Prognostic Factors Influencing Prostate Cancer-Specific Survival in Non-Castrate Patients with Metastatic Prostate Cancer

BACKGROUNDIn non-castrate prostate cancer (PCa), the prognostic value of the number of metastases on prostate cancer-specific survival (PCSS) is not well studied. METHODSWe retrospectively analyzed the medical records of 1,206 patients, referred for radiotherapy of the prostate (bed) following diagnosis of PCa. Distant metastases (nodal, skeletal, and/or visceral) developed in 121 patients following curative treatment, of which 80 with complete records were not castrated at time of metastasis. The treatment at time of metastases was androgen deprivation therapy (ADT; n=22), active surveillance (n=10) or metastasis-directed therapy (MDT; n=48). Cox-regression analyses were used to examine the influence of different variables on PCSS. RESULTSThe median follow-up from primary PCa treatment was 6.9 years with a median interval from diagnosis to first metastatic event of 4.1 year (range: 0.2-15 years). The primary site of metastases was limited to lymph nodes (48%), bone (39%), and viscera (1%) or a combination (12%). Median PCSS from diagnosis of noncastrate metastases was 6.6 years (95% confidence interval [CI], 5.6-7.7 years). A longer premetastatic PSA doubling time (DT) (hazard ratio [HR] 0.73; 95% CI: 0.57-0.92), a lower number of metastases at first presentation (HR 1.07; 95% CI: 1.02-1.12) and pattern of metastatic spread (HR 3.6; 95% CI: 1.13-11.8 for extensive vs. minimal) were associated with improved PCSS. CONCLUSIONA longer PSA DT, involvement of nodes or axial skeleton and a lower number of metastases are associated with an improved PCSS in non-castrated patients developing metastases. Prostate 74:297-305, 2014. (c) 2013 Wiley Periodicals, Inc.