Comprehensive Urinary Proteome Profiling Analysis Identifies Diagnosis and Relapse Surveillance Biomarkers for Bladder Cancer
Qi Chang,Yongqiang Chen,Jianjian Yin,Tao Wang,Yuanheng Dai,Zixin Wu,Yufeng Guo,Lingang Wang,Yufen Zhao,Hang Yuan,Dongkui Song,Lirong Zhang
DOI: https://doi.org/10.1021/acs.jproteome.4c00199
2024-01-01
Journal of Proteome Research
Abstract:Bladder cancer (BCa) is the predominant malignancy of the urinary system. Herein, a comprehensive urine proteomic feature was initially established for the noninvasive diagnosis and recurrence monitoring of bladder cancer. 279 cases (63 primary BCa, 87 nontumor controls (NT), 73 relapsed BCa (BCR), and 56 nonrelapsed BCa (BCNR)) were collected to screen urinary protein biomarkers. 4761 and 3668 proteins were qualified and quantified by DDA and sequential window acquisition of all theoretical mass spectra (SWATH-MS) analysis in two discovery sets, respectively. Upregulated proteins were validated by multiple reaction monitoring (MRM) in two independent combined sets. Using the multi-support vector machine-recursive feature elimination (mSVM-RFE) algorithm, a model comprising 13 proteins exhibited good performance between BCa and NT with an AUC of 0.821 (95% CI: 0.675-0.967), 90.9% sensitivity (95% CI: 72.7-100%), and 73.3% specificity (95% CI: 53.3-93.3%) in the diagnosis test set. Meanwhile, an 11-marker classifier significantly distinguished BCR from BCNR with 75.0% sensitivity (95% CI: 50.0-100%), 81.8% specificity (95% CI: 54.5-100%), and an AUC of 0.784 (95% CI: 0.609-0.959) in the test cohort for relapse surveillance. Notably, six proteins (SPR, AK1, CD2AP, ADGRF1, GMPS, and C8A) of 24 markers were newly reported. This paper reveals novel urinary protein biomarkers for BCa and offers new theoretical insights into the pathogenesis of bladder cancer (data identifier PXD044896).