CA 50 determination in body fluids: can we screen patients at risk for gastric cancer?

CA 50 (a tumor-associated gangliosidic antigen) levels have been determined by an RIA test in serum, gastric juice and urine of patients undergoing upper gastrointestinal tract endoscopy: 22 control subjects (no macroscopic or microscopic lesions), 29 patients with chronic atrophic gastritis, 20 with epithelial dysplasia and 16 with gastric cancer entered the study. Gastric juices were also tested for pH, protein concentration and specific gravity, urines for protein concentration and osmolarity. Serum and gastric juices were also tested for CEA levels and the results obtained with the two markers compared. In patients with gastric cancer, CA 50 gastric juice levels were statistically higher than in controls; a wide overlap was however present among groups, and sensitivity and specificity were respectively 38% and 85% for serum and 69% and 82% for gastric juice. Sensitivity and specificity were respectively 23% and 89% for CA 50 determination in urines. In this case, no statistically significant difference was observed between gastric cancer and control patients. A trend toward higher median values was observed in advanced with respect to early gastric cancer. A correlation was found between gastric juice and serum CA 50 levels, as well as between serum and urine levels of the marker. A correlation was also observed between CA 50 values and protein concentration in gastric juice and with osmolarity in urines. Overall, CA 50 levels were statistically higher in patients with intestinal metaplasia than in those who did not present the lesion. Increased CA 50 gastric juice levels are also observed in patients with chronic atrophic gastritis and epithelial dysplasia. CA 50 gastric juice and urine levels appear to be dependent, at least in part, on the concentration of the fluid.