Prostate-specific antigen velocity is not better than total prostate-specific antigen in predicting prostate biopsy diagnosis

Abstract

Objectives1.) Identify whether prostate-specific antigen velocity improves the ability to predict prostate biopsy diagnosis. 2.) Test whether there is an increase in the predictive capability of models when Gleason 7 prostate cancers are separated into a 3+4 and a 4+3 group. Design and methodsCalgary Laboratory Services' Clinical Laboratory Information System was searched for prostate biopsies reported between January 1, 2009 and December 31, 2013. Total prostate-specific antigen tests were recorded for each patient from January 1, 2007 to the most recent test before their recorded prostate biopsy. The data set was divided into the following three groups for comparison; benign, all prostate cancer and Gleason 7–10. The Gleason grade 7–10 group was further divided into 4+3 and 3+4 Gleason 7 prostate cancers. Prostate-specific antigen velocity was calculated using four different methods found in the literature. Receiver operator curves were used to assess operational characteristics of the tests. Results4622 men between the ages of 40–89 with a prostate biopsy were included for analysis. Combining prostate-specific antigen velocity with total prostate-specific antigen (AUC=0.570–0.712) resulted in small non-statistically significant changes to the area under the curve compared to the area under the curve of total prostate-specific antigen alone (AUC=0.572–0.699). There were marked increases in the area under curves when 3+4 and 4+3 Gleason 7 cancers were separated. ConclusionsProstate-specific antigen velocity does not add predictive value for prostate biopsy diagnosis. The clinical significance of the prostate specific antigen test can be improved by separating Gleason 7 prostate cancers into a 3+4 and 4+3 group.