Abstract
BACKGROUND: Currently, serum biomarkers, which are sufficiently sensitive and specific for early detection and risk classification of gastric adenocarcinoma do not exist. Therefore, this study identified a panel of serum biomarkers for the diagnosis of gastric adenocarcinoma. METHODS: A 29-plex array platform with 29 biomarkers, consisting of 11 proteins discovered through proteomics and 18 previously known to be cancer-associated, was constructed. A test/training set consisting of 120 gastric adenocarcinoma and 120 control samples were examined. After 13 proteins were selected as candidate biomarkers, multivariate classification analyses were used to identify algorithms for diagnostic biomarker combinations. These algorithms were independently validated using a set of 95 gastric adenocarcinoma and 51 control samples. RESULTS: Epidermal growth factor receptor (EGFR), pro-apolipoprotein A1 (proApoA1), apolipoprotein A1, transthyretin (TTR), regulated upon activation, normally T-expressed and presumably secreted (RANTES), D-dimer, vitronectin (VN), interleukin-6, α-2 macroglobulin, C-reactive protein and plasminogen activator inhibitor-1 were selected as classifiers in the two algorithms. These algorithms differentiated between the majority of gastric adenocarcinoma and control serum samples in the training/test set with high accuracy (>88%). These algorithms also accurately classified in the validation set (>85%). CONCLUSION: Two panels of combinatorial biomarkers, including EGFR, TTR, RANTES, and VN, are developed, which are less invasive method for the diagnosis of gastric adenocarcinoma. They could supplement clinical gastroscopic evaluation of symptomatic patients to enhance diagnostic accuracy.