[860] Clinical Stage Drives Outcome at Radical Prostatectomy (RP) with Gleason Score 9-10 (GS910) Prostatic Adenocarcinoma on Biopsy

Carla L Ellis, Jonathan I Epstein. Johns Hopkins Hospital, Baltimore, MD

Background: There is little data on the prognosis at RP when there is GS910 on biopsy.
Design: We identified 259 men (1987-2012) with GS910 on biopsy that underwent RP at our institution. We categorized age, race, pre-operative PSA (prePSA) level, location of adenocarcinoma, # of total biopsy cores, # of total positive cores, # of positive cores with GS910, maximum % of core length with GS910, maximum % of adenocarcinoma overall, and biochemical free survival (BFS) at 3 years. Clinical stages were combined into the following groups: Cstage1=T1c, T2a or T2b; Cstage2 = T2c; & Cstage3 = >T3. Pathological stage was grouped as: Organ confined (OC); Extra-prostatic extendion (EPE); Seminal vesicle invasion (SV); Lymph node (LN) metastases, based on the highest stage finding in a case (ie. case with SV & EPE designated as SV).
Results: 243/259 (94%) were GS9. The mean age was 60 yrs. and the majority were Caucasian. The mean prePSA level = 8.9, mean # of total cores = 12, mean # positive cores = 4, mean # of positive cores with GS910 = 2, mean maximum % adenocarcinoma per core = 66%; mean maximum % GS910 per core = 56%. At RP, 66% of men had GS910. Only 29% had OC. Many of the single variables were predictive of RP stage. Cstage was the best predictor at RP, with Cstage1 or 2 having a relatively favorable prognosis.

Table 1
CstageTotal %%OC% Margins (+)% LN (+)% SV (+)% BFS

Although multiple single variables (prePSA and various measures of extent of cancer on biopsy) predicted OC in univariate analysis, only clinical stage was predictive in multivariate analyses. The only factors predictive in uni- and multivariate analysis of positive margins were prePSA (p=0.02) and Cstage (p<0.001), with the latter being much more predictive. In terms of SV and LN, Cstage was overwhelmingly predictive.
Conclusions: It is rare for patients with GS10 to be cadidates for RP due to advanced disease. Even then, most patients who are candidates for RP with GS910 on prostate biopsy have clinical T3 disease and a poor outcome at RP. However, for the highly selected small number of patients with clinical T2 disease who are candidates for RP, there is increased likelihood for a favorable pathological outcome at RP and a low risk of BFR at 3 years after RP. Additional pathological measures of tumor exent on biopsy are not further predictive, such that patients with mulitple positive cores of GS910 were still likely to have OC disease if their Cstage was low.
Category: Genitourinary (including renal tumors)

Monday, March 4, 2013 1:15 PM

Proffered Papers: Section A, Monday Afternoon


Close Window