Purpose: In the 1997 the TNM staging system for prostate cancer was changed, reclassifying, T2 cancers from 3 groups (T2a, less than one half of one lobe; T2b, one lobe; and T2c, both lobes) to 2 groups (T2a, one lobe; and T2b, both lobes), combining the 1992 T2a and T2b into the 1997 T2a subclassification. We investigated the pathological stage and prognosis of cancers in the 1992 and 1997 subclassification to determine whether this change was warranted.
Material and methods: We studied a consecutive series of 555 patients with clinical stage T2 prostate cancer treated with radical prostatectomy (RP) between 1983 and 1998. We analyzed the clinical, pathological features and PSA non-progression rate after prostatectomy for patients classified according to the 1992 and the 1997 TNM system. Median follow-up was 51.3 months.
Results: In the 1992 TNM system T2a tumors were more likely to have a low PSA (5.8 versus 7.2 and 8.1 ng/ml, p = 0.034, p = 0.012), be confined to the prostate (67% versus 45% and 40%, p < 0.001 for both), be poorly differentiated (48% versus 63% and 66%, p = 0.002 for both) and have a low cancer volume (1.22 versus 2.27 and 2.63 cm3, p = 0.005 for both) than T2b and T2c tumors. But there were no significant differences between T2b and T2c. Reflecting these results, the patients with T2a cancer had a significantly better prognosis with 82 +/- 4% PSA non-progression rate at 5 years compared to 68 +/- 4% of patients with T2b and 73 +/- 4% of patients with T2c (p = 0.007, p = 0.048, respectively). In the 1997 TNM system T2a tumors were also different from T2b tumors in terms with the frequency of confined cancer (54% versus 40%, p = 0.006) and cancer volume (1.78 versus 2.63 cm3, p = 0.013). However, the those differences were smaller than those in 1992 system. There were no significant differences between 1997 T2a and T2b cancers in the serum PSA level and the frequency of a poorly differentiated cancer. In fact, the 5-years recurrence-free survival rate for patients with T2a (73 +/- 3%) was identical to that for T2b cancer. In a Cox proportional hazard regression analysis, however, neither the 1992 nor the 1997 TNM staging subclassifications of T2 cancers were independent predictor of PSA non-progression when the age of patient, serum PSA level and biopsy Gleason grade were included in the analysis.
Conclusion: Since a palpable tumor less than half of one lobe (1992 T2a) has a distinctly different pathological and prognostic significance compared to T2b and T2c cancers, the T2a subclassification should be retained in future revisions of TNM staging system. However, because the digital rectal examination provides limited information, both PSA results and histological grade in a biopsy specimen should be incorporated into future revision of the TNM staging system.