BACKGROUND: A novel urine test for early detection of prostate cancer (PCA), distributed and marketed by the company DiaPat, is advertised by the statement "correct analysis in 9 of 10 cases." PATIENTS AND METHODS: The test separates urinary polypeptides by means of capillary electrophoresis and characterizes the peptides in a time-of-flight mass spectrometer. The DiaPat test was performed on the urine of 18 men prior to multiple ultrasound-guided prostate biopsies. RESULTS: Sixteen of the 18 samples met the requirements for sample quality as established by the manufacturer. Eight of these 16 urine samples had been collected from patients in whom biopsies consecutively detected PCA; the remaining eight patients had benign biopsy results. Among the eight patients with detected PCA, the urine test yielded a low probability for PCA in three cases and a high probability in five. Within the group of eight patients with benign biopsy results, the urine test predicted a high probability for PCA in five men and a low probability in three. For the given PCA incidence of 50% within the investigated population, the DiaPat test correctly predicted biopsy results in one half of the population, whereas prediction in the remaining half was incorrect. CONCLUSION: Unless reliable validation of the DiaPat urine test for PCA is available, no clinical consequences should be drawn from the test results.
BACKGROUND: A novel urine test for early detection of prostate cancer (PCA), distributed and marketed by the company DiaPat, is advertised by the statement "correct analysis in 9 of 10 cases." PATIENTS AND METHODS: The test separates urinary polypeptides by means of capillary electrophoresis and characterizes the peptides in a time-of-flight mass spectrometer. The DiaPat test was performed on the urine of 18 men prior to multiple ultrasound-guided prostate biopsies. RESULTS: Sixteen of the 18 samples met the requirements for sample quality as established by the manufacturer. Eight of these 16 urine samples had been collected from patients in whom biopsies consecutively detected PCA; the remaining eight patients had benign biopsy results. Among the eight patients with detected PCA, the urine test yielded a low probability for PCA in three cases and a high probability in five. Within the group of eight patients with benign biopsy results, the urine test predicted a high probability for PCA in five men and a low probability in three. For the given PCA incidence of 50% within the investigated population, the DiaPat test correctly predicted biopsy results in one half of the population, whereas prediction in the remaining half was incorrect. CONCLUSION: Unless reliable validation of the DiaPat urine test for PCA is available, no clinical consequences should be drawn from the test results.
Authors: Dan Theodorescu; Danilo Fliser; Stefan Wittke; Harald Mischak; Ronald Krebs; Michael Walden; Mark Ross; Elke Eltze; Olaf Bettendorf; Christian Wulfing; Axel Semjonow Journal: Electrophoresis Date: 2005-07 Impact factor: 3.535
Authors: Dan Theodorescu; Eric Schiffer; Hartwig W Bauer; Friedrich Douwes; Frank Eichhorn; Reinhard Polley; Thomas Schmidt; Wolfgang Schöfer; Petra Zürbig; David M Good; Joshua J Coon; Harald Mischak Journal: Proteomics Clin Appl Date: 2008-03-07 Impact factor: 3.494