Evaluating the Phoenix definition of biochemical failure after (125)I prostate brachytherapy: Can PSA kinetics distinguish PSA failures from PSA bounces?

Abstract

To evaluate the prostate-specific antigen (PSA) kinetics of PSA failure (PSAf) and PSA bounce (PSAb) after permanent (125)I prostate brachytherapy (PB). The study included 1,006 consecutive low and "low tier" intermediate-risk patients treated with (125)I PB, with a potential minimum follow-up of 4 years. Patients who met the Phoenix definition of biochemical failure (nadir + 2 ng/mL(-1)) were identified. If the PSA subsequently fell to ≤0.5 ng/mL(-1)without intervention, this was considered a PSAb. All others were scored as true PSAf. Patient, tumor and dosimetric characteristics were compared between groups using the chi-square test and analysis of variance to evaluate factors associated with PSAf or PSAb. Median follow-up was 54 months. Of the 1,006 men, 57 patients triggered the Phoenix definition of PSA failure, 32 (56%) were true PSAf, and 25 PSAb (44%). The median time to trigger nadir + 2 was 20.6 months (range, 6-36) vs. 49 mo (range, 12-83) for PSAb vs. PSAf groups (p < 0.001). The PSAb patients were significantly younger (p < 0.0001), had shorter time to reach the nadir (median 6 vs. 11.5 months, p = 0.001) and had a shorter PSA doubling time (p = 0.05). Men younger than age 70 who trigger nadir +2 PSA failure within 38 months of implant have an 80% likelihood of having PSAb and 20% chance of PSAf. With adequate follow-up, 44% of PSA failures by the Phoenix definition in our cohort were found to be benign PSA bounces. Our study reinforces the need for adequate follow-up when reporting PB PSA outcomes, to ensure accurate estimates of treatment efficacy and to avoid unnecessary secondary interventions.