Received on December 2, 2005
Accepted on February 22, 2006

Cancer Diagnostics

Enhanced Discrimination of Benign from Malignant Prostatic Disease by Selective Measurements of Cleaved Forms of Urokinase Receptor in Serum

¹ The Finsen Laboratory, Rigshospitalet, Copenhagen, Denmark, and Current affiliation: Schering Oy, Turku, Finland
² Department of Urology, University Clinic Hamburg Eppendorf, Hamburg, Germany
³ Department of Surgical Gastroenterology, Hvidovre University Hospital, Hvidovre Denmark
⁴ Institute of Pathobiology, Royal Veterinary and Agricultural University, Copenhagen, Denmark
⁵ The Finsen Laboratory, Rigshospitalet, Copenhagen, Denmark
⁶ Department of Laboratory Medicine, Division of Clinical Chemistry, Lund University, University Hospital (UMAS), Malmö, Sweden, and Departments of Clinical Laboratories, Urology, and Medicine, Memorial Sloan-Kettering Cancer Center, New York, NY

^* To whom correspondence should be addressed. E-mail: LiljaH{at}mskcc.org.

Background: Early detection of prostate cancer (PCa) centers on measurements of prostate-specific antigen (PSA), but current testing practices suffer from lack of specificity and generate many unnecessary prostate biopsies. Soluble urokinase plasminogen activator receptor (uPAR) is present in blood in both intact and cleaved forms. Increased uPAR in blood is correlated with poor prognosis in various cancers, but uPAR has not been shown to be useful in PCa diagnostics. We assessed the ability of immunoassays for specific uPAR forms to discriminate PCa from benign conditions.

Methods: We measured total PSA (tPSA), free PSA (fPSA), intact uPAR [uPAR(I-III)], intact uPAR + cleaved uPAR domains II+III [uPAR(I-III) + uPAR(II-III)], and cleaved uPAR domain I [uPAR(I)] in sera from 224 men with and 166 men without PCa. We assessed differences in serum concentrations between the PCa and noncancer groups within the entire cohort and in men with tPSA concentrations of 2-10 µg/L. The diagnostic accuracy of individual analytes and analyte combinations was explored by logistic regression and ROC analyses and evaluations of sensitivity and specificity pairs.

Results: Serum uPAR(I) and uPAR(II-III) were higher in PCa than in benign disease. In men with tPSA between 2 and 10 µg/L, the combination of %fPSA with the ratio uPAR(I)/uPAR(I-III) had a greater area under the ROC curve (0.73) than did %fPSA (0.68).

Conclusions: Specific measurements of different uPAR forms in serum improve the specificity of PCa detection. The uPAR forms may therefore be complementary to PSA for PCa detection, most importantly in men with moderately increased PSA.