Analysis On Differential Proteomic Fingerprints For Advanced Gastric Cancer Patients With LGT Fingerprints Changing

Object:To investigate the relationship between the features (sex, age, surgery, TNM Staging, chemotherapy, tumor marker, LGT fingerprints) and prognosis of gastric cancer.Methods:Clinical data of 149 patients with primary gastric cancer who first time be in hospital between 2007 and 2008. Select sex, age, surgery, TNM Staging, chemotherapy, tumor marker, and LGT fingerprints as observed factors, investigate the relationship between them and prognosis of gastric cancer.The emphasis is to research the relationship between LGT fingerprints and prognosis of gastric cancer, and between LGT fingerprints and tumor markers.Statistical calculations were performed using Statistical Package for Social Science software SPSS 16.0. The statistical methods were Kaplan-Meier, Log-Rank test, COX proportional hazard model, and Pearson correlation analysis.Results:In Single factor analysis:Surgery, TNM Staging, chemotherapy, tumor markers (CEA, CA199, CA242, CA724), the cluster and the abundance of lost goodwill target (LGT) fingerprints were influencing factors about prognosis of gastric cancer. In multiple factor analysis:Surgery, the cluster of LGT fingerprints, TNM Staging, the abundance of LGT fingerprints, and CA724 were independent influencing factor.The patients with negative expression of LGT had low risk of death, which were lower than those of single-peak and double-peak groups, especially lower than those of tri-peak and multi-peak groups; the risk of death among the tri-peak and multi-peak groups were much higher than single-peak, double-peak groups, the difference had statistical significance (P<0.05), and the death rate is rather high.In the correlation analysis, the cluster and the abundance of LGT fingerprints was positive correlation. In the survival group, the abundance of LGT was independence with tumor markers, in the death group, the abundance of LGT was positive correlation with CEA, CA199, CA242, but the correlation coefficient is low, and was independence with CA724. In the survival group, the abundance of LGT fingerprints was independence with CEA and CA724, and was positive correlation with CA199, CA242, in the death group, the abundance of LGT fingerprints was positive correlation with CEA, CA199, CA242 and CA724. And between the four tumor markers, the correlations were existence too.Conclusion:1. In Single factor analysis:Surgery, TNM Staging, chemotherapy, tumor markers (CEA, CA199, CA242, CA724), the cluster and the abundance of lost goodwill target (LGT) fingerprints were influencing factors about prognosis of gastric cancer. 2. In multiple factor analysis:Surgery, the cluster of LGT fingerprints, TNM Staging, the abundance of LGT fingerprints, and CA724 were independent influencing factor.3. The expression of LGT with tri-peak and multi-peak suggested that patient's state of illness is about to deteriorate. This kind of fingerprint spectrum can be regarded as positive indicators, the expression of LGT with single-peak, double-peak can be regarded as suspect indicators, and the negative expression of LGT suggested that patient will has a good prognosis with a very low risk of death in the near future.4. In the death group, the abundance of LGT fingerprints was positive correlation with CEA, CA199,CA242 and CA724.