PSA doubling time (PSADT) and serum testosterone (T) during intermittent androgen deprivation (IAD) in patients with biochemically relapsed prostate cancer (BRCP; M0): Potential predictive implications.

Abstract

150 Background: The systemic management of patients with BRPC remains controversial. IAD is commonly employed. Aims: To evaluate the PSA dynamics and serum T in pts with BRPC treated with IAD until the development of PSA refractoriness or clinical evidence of metastatic disease. Methods: Data were retrospectively analyzed in all pts with BRCP treated with GnRH at PSA thresholds according to pre-treatment PSADT (10-15ng/mL, 15-20, and 20-30 for PSADT ≤3 mos, 3-9 mos, and ≥ 9 ms, respectively) and continued until PSA nadir. Antiandrogen (AA) was added for PSA > 1.0 ng/mL after 3 mos). Follow-up (FU) consisted of PSA and T q3 mos. Cycles were repeated at the above preselected PSA thresholds and continued until lack of PSA response. Scans were obtained prior to cycles and at the time of CRPC state. Mixed effects model was used to study PSADT change over cycles. Multivariate cox regression model was used to identify prognostic variables. Results: From 1995-2010, with a mean FU of 71 mos (range 22-183 months), 96 pts received a mean of 2.8 cycles (range 1-9) of IAD; 58 (60%) remain on treatment and 38 (40%) were switched to continuous ADT due to PSA refractoriness (n=11) or positive scans (n=27). PSADT at the first off treatment (tx) interval (mean 3.1, 0.59-30.5 range, median 2.3) was significantly shorter than the baseline (p<0.0001; mean 9.7, range 0.27-53.9, median 7.34) but remained relatively stable (p=0.29) in subsequent cycles. PSADT adjusted for T recovery (≥3 ms after T recovery to ≥ 150 ng/dL) was significantly longer (p=0.006) than that based only on all PSA determinations (mean 5.4, range 1.31-30.5, median 3.7 versus mean 3.1, range 0.59-30.5, median 2.3). Significant factors associated with probability of PSA refractoriness were pre-IAD PSADT (≥ 6 vs <6 ms), first off tx interval PSADT (≥3 ms vs <3m), the use of AA during first tx cycle, and PSA nadir during the first tx interval (<0.1 vs ≥0.1 ng/mL). Conclusions: Our data suggest that PSADT becomes shorter after the initial cycle of IAD and correlate with T recovery. PSA dynamics and need for AA to enhance PSA nadir are associated with PSA refractoriness in pts BRPC treated with IAD. No significant financial relationships to disclose.