Time trends in clinical risk stratification for prostate cancer: implications for outcomes (data from CaPSURE)

Abstract

Purpose: Many instruments designed to predict prostate cancer risk use a combination of clinical T stage, biopsy Gleason score and serum prostate specific antigen (PSA). We designed a study to characterize time trends in these parameters and their impact on patient risk stratification.

Materials and methods: Data were abstracted from CaPSURE (Cancer of the Prostate Strategic Urological Research Endeavor), a disease registry of 8,685 men with prostate cancer. The 6,260 men diagnosed since 1989 who had complete clinical information reported were categorized into low, intermediate or high risk groups based on established parameters for T stage, Gleason score and PSA.

Results: Between 1989 to 1990 and 2001 to 2002 the proportion of patients presenting with high, intermediate and low risk disease changed from 40.9%, 28.0% and 31.2% to 14.8%, 37.5% and 47.7%, respectively (p <0.0001). The incidence of T1 tumors increased from 16.7% to 48.5% and that of T3-4 tumors decreased from 11.8% to 3.5%, respectively (p <0.0001). The incidence of Gleason 2 to 6 tumors decreased from 77.1% to 66.4%, while that of Gleason 7 tumors increased from 12.9% to 24.8%, respectively (p = 0.0030). PSA levels 10 ng/ml or less increased from 43.6% to 77.7%, respectively, while PSA 10 to 20 and greater than 20 ng/ml decreased accordingly (p <0.0001). These trends were mirrored in subset analysis of black patients.

Conclusions: A significant downward risk migration has occurred over time. Gleason score is now more likely and PSA less likely than previously to drive risk assignment. This shift is most likely attributable to changes in practice patterns with respect to screening and pathological grading. These changes should be considered when applying nomograms derived from earlier datasets to contemporary cases.