An Examination of Clinical Outcomes in Patients Experiencing Biochemical Relapse in a Cohort of 966 Permanent Prostate Brachytherapy Monotherapy Patients

Abstract

Low rates of prostate specific antigen (PSA) relapse are observed following permanent prostate brachytherapy monotherapy (PPBM). We report a large single institution experience of outcomes following such relapse. We examined 966 consecutive pts treated with I-125 PPBM between 4/98 and 1/13 with at least one post PPBM serum PSA to identify pts who experienced PSA relapse by the nadir + 2 (Phoenix) criteria. Survival estimates used the Kaplan-Meier method. Characteristics of the 966 pt cohort include median age of 69 years (range 42-86) and NCCN risk group of low (n=685, 71%) or intermediate (n=281, 29%). Overall survival (OS) at 5 and 10 years was 94% (95% confidence interval (CI) 92-95) and 74% (95% CI 69-78). Ten year prostate cancer-specific mortality (PCSM) was 2% (95% CI 0-3) overall and 1% (95% CI 0-3) vs 4% (95% CI 0-8, p=0.02) for low vs intermediate risk prostate cancer. Freedom from PSA relapse at 5 and 10 years is 97% (95% CI 95-98) and 85% (95% CI 80-90). PSA relapse was detected in 45 pts at a median of 52 (IQR 36-67) months after PPBM with a median PSA at relapse of 3.9 (IQR 2.7-6) ng/dL. Characteristics of pts with relapse include NCCN low risk (n=20) or intermediate risk (n=25) at initial diagnosis. Relapse was found to be local (n= 10, 22%), nodal (n=9, 20%), local and nodal (n=1, 2%), distant (n=9, 20%), local and distant (n=1, 2%) or biochemical only (n=15, 33%). Positive prostate biopsy was found in 12 of 23 (52%) (representing 12 local relapses in 966 pts (1.2%)). Biopsy pathology showed the same Gleason score (n=3), higher Gleason score (n=8) or new neuroendocrine features (n=1) compared to initial diagnosis. Locoregional salvage therapy for local and/or nodal relapse was prostatectomy and lymph node dissection (n=4), external beam radiotherapy (n=2), cryotherapy (n=1) or no local treatment (n=13). Locoregional therapy without (n=3) androgen deprivation therapy (ADT) resulted in persistent PSA in 1 pt and undetectable PSA for >72 months in 2 pts. Locoregional therapy with ADT (n=4) resulted in PSA relapse in 1 pt at 36 months following salvage treatment and no PSA relapse at 5, 7 and 21 months in the remaining pts. ADT was given to 28 pts a median of 6 (IQR 3-17) months after PSA relapse with subsequent PSA decline in all pts. Median time to disease progression after ADT initiation was 36 months. One patient was treated with chemotherapy (carboplatin) as the initial management strategy after PSA relapse. Median OS after PSA relapse for patients with low or intermediate risk disease was 74 vs 52 months (p=0.07), respectively. Mortality occurred in 18 pts. Cause of death was prostate cancer (n=7), lung cancer (n=2), non-malignant causes (n=6) or unknown causes (n=3). PCSM occurred in pts with metastatic disease at PSA relapse (n=3), nodal disease (n=3) or in pts with PSA relapse only who subsequently developed evidence of metastases after progression on ADT (n=2). Of 10 pts with local relapse only, mortality occurred in 4 pts from metastatic lung cancer (n=2) or neurodegenerative disease (n=2). PCSM at 5 years after PSA relapse for the whole cohort is 40% (95% CI 21-64) and is 27% (95% CI 9-59) vs 63% (95 CI 25-90, p=0.06) for low and intermediate risk pts, respectively. Excellent long term results for patients undergoing PPBM similar to other reports are observed. Patients experiencing biochemical relapse have a decrement in subsequent PCSM. Consequently, these data may provide insight into selection of management approaches.