The impact of race on biochemical outcome in patients receiving irradiation for prostate cancer

Purpose/Objective: African-Americans tend to present with a higher stage and grade prostate cancer than whites and hence previous studies have attempted to delineate the importance of race in outcome with radiotherapy. However, these studies have had limitations including insufficient number of African-American patients, lack of a similar quality of care or uniform treatment policy. In addition, race as a prognostic variable has not been analyzed in regards to PSA based outcome criteria. The current study was performed in order to determine the impact of race on survival and biochemical control in patients with prostate cancer treated at a single center using a standardized radiation protocol. Materials and Methods: Between 1988 and 1995, 455 patients with clinically localized adenocarcinoma of the prostate received external beam irradiation for a median dose of 68 Gy using a four field technique. Of the 455 patients, 211 were African-American and 244 were white. Pretreatment PSA were: 0-4 ng/ml (51), 4-10 ng/ml (129), 10-20 ng/ml (117), > 20 ng/ml (136), unknown (22). Clinical stages were: T1 (108), T2 (238), T3 (99), not available (10). There was no significant difference in pretreatment characteristics (stage, grade and PSA) or radiation dose between the African-American and white group of patients. Median follow-up is 37.8 months. Biochemical failure was defined as two or more consecutive PSA values that are greater than the posttreatment nadir. Race, pretreatment PSA, grade, age, stage and dose were assessed with univariate and multivariate Cox regression analysis as prognostic factors for overall survival and biochemical disease free survival. Results: The 5 year actuarial overall survival (OS) was 79% and biochemical disease free survival (bNED) was 35% for the entire group of patients. There was no significant difference in 5 year OS (71% vs. 85%) (p=0.3) or bNED (26% vs. 40%) (p=0.26) for African-Americans in comparison to whites. Univariate analysis revealed significance only for stage (p=0.02) and pretreatment PSA (p=0.01) in regards to OS. Grade (p=0.02) and pretreatment PSA (p<0.005) were significant determinants for bNED. In the multivariate regression analysis, only pretreatment PSA (p=0.01) remained significant in regards to bNED. The 5 year OS, with respect to pretreatment PSA, for whites in comparison to African-Americans was: 65% vs. 93% (0-4 ng/ml), 90% vs. 100% (4-10 ng/ml), 82% vs. 69% (10-20 ng/ml) and 86% vs. 54% (> 20 ng/ml). The 5 year bNED, with respect to pretreatment PSA, for whites in comparison to African-Americans was: 92% vs. 50% (0-4 ng/ml), 64% vs. 41% (4-10 ng/ml), 31% vs. 52% (10-20 ng/ml) and 22% vs. 14% (> 20 ng/ml). The risk ratio for OS and bNED for white patients relative to African-Americans in the multivariate analysis was 0.78 (p=0.58) and 0.79 (p=0.4) respectively. Conclusion: This is the largest study from a single center of African-American patients treated in a uniform manner with a similar quality of care. There is a trend toward poorer biochemical disease free survival for African-Americans in comparison to whites which persists upon stratification by pretreatment PSA although not statistically significant. The relative risk for biochemical failure is greater by a factor of 1.3 for African-Americans in comparison to the white patients, again not statistically significant. Therefore, African-Americans do not appear to have a more aggressive disease course, in comparison to white patients, for a similar extent of disease; however, a longer follow-up is needed