It is of highest importance to find
proteins responsible for
breast cancer dissemination, for use as
biomarkers or treatment targets. We established and performed a combined nontargeted LC-MS/MS and a targeted LC-SRM workflow for discovery and validation of
protein biomarkers. Eighty
breast tumors, stratified for
estrogen receptor status and development of distant recurrence (DR ± ), were collected. After enrichment of N-glycosylated
peptides, label-free LC-MS/MS was performed on each individual
tumor in triplicate. In total, 1515
glycopeptides from 778
proteins were identified and used to create a map of the
breast cancer N-glycosylated
proteome. Based on this specific
proteome map, we constructed a 92-plex targeted label-free LC-SRM panel. These
proteins were quantified across samples by LC-SRM, resulting in 10
proteins consistently differentially regulated between DR+/DR-
tumors. Five
proteins were further validated in a separate cohort as prognostic
biomarkers at the gene expression level. We also compared the LC-SRM results to clinically reported HER2 status, demonstrating its clinical accuracy. In conclusion, we demonstrate a combined mass spectrometry strategy, at large scale on clinical samples, leading to the identification and validation of five
proteins as potential
biomarkers for
breast cancer recurrence. All MS data are available via ProteomeXchange and PASSEL with identifiers PXD001685 and PASS00643.