10.0 ng/ml.

MeSH terms

  • Aged
  • Biopsy / statistics & numerical data
  • Humans
  • Male
  • Predictive Value of Tests
  • Prostate-Specific Antigen / blood*
  • Prostatic Neoplasms / blood*
  • Prostatic Neoplasms / pathology*

Substances

  • Prostate-Specific Antigen