Abstract 347P
Background
Widespread reliance on PSA screening has resulted in a significant number of unnecessary invasive biopsies in patients without prostate cancer. Consequently, identifying highly sensitive and specific non-invasive molecular biomarkers is paramount for early prostate cancer detection. Mass spectrometry-based urinary proteomics has emerged as a powerful tool for cancer biomarker analysis and holds immense potential for developing early prostate cancer screening biomarkers.
Methods
A cohort of 40 patients at our center who underwent prostate biopsy due to elevated PSA levels (≥ 4 ng/ml) were enrolled in the study. Preoperative peripheral blood and morning midstream urine (1 ml ) were collected from these patients. Twenty-two (55%) patients were diagnosed with prostate cancer after biopsy. Protein enrichment was performed using the n-LAPE/MSTM platform on 0.5 ml aliquots of both blood and urine samples. Subsequently, data-independent acquisition (DIA) mass spectrometry was employed for the analysis of protein-derived peptide fragments. Following data quality control, comprehensive proteomic analysis and differential protein analysis were performed. Based on differentially expressed proteins with an appearance frequency ≥ 30% and a fold change ≥ 1.5, the five most significantly upregulated and downregulated proteins (appearance frequency ≥ 70% and the highest AUC value) were selected for model construction.
Results
Preliminary results demonstrated that the urine proteomics-based prostate cancer early screening model achieved an AUC of 0.9074 in distinguishing biopsy tumor from biopsy non-tumor individuals. When paired blood samples were included in the comparison, the AUC was 0.8857. Table: 347P
| ID | Gene | Protein descriptions | FC | P | AUC | |
| top5 Up-regulated | Q9Y6E2 | BZW2 | eIF5-mimic protein 1 | 2.30 | 7.33E-04 | 0.833 |
| Q07075 | ENPEP | Glutamyl aminopeptidase | 2.11 | 1.16E-03 | 0.820 | |
| Q9HBL8 | NMRAL1 | NmrA-like family domain-containing protein 1 | 2.44 | 2.88E-03 | 0.794 | |
| O14908 | GIPC1 | PDZ domain-containing protein GIPC1 | 1.69 | 3.94E-03 | 0.784 | |
| Q06828 | FMOD | Fibromodulin | 2.32 | 4.58E-03 | 0.780 | |
| top5 Down-regulated | P08311 | CTSG | Cathepsin G | 0.20 | 6.01E-04 | 0.825 |
| P41218 | MNDA | Myeloid cell nuclear differentiation antigen | 0.18 | 9.01E-04 | 0.816 | |
| P17213 | BPI | Bactericidal permeability-increasing protein | 0.07 | 1.31E-03 | 0.813 | |
| P11215 | ITGAM | Integrin alpha-M | 0.26 | 2.53E-03 | 0.797 | |
| P16401 | H1-5 | Histone H1.5 | 0.25 | 2.46E-03 | 0.791 | |
| Multi-ROC | 0.907 |
Conclusions
The urine proteomics-based model accurately differentiates prostate cancer patients from non-cancer individuals, offering a promising tool for early detection.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
Has not received any funding.
Disclosure
All authors have declared no conflicts of interest.