|| Checking for direct PDF access through Ovid
A total of 5,209 asymptomatic, middle-aged subjects, whose serum pepsinogen (PG) andHelicobacter pyloriantibody levels had been assessed, were followed for 10 years. Subjects with positive serumH. pyloriantibodies (>50 U/mL) had an increased cancer risk (HR = 3.48, 95% CI = 1.26–9.64). Risk of gastric cancer increased as the antibody level increased; theH. pylori-positive group with antibody levels >500 U/mL had the highest incidence rate (325/100,000 person-years). Cancer development also increased with a reduced serum PG I level or a reduced PG I/II ratio; the risk was significantly elevated with serum PG I level ≤30 ng/mL (HR = 3.54, 95% CI = 1.95–6.40) or PG I/II ratio ≤3.0 (HR = 4.25, 95% CI = 2.47–7.32). Furthermore, the risk of diffuse-type cancer increased as PG II level increased; it was significantly elevated with PG II level ≥30 ng/mL (HR = 3.81, 95% CI = 1.10–13.21). UsingH. pyloriantibody and PG levels, subgroups with an especially high or low cancer incidence rate could be identified.H. pylori-negative or indeterminate subjects with low PG level (PG I ≤30 ng/mL or PG I/II ratio ≤2.0) orH. pylori-positive subjects with antibody levels >500 U/mL and a low PG level were among the subgroups with a high cancer incidence rate (over 400/100,000 person-years). In contrast,H. pylori-negative subjects with a PG I level >70 ng/mL or a PG I/II ratio >3.0 had the lowest risk; none of these subjects developed cancer. Thus, serum PG levels and/orH. pyloriantibody levels can be used to predict the risk of cancer in individuals withH. pylori-related gastritis from the general population. © 2008 Wiley-Liss, Inc.