Prognostic significance of race on biochemical control in patients with localized prostate cancer treated with permanent brachytherapy: a matched pair analysis

Abstract

Purpose: To evaluate PSA relapse-free survival (PSA-RFS) via a matched pair analysis between African American (AA) and White (W) men treated with permanent prostate brachytherapy (PPB) for clinically localized prostate cancer. Materials and Methods: Eight hundred thirty- five W and 246 AA consecutive patients (pts) comprise this study cohort treated between 9/92 and 9/99. All patients were treated with PPB with either Pd-103 or I-125 either as monotherapy or in conjunction with external beam radiatiotherapy (EBT). Risk group assignment was based on a pretreatment PSA value <10 ng/ml and Gleason score <7 (low risk: PSA <10, Gleason <7, Intermediate risk PSA > 10 OR Gleason >6, High risk: PSA > 10 AND Gleason >6). Presenting characteristics were compared by the Mann-Whitney test. PSA-RFS was calculated using the Kattan modification of the ASTRO definition. Cox multivariate hazards analysis was performed. A computer generated matching was performed to create 2 identical cohorts of W and AA pts, respectively. Kaplain-Meier PSA-RFS curves were compared via the log rank test. Results: AA pts were found at presentation to our clinic to have lower stage disease (p=0.01), were younger (p=0.001), had lower Gleason sum (p=0.017) and were more likely treated with neoadjuvant androgen ablation (NAAD; p=0.001) as compared to the W pts. No difference in the pretreatment PSA value, isotope selection, the addition of EBT, median follow-up and risk group classification was found. Multivariate analysis found pretreatment PSA value and Gleason sum score significant to predict PSA-RFS, while race was an insignificant factor. The matching process was performed creating identical cohorts of 246 pts each based on race by the following factors: the use of NAAD, Gleason score sum and pretreatment PSA value. The PSA-RFS at 5-years was 84.0% for the AA Pts and 81.2% for the W Pts (p=0.384). No significant difference based on race was observed for 5-year PSA-RFS between low, intermediate and high-risk patients for 5-year PSA-RFS. There was no difference in 5-year PSA-RFS based on race for those patients treated with or without NAAD from the matched cohort. Conclusion: This study presents the largest cohort of AA men treated with PPB to date. The AA cohort of pts presenting to our clinic did so with lower stage and lower Gleason score disease and were more likely to be treated with NAAD. Race was not an independent predictor of PSA-RFS in pts treated with PPB. Furthermore, a matched pair analysis that corrected for the use of NAAD, the Gleason score sum and the pretreatment PSA value failed to identify any difference in PSA-RFS based on race. This outcome in men treated with PPB is similar to other studies that have identified that race is not an independent factor for biochemical outcome with localized prostate cancer.