Preoperative PSA level significantly associated with interval to biochemical progression after radical retropubic prostatectomy

Objectives. To evaluate the association between the preoperative prostate-specific antigen (PSA) level and the interval to biochemical progression after radical retropubic prostatectomy (RRP) for prostate cancer. Methods. We studied 5534 men who underwent RRP for prostate cancer during a 15-year period and were enrolled in prospective follow-up studies. Of these men, 857 (15%) subsequently had biochemical evidence of cancer progression (PSA greater than 0.2 ng/mL). All men with biochemical progression after RRP were arbitrarily stratified into five groups according to their preoperative PSA level: less than 2.6 ng/mL, 2.6 to 4 ng/mL, 4.1 to 10 ng/mL, 10. 1 to 20 ng/mL, and greater than 20 ng/mL. The time to biochemical progression after RRP was the primary endpoint of the analysis. Other prognostic factors were also evaluated in multivariable analysis. Results. The median time to progression for men with a PSA level greater than 20 ng/mL was significantly shorter (16 months) than for men with a PSA level of 10. 1 to 20 ng/mL (27 months) or those with lower PSA levels (P = 0.0005). Men with a PSA level of 4.1 to 10 ng/mL did not have a statistically significantly different median time to progression (31 months) from men with a PSA level of 10. 1 to 20 ng/mL (27 months; P = 0.3), 2.6 to 4 ng/mL (32 months; P = 0.8), or less than 2.6 ng/mL (38 months; P = 0.3). Also, no statistically significant difference was found in the median time to progression between men with a PSA level of 2.6 to 4 ng/mL and less than 2.6 ng/mL (P = 0.4). In an analysis of covariance multivariate analysis, the preoperative PSA level was not an independent predictor of the time to biochemical progression (P = 0. 12), after accounting for tumor stage and Gleason grade. Conclusions. The preoperative PSA level was significantly associated with the interval to biochemical cancer progression after RRP; however, this association appeared to be because the preoperative PSA level serves as a surrogate marker for other prognostic factors, such as the tumor volume, tumor stage, and Gleason grade. (C) 2004 Elsevier Inc.