Bladder neck involvement in pathological stage pT4 radical prostatectomy specimens is not an independent prognostic factor

PURPOSE: Bladder neck invasion by prostate cancer in radical prostatectomy specimens is uncommon and, thus, its influence on disease recurrence has not been well defined. Consequently the classification of bladder neck invasion in the TNM staging system is controversial. We studied our cohort of patients with stage pT4 disease and bladder neck invasion to clarify the true clinical behavior and prognostic significance of bladder neck invasion in radical prostatectomy specimens. MATERIALS AND METHODS: The study group consisted of 4,090 consecutive patients treated with radical prostatectomy at one of our institutions between 1983 and 2001. Median followup was 53.1 months (range 1 to 189). After excluding from analysis patients treated with neoadjuvant androgen withdrawal or preoperative irradiation 72 of the remaining 2,571 (2.8%) with bladder neck invasion were classified with stage pT4 disease and their specimens were reviewed. Progression-free probability was determined by Kaplan-Meier analysis. Using the Cox proportional hazards model the independent prognostic significance of bladder neck invasion was assessed after controlling for pretreatment prostate specific antigen, final Gleason sum, extracapsular extension, surgical margins status, seminal vesicle invasion and lymph node involvement. RESULTS: Of the 72 patients categorized with stage pT4 disease 14 (19%) had poorly differentiated Gleason sum 8 to 10 cancer, 38 (53%) had established extracapsular extension, 24 (33%) had seminal vesicle invasion and 8 (11%) had lymph node involvement. However, 26 patients (36%) had cancer confined to the prostate and 28 (39%) had negative surgical margins except for the bladder neck site. The mean 5-year progression-free probability plus or minus SD in all stage pT4 cases was 68% +/- 7%, which was better than in cases of seminal vesicle invasion (52% +/- 5%, log rank test p = 0.0156) but worse than in those of extracapsular extension (84% +/- 4.1%). Univariate analysis of the stage pT4 cohort revealed that higher prostatectomy Gleason sum, more extensive extracapsular extension and seminal vesicle invasion were significantly associated with an adverse prognosis. However, in a multivariate model that included all radical prostatectomy cases the finding of bladder neck invasion or stage pT4 disease did not independently predict prostate specific antigen recurrence. CONCLUSIONS: Stage pT4 disease comprises a heterogeneous group of tumors with various pathological features and inconsistent outcomes. Assigning the pT4 stage to cases of microscopic bladder neck invasion provides no independent ability for predicting disease progression after adjusting for other adverse disease features. Due to this and previously reported data the definition of stage pT4 disease should be modified in the next version of the TNM staging system.