Prostate-specific antigen doubling-time (PSA DT) before onset of chemotherapy as survival predictor for hormone refractory prostate cancer (HRPC) patients

International audienceBackground: PSA DT has been reported as a prognostic factor in prostate cancer patients (pts) who relapse after radical prostatectomy, radiation or hormonal therapy. The rate at which the PSA is rising is known to correlate with cancer aggressiveness. We evaluated the interest of the initial PSA DT value, before start of chemotherapy (CT), as a surrogate end point for survival of metastatic HRPC pts. Methods: PSA DT was calculated as log x 2 divided by the slope of the log PSA line (the difference in the 2 log PSA values divided by the time between). A minimum of three PSA measurements were required, maximum three months before start of CT. The primary outcome measure was overall survival. Differences in survival rates and PSA DT were calculated using the log-rank and Cox hazard regression methods, stratified according to CT regimen. All statistical tests were two-sided. Hazard ratio (HR) and relative risk reduction were also estimated with 95% confidence interval (CI). Results: A retrospective analysis was performed on 202 metastatic HRPC pts, registered from June 1999 to July 2004 in a single institution. Pts received as first-line CT: docetaxel- (77%) or mitoxantrone (23%) based regimen. The median PSA DT value was 44 days (range 5-1028 days) and median interval between two consecutive PSA determinations before CT (used tu estimate the PSA DT) was 21 days. Overall, there were 124 deaths (61%). Median survival was 14.3 months (95% CI, 10.5 to 18 months) for pts with PSA DT 44 days. There were significant differences between survival distributions of PSA DT categories (log-rank, p=0.006), independently of CT regimen. The stratified HR was 0.61 (95% CI, 0.43 to 0.87) with a relative risk reduction of 39% (95% CI, 13% to 57%). Conclusions: A low PSA DT before onset of CT in HRPC pts was associated with an increased risk of death. This parameter could be a potential auxiliary end point in the evaluation of new cytotoxic drugs in metastatic HRPC setting. These data may be also useful in the design of clinical trials and the identification of men for enrolment into experimental protocols