Rv. Iyer et al., Outcome evaluation of the 1997 American Joint Committee on Cancer staging system for prostate carcinoma treated by radiation therapy, CANCER, 85(8), 1999, pp. 1816-1821
BACKGROUND. The 1997 American Joint Committee on Cancer (AJCC) staging syst
em condensed unilobular tumors into one entity and continues the use of bot
h imaging and biopsy to alter classification status in T2 and T3 carcinomas
. This study analyzes the biochemical freedom from disease recurrence (bNED
) outcome in a large database to determine whether these changes reflect ou
tcome differences.
METHODS. Five hundred and thirty-seven patients with adenocarcinoma of the
prostate were treated with radiation therapy to a median dose of 7180 centi
grays (cGy) (range, 6316-8074 cGy) between November 1987 and November 1994.
The median age of the patients was 70 years and the median follow-up was 5
1 months. The median pretreatment prostate specific antigen (PSA) was 11.0
ng/mL. Patients were analyzed using 1992 AJCC stage comparing bNED outcome
after radiation therapy for T2a versus T2b versus T2c tumors using Kaplan-M
eier estimation and the log rank test. Patients then were analyzed multivar
iately using Cox regression with the known prognostic variables of dose, pr
etreatment PSA, palpation stage, and grade in addition to palpation plus im
aging stage and palpation plus biopsy stage. The prognostic endpoint was bN
ED with failure as defined by the 1997 American Society for Therapeutic Rad
iology and Oncology Consensus Panel.
RESULTS. The 1992 ATCC palpation classifications T2a versus T2b versus T2c
have a significantly different (P = 0.02) bNED outcome. Prognostic signific
ance is lost by pooling these three classifications in the 1997 ATCC stagin
g system. Adding imaging information to palpation did not improve the abili
ty of palpation alone to assess bNED status (P = 0.33). However, the additi
on of biopsy information to palpation significantly (P = 0.02) increased th
e accuracy of palpation stage alone to predict for bNED outcome far T2 and
T3 tumors.
CONCLUSIONS. The subdivision of T2 tumors in the 1992 AJCC classification (
T2a, T2b, and T2c) should be used in the next revision of the 1997 AJCC sta
ging system. The addition of imaging data does not discriminate bNED Outcom
e any better than palpation stage alone in T2 and T3 tumors and should not
be used. The addition of biopsy information to palpation stage did signific
antly improve the predicted outcome compared with palpation alone and shoul
d continue to be used. (C) 1999 American Cancer Society.