Analysis of prostate-specific antigen bounce after I(125) permanent seed implant for localised prostate cancer.

ABSTRACT

BACKGROUND AND PURPOSE: To report on the incidence of benign prostate-specific antigen bounce following permanent I(125) prostate brachytherapy, to describe the associations in our population and review the relationship of bounce to subsequent biochemical failure. MATERIALS AND METHODS: From February 2000 to May 2005, 374 patients with localised prostate cancer were treated with I(125) permanent prostate brachytherapy at a single institution. A prospectively collected database was used to identify cases of prostate-specific antigen (PSA) bounce, defined as a rise of 0.2 ng/ml above an initial PSA nadir with subsequent decline to or below that nadir without treatment. The patients who received neo-adjuvant or adjuvant hormone manipulation were excluded. Biochemical failure was determined using the both the ASTRO consensus definition and Phoenix (nadir +2 ng/mL) definition. RESULTS: Two hundred and five patients were identified with a median follow-up of 45 months (24-85). PSA bounce was noted in 79 (37%) men, occurring at a median of 14.8 months (1.7-40.6) following implant. The median peak PSA was 1.8 ng/ml (0.4-7.4) with a bounce magnitude of 0.91 ng/ml (0.2-5.8). When pre- and post-implant factors were assessed for association to bounce, only younger age was statistically significant (p=0.002). The threshold for biochemical failure as defined by the ASTRO consensus definition (1997) was met in 4 (5%) patients after experiencing bounce as opposed to 19 (15%) non-bounce patients (p=0.01). The threshold for Phoenix (nadir +2 ng/mL) was met in 6 (7.5%) patients following bounce versus 22 (17%) of non-bounce patients (p=0.003). Both definitions are prone to false positive calls during bounce. Median PSA velocity during the bounce was 0.08 ng/mL/month (0.02-0.98) and was statistically significantly lower than the median velocity prior to the Phoenix biochemical failure at 0.28 ng/mL/month (0.07-2.04) (p=0.0005). CONCLUSION: PSA bounce is a common finding in our population and is associated with a lower rate of subsequent biochemical failure. The noted differences in PSA velocity will require verification in a future analysis to reduce the influence of median follow-up on this finding. Patients should be advised of the potential of bounce in PSA follow-up after permanent I(125) prostate brachytherapy and physicians involved in follow-up of prostate brachytherapy patients should be aware of this phenomenon, allowing them to commit to appropriate PSA surveillance, avoiding the premature and inappropriate initiation of salvage therapy during PSA bounce.