The prognostic role of the pathological T2 subclassification for prostate cancer in the 2002 Tumour-Nodes-Metastasis staging system.

OBJECTIVE: To evaluate the prognostic role of the 2002 Tumour-Nodes-Metastasis (TNM) pT2 subclassification for biochemical recurrence (BCR) after radical prostatectomy (RP) for prostate cancer. PATIENTS AND METHODS: The 1997 TNM staging system is based on one subdivision for organ-confined prostate tumours (T2a, unilateral; T2b, bilateral involvement). The 2002 TNM staging system subdivides unilateral involvement into T2a (half of one lobe or less) and T2b (more than one half of one lobe), while bilateral involvement is classified as T2c. In all, 542 patients were treated with RP at our institute; the RP specimens were completely embedded and histopathologically evaluated for Gleason grade, tumour volume and anatomical extent, and were staged according to the 2002 TNM staging criteria. Patients were followed for a median of 39.5 months. BCR was defined as two subsequent increasing prostate-specific antigen (PSA) levels of >0.10 ng/mL. Kaplan-Meier and proportional hazards regression analyses were used to evaluate the univariable and multivariable prognostic effect of tumour stage. RESULTS: According to the 2002 TNM staging system, 360 specimens were found to have pT2 tumours; 79 (22%) of the RP specimens were staged as pT2a and 281 (78%) as pT2c; no pT2b specimens were identified. Patients with unilateral involvement (pT2a) had a 5-year risk of BCR of 13%, while those with bilateral involvement (pT2c) had a risk of 23% (log rank test, P = 0.056). Patients with pT2c disease were more likely to have a larger tumour volume (Mann-Whitney U-test P < 0.001) and positive surgical margins (Fishers' exact test, P = 0.001)than those with pT2a tumours. Mann-Whitney U-tests showed no differences between the groups for preoperative PSA levels (P = 0.167). Also, the RP Gleason score was no different between groups (Pearson chi-square, P = 0.807). In the multivariable analyses, positive surgical margins appeared to increase the risk of BCR (hazard ratio 4.4, 95% confidence interval 2.5-7.9); pT2c vs pT2a had only a marginally (insignificant) additional effect (1.3, 0.6-2.7). CONCLUSION: The absence of a true unilateral pathological T2b tumour in a series of 360 pT2 RP specimens questions the relevance of substaging unilateral disease. The limited differences in BCR and in pathological features of unilateral vs bilateral pT2 prostate cancer justify modifying the TNM staging system to one with no subclassification of pT2 disease, or at most as only one subdivision into unilateral (T2a) and bilateral (T2b) disease, combining the T2b and T2c substages