Non-muscle-invasive bladder cancer (
NMIBC) is one of the most common malignant
tumors in the urological system with a high risk of recurrence, and effective non-invasive
biomarkers for
NMIBC relapse are still needed. The human urinary
proteome can reflect the status of the microenvironment of the urinary system and is an ideal source for clinical diagnosis of urinary system diseases. Our previous work used proteomics to identify 1643 high-confidence urinary
proteins in the urine from a healthy population. Here, we used bioinformatics to construct a
cancer-associated
protein-
protein interaction (PPI) network comprising 16 high-abundance urinary
proteins based on the urinary
proteome database. As a result,
platelet-derived growth factor receptor beta (
PDGFRB) was selected for further validation as a candidate
biomarker for
NMIBC diagnosis and prognosis. Although the levels of urinary
PDGFRB showed no significant difference between patients pre- and post-surgery (n = 185, P>0.05), over 3 years of follow-up, urinary
PDGFRB was shown to be significantly higher in relapsed patients (n = 68) than in relapse-free patients (n = 117, P<0.001). The levels of urinary
PDGFRB were significantly correlated with the risk of 3-year recurrence of
NMIBC, and these levels improved the accuracy of a
NMIBC recurrence risk prediction model that included age,
tumor size, and
tumor number (area under the curve, 0.862; 95% CI, 0.809 to 0.914) compared to PDGFR alone. Therefore, we surmise that urinary
PDGFRB could serve as a non-invasive
biomarker for predicting
NMIBC recurrence.