Assessment of high-throughput high-resolution MALDI-TOF-MS of urinary peptides for the detection of muscle-invasive bladder cancer

• Published in: Proteomics. Clinical applications 2011-10, Vol.5 (9-10), p.493-503
• Main Authors: Bryan, Richard T., Wei, Wenbin, Shimwell, Neil J., Collins, Stuart I., Hussain, Syed A., Billingham, Lucinda J., Murray, Paul G., Deshmukh, Nayneeta, James, Nicholas D., Wallace, D. Michael A., Johnson, Philip J., Zeegers, Maurice P., Cheng, K. K., Martin, Ashley, Ward, Douglas G.
• Format: Article
• Language: English
• Subjects: Adult; Aged; Aged, 80 and over; Area Under Curve; Biological and medical sciences; Biomarker; Biomarkers; Biomarkers, Tumor - urine; Bladder; Bladder cancer; Cancer; Diverse techniques; Female; Fundamental and applied biological sciences. Psychology; Hematuria - metabolism; Humans; Male; Medical research; Medical sciences; Middle Aged; Molecular and cellular biology; Neoplasm Invasiveness; Neoplasm Staging; Nephrology. Urinary tract diseases; Peptides; Peptides - urine; Peptidome; ROC Curve; Spectrometry, Mass, Matrix-Assisted Laser Desorption-Ionization; Tumors of the urinary system; Urinary Bladder Neoplasms - pathology; Urinary Bladder Neoplasms - urine; Urinary tract. Prostate gland; Urine
• ISSN: 1862-8346; 1862-8354
• DOI: 10.1002/prca.201100011

Purpose: There is a need for better biomarkers to both detect bladder cancer and distinguish muscle‐invasive (stage T2+) from non‐invasive (stage Ta/T1) disease. We assess whether MALDI‐TOF‐MS of the urine peptidome can achieve this. Experimental design: We analysed urine from 751 patients with bladder cancer and 127 patients without bladder cancer. Endogenous peptide profiling was performed using a Bruker Ultraflextreme MALDI‐TOF‐MS. Results: Significant differences were seen between the spectra of urine from patients with and without T2+ disease. Albumin, total protein and haematuria were also elevated in T2+ patients. Haematuria was detected in 39% of patients with Ta/T1 disease and in 77% of patients with T2+ disease. Class prediction models based on MALDI data produced areas under receiver‐operator characteristic curves of up to 0.76 but did not significantly outperform a model based on total protein alone. Many peptides significantly associated with invasive disease are fragments of abundant blood proteins and are also associated with haematuria. Conclusions and clinical relevance: Microscopic haematuria is strongly associated with invasive disease; even traces of blood/plasma strongly influence the urinary peptidome. This needs to be taken into consideration when using ‘omic’ methods to search for urinary biomarkers as blood proteins may give false‐positive results.