107: Comparison of Biochemical Failure Rates Between Permanent Prostate Brachytherapy and Radical Retropubic Prostatectomy Patients as a Function of Post-Therapy PSA Nadir Plus ‘X’

Abstract

Ongoing efforts are examining biochemical failure definitions (BFD) for prostate cancer patients (pts) treated with external beam radiation therapy (EBRT) and/or permanent prostate brachytherapy (PPB) that correlate with subsequent clinical failure. Recent studies have shown that serum prostate specific antigen (PSA) nadir plus a defined subsequent rise may be useful in this effort. The purpose of this study is to compare biochemical failure (BF) between PPB and radical retropubic prostatectomy (RRP) pts as a function of a varying nadir + ‘x’. Between 1998 and 2004, a total of 613 pts underwent PPB at our institution and 5,821 pts underwent RRP. Only primary pts not receiving androgen deprivation therapy (ADT) prior to or during therapy were included in the study. Three RRP pts were subsequently matched to each PPB pt for biopsy Gleason, stage, PSA, age, and year of procedure. Nadir was defined as the lowest PSA achieved within 3 years of treatment. Median time to nadir was 3 months after RRP and 2.4 years after PPB. Subsequent failure was then defined based on PSA nadir plus ‘x’, where x varied from 0.1-5 ng/ml. Pts were also considered failures if they did not have a post-therapy nadir but did have the defined PSA rise or if ADT was implemented for a rising PSA following therapy. A total of 1,164 pts were used for analysis: 873 RRP pts and 291 PPB pts. Pts. were equally matched by clinical stage (p=0.47) (T1=796; T2=368), biopsy Gleason sum (p=0.74) (≤6=1,019; 7=140; > 8=5), primary Gleason (p=0.25), age at treatment (p=0.11), and pre therapy PSA (p=0.95), median = 5.8 ng/ml with interquartile range of 4.1-7.6 ng/ml. Median follow-up was 3.1 yrs in the RRP pts and 3.6 yrs in the PPB group (p=0.01). The rate of systemic progression was 0.7% overall and not significantly different between groups (p=0.1). Comparison of BF rates by modality and nadir + ‘x’ are shown in the figure. At nadir + 0.1 ng/ml the 5-year BF rate(SE) for PPB pts was 16.3(2.3)% whereas with RRP pts it was 13.5(1.5)% (p=0.007), nadir + 1.0 ng/ml it is 6.7(2.1)% versus 4.9(1.1)% (p=0.086), but at nadir + 2 or greater the BF rates are <3% and indistinguishable. No differences in systemic progression or cause specific mortality have been observed in these two well-matched cohorts having median followup < 4 years. Five-year BF rates become similar for the two cohorts when using a nadir+2ng/ml or greater definition of BF. Further followup of this cohort is required in order to determine if such a BFD can be considered as a surrogate definition for clinical failure and for early comparison of RRP and PPB outcomes.