ASPARTIC PROTEINASES AND GASTRIN IN THE DIAGNOSIS OF GASTRIC-CANCER AND GASTRIC PRECANCEROUS CHANGES

Objective: Gastric cancer patients have been reported to have low peps inogen I (PCA), increased pepsinogen II (PGC) levels and a reduced PGA /PGA ratio. We tested PGA, PGC and gastrin (C) levels, and the PCA/PGC ratio to verify the usefulness of these markers and of a new index (P GA x C) in the diagnosis of gastric cancer. Patients: We enrolled 51 p atients with gastric cancer; 23 patients with epithelial dysplasia, 14 5 with chronic atrophic gastritis, 40 with gastric ulcer, 25 with duod enal ulcer, and 53 subjects lacking major or minor endoscopic and micr oscopic changes at gastroscopy were included as controls. Methods: PGA , PGC and gastrin levels were determined by radioimmunoassay. Statisti cal analysis was performed by one-way analysis of variance, the Krushk all-Wallis test, the Kolmogorov-Smirnov test, receiver operating chara cteristic curves and the Youden J test. Results: PGA levels and the PG A/PGC ratio were significantly reduced in gastric cancer patients (P < 0.001). No significant variations were detected in PGC or gastrin lev els. The index number (PCA x G) was also clearly reduced in gastric ca ncer patients (P < 0.001). With a cut-off point chosen using the recei ver operating characteristic curve, this 'marker' showed, in our endos copic population, very high sensitivity (92%), specificity (94%), posi tive predictive value (73%) and overall accuracy (72% by the Youden J test) for cancer. Conclusions: If these results are confirmed in popul ations at high or very low risk, PGA x G could become a useful marker for gastric cancer.