Detection of prostate cancer and predicting progression: current and future diagnostic markers.

Publication Type
Journal Article
Year of Publication
Tricoli, James V; Schoenfeldt, Mason; Conley, Barbara A
Clin Cancer Res
Date Published
2004 Jun 15
Antigens, Neoplasm; Antigens, Surface; Biomarkers, Tumor; Chromogranin A; Chromogranins; Disease Progression; DNA-Binding Proteins; Glutamate Carboxypeptidase II; Glutathione Transferase; GPI-Linked Proteins; Humans; Male; Membrane Glycoproteins; Neoplasm Proteins; prostatic neoplasms; Telomerase

Carcinoma of the prostate is the second leading cause of male cancer-related death in the United States. Better indicators of prostate cancer presence and progression are needed to avoid unnecessary treatment, predict disease course, and develop more effective therapy. Numerous molecular markers have been described in human serum, urine, seminal fluid, and histological specimens that exhibit varying capacities to detect prostate cancer and predict disease course. However, to date, few of these markers have been adequately validated for clinical use. The purpose of this review is to examine the current status of these markers in prostate cancer and to assess the diagnostic potential for future markers from identified genes and molecules that display loss, mutation, or alteration in expression between tumor and normal prostate tissues. In this review we cite 91 molecular markers that display some level of correlation with prostate cancer presence, disease progression, cancer recurrence, prediction of response to therapy, and/or disease-free survival. We suggest criteria to consider when selecting a marker for further development as a clinical tool and discuss five examples of markers (chromogranin A, glutathione S-transferase pi 1, prostate stem cell antigen, prostate-specific membrane antigen, and telomerase reverse transcriptase) that fulfill some of these criteria. Finally, we discuss how to conduct evaluations of candidate prostate cancer markers and some of the issues involved in the validation process.