Correlation Between The Sera N-glycome Characteristics And The TCM Syndromes In Gastric Cancer

OBJECTIVE: To analyze the profile of sera N-glycome from gastric cancer patients,atrophic gastrisis patients and healthy control and set up the diagnosic model ofN-glycome for gastric cancer. To make initial clinical application with N-glycomeprofile and the diagnostic model of N-glycome, and explore the mechanism of thecharacteristic change of N-glycome in gastric cancer patients. To analyze thecorrelation between the TCM syndromes and the tumor stage in gastric cancer.METHODS: Collect the sera samples from128gastric patients,108atrophicgastrisis patients and120healthy control, and profile of N-glycome was analyse byDNA sequencer-assisted fluorophore-assisted carbohydrateelectrophoresis(DSA-FACE). Set up the diagnosic model of N-glycome to identifygastric cancer from healthy control and from atrophic gastrisis respectively withlogistic step-wise regression analysis. Validate the diagnosic model of N-glycomewith reciever operating characteristic curve analysis. Verify their diagnostic efficiacywith clinical data of validation group. Compare and analyse the N-glycom profilefrom the patients with different tumor stages, and the profile collected before and afterthe radical gastrectomy for cancer. Analyze the correlation within N-glycome profileand TNM staging. Measure and compare the sera level of total core fucose residues,by lectin blottin, of the gastric cancer patients, atrophic gastrisis patients and healthycontrol. Measure and compare the levels of total core fucose residues and mRNAexpress of fucosyltransferase (Fut8) and GDP-fuc-Tr of the tumor tissue andsurrounding tissue. Analyze the correlation between the TCM syndrome and tumorstaging, and the variations of the commonly used tumor marker and N-glycomeprofile among different TCM syndromes.RESULTS: The sera N-glycome profile showed that levels of Peak1,3,5,6,7,9andsumfuc among Group Healthy Control, Atrophic Gastitis and Gastric Cancer arestastistically different. Set up diagnosic model of N-glycome, GCglyco A（GCglycoA=-16.357+0.105Peak5+0.570Peak6+0.319Peak8-0.532Peak9）, to identify gastriccancer from healthy control. Set up diagnosic model of N-glycome, GCglyco B（GCglyco B=-1.139+0.368Peak3-1.151Peak4+0.333Peak6-0.306Peak9）, to identify gastric cancer from atrophic gastrisis. The area under the ROC curve ofGCglyco A is bigger when comparing with those of CEA, CA19-9, CA125, andCA724(0.903vs0.767,0.779,0.772and0.714). The area under the ROC curve ofGCglyco B is bigger when comparing with those of CEA, CA19-9, CA125, andCA724(0.847vs0.618,0.610,0.640and0.606). The value of GCglyco A andGCglyco B among Group Healthy Control, Atrophic Gastitis and Gastric Cancer arestastistically different too. The Sensitivity and Accuracy of GCglyco A are higher thanCEA,78.13%,84.68%vs46.09%,70.16%, while the Specificity of GCglyco A arelower than CEA,91.67%vs95.83%. The Sensitivity and Accuracy of GCglyco B arehigher than CEA,67.19%,76.27%vs46.09%,58.90%, while the Specificity ofGCglyco B are lower than CEA,74.07%vs87.03%. Levels of GCglyco A andGCglyco B among Group Healthy Control, Atrophic Gastitis and Gastric Cancer arestastistically different too. Levels of sera Peak6,9, sumfuc, GCglyco A and GCglycoB are stastically different among different tumor stage groups. Levels of sera Peak6,GCglyco A and GCglyco B are negatively correlated with TNM stage, while levels ofPeak9are positively correlated with TNM stage. Levels of Peak1and4falled backafter radical gastrectomy for cancer, while the levels of GCglyco A and GCglyco Brebounded. Among the gastric cancer, atrophic gastrisis and healthy control, sera levelof total core fucose residues of gastric cancer is the lowest, Fut8mRNA expressiondecreased in cancer tissue when comparing with that in adjacent tissue, however, levelof total core fucose residues and the expression of GDP-fuc-Tr mRNA keptunchanged statistically. The TCM syndromes varied among different TNM stagegroups. Among CEA, CA19-9, CA724, and CA125，only sera CA724levels of thegastric cancer patients in different TCM syndrome groups are statistically different.N-glycome profiles changed among gastric cancer patients with different TCMsyndromes, embodied in levels of Peak1,6and GCglyco A.CONCLUSIONS: The sera N-glycome profile of the gastric cancer patients’ changeddistinguishedly when compared with those of the atrophic gastrisis patients’ andhealthy controls’. N-glycome markers are highly efficiated and can be used as tumormarkers. N-glycome profile changed with the advancement of TNM stage andcurative operation. Decreased level of Fut8may be responsible for the reduction oftotal core fucose. TCM syndromes are correlated with TNM staging, some commonlyused tumor markers and N-glycome profile.