Identification of potential serum protein biomarkers for recurrence in gastric cancer patients using a quantitative multiple reaction monitoring approach

Abstract

Gastric cancer (GC) is one of the most common cancers representing the second leading cause of cancer-related mortality. Despite improvements in clinical therapies of GC, the recurrence rate of GC patients remains high (~55%) with advanced stage of the disease. Therefore, it is essential to understand of GC recurrence mechanisms that would help effective clinical application for GC diagnosis and prognosis. Here, we aimed to identify potential serum biomarkers for recurrence in gastric cancers with an established quantitative multiple reaction monitoring (MRM) approach using GC patient serum samples. To build up a serum biomarker development platform, we first generated serum biomarker candidates through comprehensive proteomic approach. By employing both preliminary MRM and automated detection of inaccurate and imprecise transitions (AuDIT) analysis with stable isotope–labeled internal standard (SIS) peptides using pooled GC patient serum samples, we established a quantitative MRM analysis of 94 proteins as final MRM target proteins. To establish the multi-biomarker panel for identification of GC recurrence, we conducted the quantitative MRM analysis with 180 individual patients divided into the two groups, i.e. response group (n=133) and recurrence group (n=47), who received chemotherapy after D2 lymph node dissection in both groups, as a training set. By a stringent statistical analysis with quantitative MRM data of training sets individual samples, the 6-marker panel, consisting of alpha-1-antichymotrypsin (SERPINA3), apolipoprotein A-II (APOA2), apolipoprotein C-I (APOC1), clusterin (CLU), inter-alpha-trypsin inhibitor heavy chain H4 (ITIH4), and leucine-rich alpha-2-glycoprotein (LRG1), was constructed. These proteins showed the differentially expressed levels (p-value < 0.05) between the two groups with an area under the curve (AUC) value of 0.810 and high prediction rates in both groups (95.5% and 61.7% in response and recurrence groups, respectively). To verify the 6-marker panel, we further applied MRM analysis with independent patient samples (n=64), i.e. response group (n=43) and recurrence group (n=21), as a test set. We demonstrated that 6 marker proteins showed the correlated expression patterns as in a training set with statistical significance (p-value < 0.05). We propose that these proteins can serve as diagnostic signatures to identify the recurrence in GC patients and our quantitative MRM assay based serum biomarker development platform could serve as a valuable tool in the clinical biomarker discovery-verification process.