Long-term cancer control outcomes in patients with biochemical recurrence and the impact of time from radical prostatectomy to biochemical recurrence

BACKGROUND: Rates of metastatic progression (MP) and prostate cancer mortality (PCSM) are variable after biochemical recurrence (BCR) in patients who underwent radical prostatectomy (RP). To describe long-term oncological outcomes of BCR patients and to analyze risk factors for further outcomes in these men with a special focus on RP-BCR time.

METHODS: We retrospectively analyzed the data of 5509 RP patients treated between 1992 and 2006. Of those, we included 1321 patients who experienced BCR (PSA level ≥0.2 ng/mL) and did not receive any neoadjuvant or adjuvant therapy. Kaplan-Meier and time dependent Cox regression models were used.

RESULTS: Median follow-up was 121 months. MP was recorded in 177 (13.4%), PCSM in 126 (9.5%), and overall mortality (OM) in 264 (20.0%) patients. Patients with MP had worse tumor characteristics such as higher Gleason Scores (GS), rapid PSA doubling-time (DT), and shorter RP-BCR time intervals. MP-free, PCSM-free, and overall survival rates were significantly worse in patients with RP-BCR time of <12 months versus patients with 12-35.9 or ≥36 months (P ≤ 0.001). Besides higher GS and rapid PSA-DT, RP-BCR time independently predicted MP, PCSM, and OM in multivariable regression analyses. Relative to the intermediate and longest RP-BCR time interval, the shortest interval (<12) carried the highest risk for all three endpoints.

CONCLUSIONS: Only a small proportion of BCR patients proceed to MP or PCSM. Besides higher GS and rapid PSA-DT a shorter RP-BCR interval (<12 months) heralds the most aggressive phenotype for progression to all three examined endpoints: MP, PCSM, and OM.