Preoperative prediction of small volume cancer (less than 0.5 ml) in radical prostatectomy specimens

Purpose: The detection of low volume and early stage prostate cancer has increased with the widespread use of prostate specific antigen screening for prostatic adenocarcinoma. This increased detection has led to efforts to stratify patient risk and the potential benefits of various treatments based on preoperative clinical and biopsy data. We examined various clinical parameters and prostate biopsy features to determine which variables are most predictive of small volume (less than 0.5 ml) cancer at prostatectomy. Materials and Methods: We studied 336 patients who underwent prostatectomy for prostate cancer. Radical prostatectomy specimens were completely embedded and whole mounted. Final tumor volume in the radical prostatectomy specimens was determined by the grid method. Clinical data were gathered by a review of patient charts. Various preoperative clinical and biopsy findings were analyzed to determine factors predictive of small volume cancer at prostatectomy. Results: A total of 55 patients (16%) were found to have small volume cancer (less than 0.5 ml). On univariate logistic regression certain variables were significant predictors of small volume cancer, namely the highest Gleason score from all positive biopsy sites (p = 0.001), the Gleason score from the biopsy site with the highest percent of adenocarcinoma (p = 0.006), the highest percent of adenocarcinoma at any biopsy site (p <0.0001), the percent of adenocarcinoma at the biopsy site with the highest Gleason score (p <0.0001), the highest percent of cores positive for adenocarcinoma at any biopsy site (p = 0.001), the percent of cores with carcinoma at the site with the highest Gleason score (p = 0.002), the number of positive sites (p <0.0001) and tumor bilaterality (p <0.0001). None of the clinical parameters that we studied, including preoperative prostate specific antigen (p = 0.52), clinical stage (p = 0.62) or patient age (p = 0.94), was predictive of small volume cancer. On multivariate analysis the highest percent of adenocarcinoma at any site (adjusted OR 0.95, 95% CI 0.92 to 0.97, p <0.0001) and the number of positive biopsy sites (adjusted OR 0.97, 95% CI 0.96 to 0.99, p <0.0001) were significant predictors of small volume cancer. Conclusions: The number of positive biopsy sites and the highest percent of adenocarcinoma at any biopsy site are significant predictors of small volume cancer in radical prostatectomy specimens.