Freedland Stephen J, Mangold Leslie A, Walsh Patrick C, Partin Alan W
The James Buchanan Brady Urological Institute, The Johns Hopkins School of Medicine, Baltimore, Maryland 21287-2101, USA.
J Urol. 2005 Oct;174(4 Pt 1):1276-81; discussion 1281; author reply 1281. doi: 10.1097/01.ju.0000173907.84852.ec.
Prostate specific antigen (PSA) has been shown to predict the presence of prostate cancer on biopsy, pathological stage, and biochemical progression following primary therapy. A recent study found only a weak association between PSA and tumor volume in the radical prostatectomy (RP) specimen and concluded that the PSA era is over. We examined the association between PSA and clinical progression in men undergoing RP.
The study population consisted of 2,312 men treated with RP between 1992 and 2004 by a single surgeon. We evaluated the association between preoperative PSA and biochemical progression on multivariate analysis.
Men with higher preoperative PSA concentrations had higher grade cancers in the biopsy and RP specimen, and more adverse pathological features. After adjusting for the clinical covariates of age, race, grade, stage, and year of surgery, preoperative PSA was significantly associated with the risk of biochemical progression. When only men with PSA less than 10 ng/ml were examined, PSA remained a significant predictor of biochemical progression on multivariate analysis (RR 1.30, 95% CI 1.18 to 1.44, p <0.001). For each 2-point increase in PSA, the risk of biochemical progression increased approximately 2-fold.
Preoperative PSA was significantly associated with high grade disease and adverse pathological findings. After adjusting for clinical covariates, PSA was significantly associated with the risk of biochemical progression, even in men with PSA less than 10 ng/ml. Despite multiple limitations, PSA remains the best prostate cancer tumor marker available.
