PSA bounce after prostate brachytherapy with or without neoadjuvant androgen deprivation

To assess the impact of PSA bounce (PB) on biochemical failure (BF) and clinical failure (CF) in brachytherapy patients treated with or without neoadjuvant androgen deprivation (AD).

From 1987 to 2003, 691 patients with clinical stage T1-T3N0M0 prostate cancer were treated with external beam radiotherapy (EBRT) and high-dose-rate (HDR) brachytherapy boost (n=407), HDR brachytherapy alone (n=93), or permanent seed implant (n=191). Three hundred seventeen patients (46%) received neoadjuvant/adjuvant AD with RT. BF was scored using 3 definitions (ASTRO--3 rises, nadir+2 ng/ml, and threshold 3 ng/ml) based on current and absolute nadir (AN) methodologies. PB was defined as any increase in PSA followed by a decrease to the prior baseline or lower. The median followup was 4.0 years.

Forty-six patients (7%) experienced CF at 5 years. PB of >or=0.1, >or=1.0, and >or=2.0 ng/ml at any time after RT occurred in 330 (48%), 60 (9%), and 22 patients (3%) respectively. The use of an AN definition reduced the likelihood of scoring PB as BF across all levels. The patients receiving AD experienced significantly longer bounce duration. Bounce <1.0 ng/ml showed no association with CF. For bounce >or=1.0 ng/ml, 10% demonstrated CF vs. 6% without bounce of this amplitude (p=0.27). Bounces >or=1.0 ng/ml were more likely to be scored as BFs for definitions based on current nadir (3 rises: 20% vs. 13%, nadir+2: 43% vs. 11%, 3 at/after nadir: 57% vs. 12%) than those based on AN (3 rises: 8% vs. 10%, nadir+2: 18% vs. 11%, 3 at/after nadir: 13% vs. 11%).

Bounces >or=1.0 ng/ml are rare after brachytherapy with or without neoadjuvant AD, occurring in less than 10% of patients. Low PBs have little impact on BF, but as PB amplitude increases, the BF rate increases. BF definitions based on AN are less sensitive to PB after brachytherapy.