1868 - Evaluation and diagnostic potential of plasma biomarkers in bladder cancer

Background

While urine biomarkers are widely used to diagnose bladder cancer (BLC), little is known about plasma protein levels in patients with BLC. The current research is aimed to evaluate diagnostic potential of 13 plasma markers including tumor antigens, inflammatory markers and apolipoproteins (Apo) as well as combinations of thereof.

Methods

In total 203 healthy volunteers (HV) and 59 patients with BLC were enrolled into the study. Concentrations of alpha-fetoprotein (AFP), carcinoembryonic antigen (CEA), carbohydrate antigen 19-9 (СА 19-9), prostate-specific antigen (PSA), beta 2 microglobulin (B2M), human-specific C-reactive protein (hsCRP), D-dimer, сytokeratin 19-fragments (CYFRA 21-1), ApoA1, ApoA2, ApoВ, transthyretin (TTR), and soluble vascular cell adhesion molecule-1 (sVCAM-1) in plasma were measured via ELISA. t-test after log-transformation was used to identify between-group differences in biomarker levels. Diagnostic accuracy of the single biomarkers as well as trained random forest (RF), linear discriminant analysis (LDA) and support vector machine (SVM) classifiers was assessed by ROC analysis.

Results

Plasma levels of ApoB, B2M, CA 19-9, CYFRA 21-1, D-dimer, hsCRP, sVCAM-1 and TTR were significantly higher (p-value<0.001) whereas ApoA1 and ApoA2 levels were significantly lower (p-value<0.0005) in patients with BLC vs HV. No differences in AFP, CEA and PSA was found between the groups. The highest discriminative power was shown for sVCAM-1 and ApoA1 with area under ROC curve (AUROC) 0.92 and 0.90, respectively, whereas AUROC for several classifiers based on measurements of 2-12 biomarkers was higher than 0.95.

Conclusions

Numerous abnormalities in plasma biomarker levels were detected in patients with BLC, hence, blood-based tests represent a promising strategy to improve performance of urinary-based tests and cystoscopy in BLC detection and prognosis. Combining several biomarkers allows to increase diagnostic test accuracy.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

I.M. Sechenov First Moscow State Medical University.

Funding

I.M. Sechenov First Moscow State Medical University.

Disclosure

All authors have declared no conflicts of interest.