The impact of clinical stage on prostate cancer survival following radical prostatectomy

Purpose: Clinical stage has been incorporated into multiple risk stratification models for patients with newly diagnosed prostate cancer. However, the independent prognostic value of this variable remains open to debate. In this study we evaluated the association of clinical stage with death from prostate cancer in men who underwent radical prostatectomy and assessed for changes in its prognostic value over time. Materials and Methods: We reviewed the records of 14,842 consecutive patients who underwent radical prostatectomy at our institution between 1970 and 2008 without having received preoperative hormone or radiation therapy. Postoperative disease recurrence was estimated using the Kaplan-Meier method and compared using the log rank test. Multivariate Cox proportional hazard regression models were used to analyze the association of clinical stage with outcome. Results: A total of 5,725 (38.6%) men were classified as having cT1 tumors, 8,160 (55.0%) cT2 tumors and 957 (6.4%) cT3 disease. On univariate analysis clinical stage was significantly associated with postoperative biochemical recurrence, systemic progression and death from prostate cancer (p <0.001 for each). Moreover on multivariate analysis clinical stage was significantly associated with death from cancer for patients treated before (1.45, p = 0.006) and those treated during (1.96, p <0.001) the prostate specific antigen era. Furthermore, the incorporation of clinical stage into contemporary risk stratification improved the prediction of cancer specific survival (c statistic 0.782 without and 0.802 with clinical stage). Conclusions: Clinical stage is significantly associated with systemic progression and death from prostate cancer. Inclusion of this variable in multivariate prediction models improves the prediction of systemic progression and cancer specific survival.