Race as an outcome predictor after radical prostatectomy: results from the Shared Equal Access Regional Cancer Hospital (SEARCH) database

Abstract

Objectives Whether race is an independent predictor of prostate-specific antigen (PSA) recurrence after RP is controversial. To compare racial differences in clinical and pathologic features and biochemical recurrence in men undergoing radical prostatectomy (RP), we used a newly established multicenter database of patients from four equal-access healthcare centers in California, the Shared Equal Access Regional Cancer Hospital (SEARCH) database. Methods A retrospective survey of 1547 patients treated with RP at four different equal-access medical centers in California between 1988 and 2001 was undertaken. Race was categorized as white (n = 1014), black (n = 338), or nonwhite-nonblack (n = 195). Patients were analyzed for racial differences in preoperative variables (age at surgery, clinical stage, PSA, and biopsy Gleason score) and surgical variables (pathologic stage, surgical Gleason score, incidence of seminal vesicle invasion, positive surgical margins, capsular penetration, and pelvic lymph node involvement). Patients were followed up for PSA recurrence. Multivariate analysis was used to determine whether race was an independent predictor of biochemical failure. Results Significant differences were found among the races in the preoperative factors of clinical stage, age, serum PSA, and biopsy Gleason score, although the absolute differences were small. No differences were found among the races in the pathologic features of the RP specimens, including Gleason score, pathologic stage, and incidence of positive surgical margins, capsular penetration, seminal vesicle invasion, or lymph node involvement. In both univariate and multivariate analyses, only serum PSA ( P <0.001) and biopsy Gleason score ( P <0.001) were significant independent predictors of time to biochemical recurrence. Conclusions In a large multicenter cohort of patients from four equal-access medical care facilities in California, although racial differences were found in clinical stage, age, biopsy Gleason score, and serum PSA level at diagnosis, we found race was not an independent predictor of biochemical recurrence after RP. Race should not be used in models or nomograms predicting PSA failure after RP. The current study represents the largest series of black patients and the first large series of nonwhite-nonblack patients treated with RP reported to date. The Shared Equal Access Regional Cancer Hospital database is a valuable resource for studying patients treated with RP.