The endothelial lipase protein is promising urinary biomarker for diagnosis of gastric cancer

BACKGROUND: Gastric cancer is one of the most common malignant tumors in the world. Finding effective diagnostic biomarkers in urine or serum would represent the most ideal solution to detecting gastric cancer during annual physical examination. This study was to evaluate the potential of endothelial lipase (EL) as a urinary biomarker for diagnosis of gastric cancer. METHODS: The expression levels of EL was measured using Western blotting and immunohistochemical staining experiments on (tissue, serum, and urine) samples of gastric cancer patients versus healthy people. We also checked the EL levels in the urine samples of other cancer types (lung, colon and rectum cancers) and benign lesions (gastritis and gastric leiomyoma) to check if EL was specific to gastric cancer. RESULT: We observed a clear separation between the EL expression levels in the urine samples of 90 gastric cancer patients and of 57 healthy volunteers. It was approximately 9.9 fold average decrease of the EL expression levels in the urine samples of gastric cancer compared to the healthy controls (P <0.0001), achieving a 0.967 AUC value for the ROC (receiver operating characteristic) curve, demonstrating it’s highly accurate as a diagnostic marker for gastric cancer. Interestingly, the expression levels of EL in tissue and serum samples were not nearly as discriminative as in urine samples (P = 0.90 and P = 0.79). In immunohistochemical experiments, positive expression of the EL protein was found in 67% (8/12) of gastric adjacent noncancerous and in 58% (7/12) of gastric cancer samples. There was no significant statistical in the expression levels of this protein between the gastric cancer and the matching noncancerous tissues (P =0.67). CONCLUSIONS: The urinary EL as a highly accurate gastric cancer biomarker that is potentially applicable to the general screening with high sensitivity and specificity. VIRTUAL SLIDES: The virtual slide(s) for this article can be found here: http://www.diagnosticpathology.diagnomx.eu/vs/4527331618757552