Using a surgical prostate-specific antigen threshold of >0.2 ng/mL to define biochemical failure for intermediate- and high-risk prostate cancer patients treated with definitive radiation therapy in the ASCENDE-RT randomized control trial

PURPOSE: To compare biochemical failure using a prostate-specific antigen (PSA) threshold of >0.2 ng/mL to that using Phoenix threshold (nadir+2 ng/mL). METHODS AND MATERIALS: Androgen suppression combined with elective nodal and dose-escalated radiation therapy (the ASCENDE-RT trial) is a randomized control trial in which 276 high-risk and 122 intermediate-risk patients were randomized to (1)a standard arm with 12 months of androgen deprivation therapy, pelvic external beam radiation therapy (EBRT) to 46 Gy, and an EBRT boost (dose-escalated EBRT [DE-EBRT]) to 78 Gy, or (2) an experimental arm which substituted a low-dose-rate prostate brachytherapy boost (LDR-PB). The primary endpoint was biochemical progression-free survival (b-PFS) using the Phoenix threshold. In this reanalysis of ASCENDE-RT, the b-PFS using phoenix is compared to the surgical PSA threshold of >0.2 ng/mL. RESULTS: Compared to nadir+2 ng/mL, the >0.2 ng/mL PSA threshold doubled the number of relapse events from 69 to 139. However, the increase was confined to the DE-EBRT subjects. The 7-year Kaplan-Meier b-PFS after DE-EBRT declined from 76% using nadir+2 ng/mL to 38% using the >0.2 ng/mL threshold (p < 0.001). Among the LDR-PB subset, there was no significant difference in b-PFS; the 7-year Kaplan-Meier b-PFS was 85% (>0.2 ng/mL) versus 88% (nadir+2 ng/mL) (p = 0.319). CONCLUSIONS: Replacing Phoenix with a surgical threshold greatly increased biochemical failure after DE-EBRT boost but had no effect after LDR-PB. As a result of this finding, PSA outcomes after surgery or brachytherapy can be directly compared by using the surgical definition of PSA failure. In this context, a brachytherapy boost appears to produce superior b-PFS compared to contemporary surgical series. (C) 2018 American Brachytherapy Society. Published by Elsevier Inc. All rights reserved.