• Gleason score 8 to 10;
  • high risk;
  • prostate cancer;
  • recurrence;
  • radical prostatectomy



Men with a biopsy Gleason sum of 8 to 10 are considered high-risk. The current study sought to identify whether there was a subset of men with high biopsy Gleason sums who would have a good pathologic and biochemical outcome with surgical monotherapy. To increase the generalizability of the findings, data were used from patients treated at 2 very different practice settings: a tertiary care referral center (Johns Hopkins Hospital) and multiple equal-access medical centers (Shared Equal Access Regional Cancer Hospital [SEARCH] Database).


The data were retrospectively reviewed from men with biopsy Gleason sums 8 to 10 treated by radical prostatectomy at the Johns Hopkins Hospital (n = 220, 3.8% of total cohort) and within the SEARCH Database (n = 149, 7.7% of total cohort). The preoperative clinical characteristics predicting unfavorable pathologic disease (nonorgan-confined and/or positive surgical margins) and time to biochemical recurrence were determined using logistic regression and Cox proportional hazards analysis, respectively.


Favorable pathologic outcome (organ-confined and negative surgical margins) was observed in 21% of the men in the Johns Hopkins cohort and 41% from the SEARCH cohort. On multivariate analysis, higher serum prostate-specific antigen (PSA) was the only variable that significantly predicted an unfavorable pathologic outcome from both the Johns Hopkins (P = .047) and SEARCH cohorts (P = .002). The 5-year and 10-year estimated biochemical-free survival rates in the Johns Hopkins cohort were 40% (95% confidence interval [CI], 33–48%) and 27% (95% CI, 18–36%), respectively, and 32% (95% CI, 22–42%) and 28% (95% CI, 18–38%) in the SEARCH cohort, respectively. Among men with favorable pathologic findings, the 5- and 10-year estimated biochemical-free survival rates in the Johns Hopkins cohort were 79% (95% CI, 62–89%) and 50% (95% CI, 25–71%), respectively, and 49% (95% CI, 32–65%) and 49% (95% CI, 32–65%) in the SEARCH cohort, respectively. No single preoperative variable significantly predicted the risk of biochemical progression in both the SEARCH or Johns Hopkins cohorts.


The majority of men with a biopsy Gleason sum of ≥8, regardless of where the patient is treated, had unfavorable pathologic disease and experienced a biochemical progression after radical prostatectomy. Even among men with organ-confined disease and negative surgical margins or pathologic Gleason sum <8, at least half of the men experienced a PSA recurrence. Patients with biopsy Gleason sum 8 to 10 cancers are good candidates for multimodal therapy. Whereas multimodal therapy has often meant radiation plus hormonal therapy, newer possibilities for multimodal therapy exist such as surgery with neoadjuvant or adjuvant chemohormonal therapy or surgery with adjuvant radiation. Cancer 2006. © 2006 American Cancer Society.