Clinical and pathologic predictors of Gleason sum upgrading in patients after radical prostatectomy: Results from a single institution series

Objectives rative Gleason scores (GSs) are often upgraded after pathologic examination of the prostate following radical prostatectomy (RP). There have been disparate reports of the impact of different factors as predictors of GS upgrading after RP. We sought to study the robustness of frequently reported predictors in an unselected single institution cohort. ts and methods l of 684 patients with biopsy-proven prostate cancer treated with RP between 2004 and 2007 were included in the study. The association between clinical and pathologic parameters and GS upgrading was retrospectively evaluated. Logistic regression analysis was used to identify predictors of pathologic grading changes. Likelihood of upgrading was compared between tertile groups for prostate volume and prostate-specific antigen (PSA) density using χ2 analysis and multivariate logistic regression. Pathologic outcomes were compared between cases with and without GS upgrading. s erall mean age was 64.3 years, with median PSA level of 7.04 ng/ml. Overall, 203 cases (29.7%) were upgraded, whereas 481 patients (70.3%) were downgraded or had identical biopsy and pathologic GS after RP. Patients with prostate volume of <31 g were upgraded in 32.6% of the cases compared with 21.9% in patients with prostate volume of >45 g (P = 0.020). On multivariate analysis preoperative PSA (P < 0.0001), prostate volume (P < 0.0001), and PSA density (P < 0.0001) were predictive of Gleason sum upgrading. Upgraded patients were more likely to have extracapsular extension, seminal vesicle invasion, positive surgical margins, and lymphonodular invasion at RP (P < 0.001, P < 0.001, P < 0.001, and P < 0.001, respectively). sions r prostate volume and higher PSA level are associated with clinically significant upgrading of GS. PSA density as a function of both is a significant predictor of GS upgrading in low- and high-risk patients. This may be of relevance in the pretreatment risk assessment of prostate cancer patients.