The significance of the free-to-complexed prostate-specific antigen (PSA) ratio in prostate cancer detection in patients with a PSA level of 4.1-10.0ng/mL
T. Okegawa et al., The significance of the free-to-complexed prostate-specific antigen (PSA) ratio in prostate cancer detection in patients with a PSA level of 4.1-10.0ng/mL, BJU INT, 85(6), 2000, pp. 708-714
Objective To compare the ratio of free prostate: specific antigen (fPSA), t
otal PSA (tPSA) and complexed PSA (cPSA, measured using a novel immunoassay
) with other variables used to detect prostate cancer in patients with inte
rmediate serum PSA levels of 4.1-10.0 ng/mL.
Patients and methods From July 1997 to August 1998, 140 consecutive patient
s were assessed; all had intermediate serum PSA levels and/or abnormal find
ings on a digital rectal examination. All patients underwent transrectal ul
trasonography (TRUS)-guided biopsy, and the prostate and transition zone vo
lumes were determined by TRUS. Free and tPSA were measured using the Tandem
-R assay (Hybritech Corp., San Diego, CA). PSA complexed with al-antichymot
rypsin (cPSA) was measured using an appropriate assay. The ability of cPSA,
free-to-total PSA ratio (f/tPSA), free-to-complexed PSA ratio (f/cPSA), tP
SA density of the whole prostate (PSAD), of the transition zone (tPSATZ), a
nd cPSA density of the whole prostate (cPSAD) and of the transition zone (c
PSATZ) to improve the power of PSA in detecting prostate cancer was evaluat
ed using receiver operating characteristic (ROC) curves.
Results Of the 140 patients, 126 had histologically confirmed benign diseas
e and 14 had prostate cancer, The cPSA alone had better specificity for det
ecting prostate cancer than had tPSA alone hut the difference was not signi
ficant. The area under the ROC curve for f/cPSA was larger than those for a
ll other variables. With a 93% sensitivity for detecting prostate cancer, a
f/cPSA threshold of 25% would result in fewer unnecessary biopsies (40% f/
cPSA specificity) than with all other PSA variables. The difference in the
resolution was significant between f/cPSA and tPSA, cPSA, tPSAD and tPSATZ,
but not with f/tPSA, cPSAD or cPSATZ. In patients with a prostate volume o
f <30 mL, the cPSATZ showed better specificity for prostate cancer than tPS
A alone.
Conclusion Measuring the level of cPSA and its derivatives may provide bett
er differentiation of prostate cancer and benign disease than tPSA alone in
patients with a tPSA level of 4.1-10.0 ng/mL.