The ability of the American Joint Committee on Cancer Staging system to predict progression-free survival after radical prostatectomy

OBJECTIVE: To examine, in a retrospective analysis of outcome based on prostate-specific antigen (PSA) levels, whether the 1992 and revised 1997 staging criteria for prostate cancer can be used to predict progression-free survival for patients after radical prostatectomy for pT2 and pT3 prostate cancer. PATIENTS AND METHODS: In all, 291 patients with a PSA determination during a 6-month interval after radical prostatectomy were analysed (mean follow-up 5.2 years). In the absence of a uniform system of pathological staging, the histopathological stage was defined according to the 1992 and 1997 American Joint Cancer Committee/Union Internationale Contre le Cancer (AJCC/UICC) tumour-nodes-metastases (TNM) staging classification. Findings were correlated with the PSA value after surgery. The subgroups of pT2 and pT3 disease were compared for the time to PSA progression, using Kaplan-Meier data analysis and the log-rank test. RESULTS: The biochemical progression-free 5-year survival rates for stage pT2 were 83% (pT2a), 81% (pT2b) and 62% (pT2c); there were no significant differences in the pT2 subgroups. The recurrence-free rates for pT3 were 79% (pT3a), 65% (pT3b) and 50% (pT3c); the actuarial recurrence-free rate was significantly different for patients with 1997 AJCC pT3a vs pT3b disease (P=0.0132). There was no significant difference in the 1992 AJCC stages pT2a vs pT2b (P=0.1232) and the recurrence-free rate was not significantly different for patients with 1992 AJCC pT3a vs pT3b disease (P=0.9). There was a significant difference in the likelihood of a PSA relapse between patients with positive and negative surgical margins (P=0.131). CONCLUSION: These results support the current revised 1997 AJCC/UICC staging system for prostate cancer. There is an urgent need to develop a pathological equivalent to the AJCC/UIC TNM clinical staging system. Greater clinical input and evaluation from different institutions are essential to reach consensus on pathological staging categories that maximize the predictability of outcome after definitive therapy. Crucial issues are the definition and quantification of extraprostatic extension and definition of surgical margin categories.