Pretreatment prostate-specific antigen and Gleason score predict the risk of extracapsular extension and the risk of failure following radiotherapy in patients with clinically localized prostate cancer

The purpose of this study is to evaluate the use of a relatively simple equation for predicting the risk of extracapsular extension (ECE) based on the pretreatment prostate-specific antigen (PSA) and Gleason score (GS) in patients with clinically localized prostate cancer. Three hundred and seventy-four patients who underwent radical prostatectomy between 1988 and 1994 and 521 men undergoing definitive radiotherapy during a similar time period were eligible for this analysis. Surgically treated patients were considered eligible if the pathological stage, preoperative PSA, and GS were available. Among these patients, the median preoperative PSA was 8.1 ng/mL (range, 0 to 195 ng/mL), and the median preoperative GS was 6 (range, 2 to 10). The empirically derived equation tested was (1.5 x PSA + [GS - 3] x 10). For this equation, the range of calculated risk was limited to 0% to 100%. Using the empirically derived equation, patients with a low calculated risk (CR) of < or = 33% had an average calculated risk (ACR) of 21.9% and an observed incidence (OI) of ECE was 17.8%. Patients with a moderate CR of 34% to 66% had an ACR of 46.3%, and an OI of ECE was 46.7%. Patients with a CR of 67% to 100% had an ACR of 83.7% and an OI of ECE of 66.7%. Of the 21 patients who had a PSA < or = 4 and a GS < or = 4, only 1 patient (4.8%) was found to have ECE. Men with an estimated risk of ECE of <33%, 33% to 67%, and >67% had a 4-year risk of biochemical failure following radiotherapy of 29%, 56%, and 78% (P < .00001). This empirically derived data appears to be reasonably accurate at estimating the incidence of ECE in patients with at low or intermediate risk before surgery. The risk of biochemical failure following radiotherapy also correlated the risk of ECE. Future staging systems for prostate cancer should use similar approach for defining risk groups.