Ion Chromatography Based Urine Amino Acid Profiling Applied for Diagnosis of Gastric Cancer

Aim. Amino acid metabolism in cancer patients differs from that in healthy people. In the study, we performed urine-free amino acid profile of gastric cancer at different stages and health subjects to explore potential biomarkers for diagnosing or screening gastric cancer. Methods. Forty three urine samples were collected from inpatients and healthy adults who were divided into 4 groups. Healthy adults were in group A (n = 15), early gastric cancer inpatients in group B (n = 7), and advanced gastric cancer inpatients in group C (n = 16); in addition, two healthy adults and three advanced gastric cancer inpatients were in group D (n = 5) to test models. We performed urine amino acids profile of each group by applying ion chromatography (IC) technique and analyzed urine amino acids according to chromatogram of amino acids standard solution. The data we obtained were processed with statistical analysis. A diagnostic model was constructed to discriminate gastric cancer from healthy individuals and another diagnostic model for clinical staging by principal component analysis. Differentiation performance was validated by the area under the curve (AUC) of receiver-operating characteristic (ROC) curves. Results. The urine-free amino acid profile of gastric cancer patients changed to a certain degree compared with that of healthy adults. Compared with healthy adult group, the levels of valine, isoleucine, and leucine increased (P < 0.05), but the levels of histidine and methionine decreased (P < 0.05), and aspartate decreased significantly (P < 0.01). The urine amino acid profile was also different between early and advanced gastric cancer groups. Compared with early gastric cancer, the levels of isoleucine and valine decreased in advanced gastric cancer (P < 0.05). A diagnosis model constructed for gastric cancer with AUC value of 0.936 tested by group D showed that 4 samples could coincide with it. Another diagnosis model for clinical staging with an AUC value of 0.902 tested by 3 advanced gastric cancer inpatients of group D showed that all could coincide with the model. Conclusions. The noticeable differences of urine-free amino acid profiles between gastric cancer patients and healthy adults indicate that such amino acids as valine, isoleucine, leucine, methionine, histidine and aspartate are important metabolites in cell multiplication and gene expression during tumor growth and metastatic process. The study suggests that urine-free amino acid profiling is of potential value for screening or diagnosing gastric cancer.