BACKGROUND: Non-invasive methods for diagnosis of urothelial carcinoma have reduced specificity in patients with non-malignant genitourinary disease or other disorders. We aimed to use mass spectrometry and bioinformatics to define and validate a cancer-specific proteomic pattern. METHODS: We used capillary-electrophoresis-coupled mass spectrometry to obtain polypeptide patterns from urine samples of 46 patients with urothelial carcinoma and 33 healthy volunteers. From signatures of polypeptide mass, we established a model for predicting the presence of cancer. The model was refined further by use of 366 urine samples obtained from other healthy volunteers and patients with malignant and non-malignant genitourinary disease. We estimated the proportion of correct classifications from the refined model by applying it to a masked group containing 31 patients with urothelial carcinoma, 11 healthy individuals, and 138 patients with non-malignant genitourinary disease. We also sequenced several diagnostic polypeptides for urothelial carcinoma. FINDINGS: We identified a diagnostic urothelial-carcinoma pattern of 22 polypeptide masses. On masked assessment, prediction models based on these polypeptides correctly classified all samples of urothelial carcinoma (sensitivity 100% [95% CI 87-100) and all healthy samples (specificity 100% [84-100]). Correct identification of patients with urothelial carcinoma from those with other malignant and non-malignant genitourinary disease ranged from 86% to 100%. A prominent polypeptide from the diagnostic pattern for urothelial carcinoma was identified as fibrinopeptide A-a known biomarker of ovarian cancer and gastric cancer. INTERPRETATION: Validation of a highly specific biomarker pattern for urothelial carcinoma in a large group of patients with various urological disorders could be used in the diagnosis of other diseases that are identified in urine samples or in other body fluids.
BACKGROUND: Non-invasive methods for diagnosis of urothelial carcinoma have reduced specificity in patients with non-malignant genitourinary disease or other disorders. We aimed to use mass spectrometry and bioinformatics to define and validate a cancer-specific proteomic pattern. METHODS: We used capillary-electrophoresis-coupled mass spectrometry to obtain polypeptide patterns from urine samples of 46 patients with urothelial carcinoma and 33 healthy volunteers. From signatures of polypeptide mass, we established a model for predicting the presence of cancer. The model was refined further by use of 366 urine samples obtained from other healthy volunteers and patients with malignant and non-malignant genitourinary disease. We estimated the proportion of correct classifications from the refined model by applying it to a masked group containing 31 patients with urothelial carcinoma, 11 healthy individuals, and 138 patients with non-malignant genitourinary disease. We also sequenced several diagnostic polypeptides for urothelial carcinoma. FINDINGS: We identified a diagnostic urothelial-carcinoma pattern of 22 polypeptide masses. On masked assessment, prediction models based on these polypeptides correctly classified all samples of urothelial carcinoma (sensitivity 100% [95% CI 87-100) and all healthy samples (specificity 100% [84-100]). Correct identification of patients with urothelial carcinoma from those with other malignant and non-malignant genitourinary disease ranged from 86% to 100%. A prominent polypeptide from the diagnostic pattern for urothelial carcinoma was identified as fibrinopeptide A-a known biomarker of ovarian cancer and gastric cancer. INTERPRETATION: Validation of a highly specific biomarker pattern for urothelial carcinoma in a large group of patients with various urological disorders could be used in the diagnosis of other diseases that are identified in urine samples or in other body fluids.
Authors: Sylvia Borchers; Elena Provasi; Anna Silvani; Marina Radrizzani; Claudia Benati; Elke Dammann; Annika Krons; Julia Kontsendorn; Joerg Schmidtke; Wolfgang Kuehnau; Nils von Neuhoff; Michael Stadler; Fabio Ciceri; Chiara Bonini; Arnold Ganser; Bernd Hertenstein; Eva M Weissinger Journal: Hum Gene Ther Date: 2011-03-30 Impact factor: 5.695
Authors: Marion Haubitz; David M Good; Alexander Woywodt; Hermann Haller; Harald Rupprecht; Dan Theodorescu; Mohammed Dakna; Joshua J Coon; Harald Mischak Journal: Mol Cell Proteomics Date: 2009-06-28 Impact factor: 5.911