Novel urinary biomarkers for the detection of bladder cancer: A systematic review

Abstract

Background: Urinary biomarkers for the diagnosis of bladder cancer represents an area of considerable research which has been tested in both patients presenting with haematuria and non-muscle invasive bladder cancer patients requiring surveillance cystoscopy. In this systematic review, we identify and appraise the diagnostic sensitive and specificity of reported novel biomarkers of different 'omic' class and highlight promising biomarkers investigated to date.

Methods: A MEDLINE/Pubmed systematic search was performed between January 2013 and July 2017 using the following keywords: (bladder cancer OR transitional cell carcinoma OR urothelial cell carcinoma) AND (detection OR diagnosis) AND urine AND (biomarker OR assay). All studies had a minimum of 20 patients in both bladder cancer and control arms and reported sensitivity and/or specificity and/or receiver operating characteristics (ROC) curve. QUADAS-2 tool was used to assess risk of bias and applicability of studies. The search protocol was registered in the PROSPERO database (CRD42016049918).

Results: Systematic search yielded 115 reports were included for analysis. In single target biomarkers had a sensitivity of 2-94%, specificity of 46-100%, positive predictive value (PPV) of 47-100% and negative predictive value (NPV) of 21-94%. Multi-target biomarkers achieved a sensitivity of 24-100%, specificity of 48-100%, PPV of 42-95% and NPV of 32-100%. 50 studies achieved a sensitivity and specificity of ≥80%. Protein (n = 59) and transcriptomic (n = 21) biomarkers represents the most studied biomarkers. Multi-target biomarker panels had a better diagnostic accuracy compared to single biomarker targets. Urinary cytology with urinary biomarkers improved the diagnostic ability of the biomarker. The sensitivity and specificity of biomarkers were higher for primary diagnosis compared to patients in the surveillance setting. Most studies were case control studies and did not have a predefined threshold to determine a positive test result indicating a possible risk of bias.

Conclusion: This comprehensive systematic review provides an update on urinary biomarkers of different 'omic' class and highlights promising biomarkers. Few biomarkers achieve a high sensitivity and negative predictive value. Such biomarkers will require external validation in a prospective observational setting before adoption in clinical practice.

Keywords: Biomarker; Bladder cancer; Diagnosis; Systematic review; Urine.