Insignificant prostate cancer in radical prostatectomy specimen: time trends and preoperative prediction.

Abstract

We analysed systematically a consecutive series of radical prostatectomy specimens performed between January 1992 and June 2002 with emphasis to time trends, tumour characteristics and preoperative prediction of insignificant prostate cancers (cancer volume < or =0.5 cm(3) and Gleason pattern < or =6). In a total of 1254 patients, prostate cancers (PC) were divided by a cancer volume of 0.5 cm(3). The two groups were compared in their clinical and pathological tumour characteristics. Correlation was determined between yearly incidence rates of T1c and insignificant PC. Furthermore, a logistic regression analysis was performed to calculate the ability to predict insignificant PC and a statistical model was established. Overall, 73 (5.8%) of 1254 men presented with insignificant PC. The incidence of insignificant PC showed no significant linear correlation with that of T1c PC (p<0.61). PSA density and percentage of cancer per biopsy set were assessed as independent prognosticators predicting insignificant PC. Using a threshold of 1% of cancer per biopsy set and a PSA density < or =0.10, positive and negative predictive values were 45.0% and 93.3%, respectively. In our series, only few men undergoing radical prostatectomy were affected by insignificant PC. Their incidence showed no statistically significant correlation with that of T1c tumours. Furthermore, insignificant PC was predictable by PSA density and percentage of cancer per biopsy set. Mainly elderly patients facing different treatment options for localized PC may benefit from this information.