Race as an independent predictor of outcome after treatment for localized prostate cancer

Purpose: We analyze the outcome after prostatectomy or radiotherapy for loc alized prostate cancer with respect to race. Materials and Methods: A total of 2,219 consecutive patients with prostate cancer were treated with radiotherapy (1,183) or radical prostatectomy (1,0 36) between June 1986 and June 1998. Initial prostate specific antigen (PSA ) and biopsy Gleason scores were available in all cases. Androgen deprivati on was used in 22% of men (492). Of the patients 86% (1,901) were white, in cluding Hispanic and Asian, and 14% (318) were black. The outcomes of inter est were biochemical relapse-free survival, clinical relapse-free survival and overall survival. Median followup was 24 months (range 2 to 140). Results: There was no difference in the incidence of familial prostate canc er, patient age at presentation, clinical stage or biopsy Gleason scores be tween black and white men. However, black men had higher initial PSA levels (median 13.3 versus 8.6 for white men, p <0.001). The 5-year biochemical r elapse-free survival rate was 59% for the entire group, 54% (95% confidence interval 44 to 63) for black men and 61% (95% confidence interval 57 to 65 ) for white men (p = 0.11). Multivariate analysis was performed for the var iables of age, race, family history of prostate cancer (brother or father), initial PSA, biopsy Gleason sum, clinical T stage, treatment modality and androgen deprivation. Familial prostate cancer (p = 0.001), higher T stage (p <0.001), higher initial PSA (p <0.001), higher biopsy Gleason score (p < 0.001) and use of androgen deprivation (p = 0.001) were independent predict ors of biochemical failure and all other factors, including race, were not (p = 0.46). The projected 10-year clinical relapse-free survival rate was 7 4% for the entire group, and was identical for black and white men (p = 0.7 7). The projected 10-year overall survival rate for black and white men was 92 and 79%, respectively (p = 0.62). Conclusions: We have demonstrated a statistically nonsignificant trend for higher biochemical failure rates in black men presenting with localized pro state cancer. This trend could be due to the higher pretreatment PSA levels in black patients. Treatment recommendations should not differ with respec t to race.