Purpose: The goal of this study was to evaluate the effect of local/regiona
l treatment, particularly external beam radiation alone vs. radical prostat
ectomy plus radiation therapy in patients with pathologic node positive pro
state cancer on survival. Methods: Medical records of all 116 patients who
received their initial treatment at the Massachusetts General Hospital betw
een 1980 and 1996 for adenocarcinoma of the prostate with pathologic confir
med nodal metastases and no distant metastases were reviewed. The mean foll
ow up was 5.5 years. Overall survival, time to PSA failure on endocrine the
rapy, and time to first intervention were evaluated. The effect of the diff
erent treatment options were compared using multivariate Cox proportional h
azard models to adjust for tumor characteristics that might influence survi
val. These included clinical T stage, clinical N stage, Gleason grade, numb
er and location of positive lymph nodes and pretreatment PSA. Results: The
combined patient population had a 5-year survival rate of 74% and a 10-year
survival rate of 48%. Patients receiving local/regional treatment had adju
sted 5 year survival rates of 80% compared to 27% for patients receiving no
local/regional treatment (p = .001) with corresponding cumulative interven
tion rates (CIR) of 11% vs. 73% (p = .01) Patients receiving external beam
radiation (XRT) alone did not differ significantly from those receiving pro
statectomy plus radiation therapy in terms of survival (75 vs. 82%, p = .23
) or cumulative intervention rates (14% vs. 14%, p = .94) Conclusion: Altho
ugh it appears that all patients with node positive prostate cancer will ev
entually develop failure, this paper suggests local/regional therapy offers
a medium term survival advantage over no local/regional treatment. The add
ition of prostatectomy did not confer a demonstrable advantage over radiati
on alone. (C) 2001 Elsevier Science Inc. All rights reserved.