Development of a Predictive Model of Prostate cancer: Integration of a Panel of Formerly N-linked Glycopeptides and Clinical Variables for Serum Testing

Abstract

Abstract Background: Prostate Cancer (PCa) represents the second leading cause of cancer-related death in men. Prostate-specific antigen (PSA) is currently used for PCa screening but because of its low specificity and sensitivity new diagnostic tools are required. Methods: In this work, 32 formerly N-glycosylated peptides were quantified by PRM in 163 serum samples (79 from PCa patients and 84 from individuals affected by benign prostatic hyperplasia (BPH)) in two technical replicates. These potential biomarker candidates were prioritized through a multi-stage biomarker discovery pipeline articulated in: discovery, LC-PRM assay development and verification phases. Because of the well-established involvement of glycoproteins in cancer development and progression, the proteomic analysis was focused on glycoproteins enriched by TiO2 strategy. Results: Machine learning algorithms have been applied to the combined matrix comprising proteomic and clinical variables, resulting in a predictive model based on six proteomic variables (LAMB1, LAMP2, LUM, TFRC, NCAM1, GPLD1) and five clinical variables (prostate dimension, proPSA, free-PSA, total-PSA, free/total-PSA). Conclusions: A predictive model combining proteomic and clinical variables able to distinguish PCa from BPH with an AUC of 0.82 was developed. This model outperformed PSA alone which, on the same sample set, was able to discriminate PCa from BPH with an AUC of 0.74. Data are available via ProteomeXchange with identifier PXD035935.