Randomized Multi-Intervention Trial to Inhibit Precancerous Gastric Lesions in Lingu, Shandong Province, China

The Division of Cancer Epidemiology and Genetics is conducting an 8-year collaborative randomized multi-intervention trial with the Beijing Institute for Cancer Research to evaluate the etiologic role of Helicobacter pylori, garlic, and certain micronutrients in the multi-step process of gastric carcinogenesis. The primary endpoint will not be cancer, but rather the precancerous lesions severe chronic atrophic gastritis, intestinal metaplasia, and dysplasia. The study is designed to detect a stabilizing or mild decrease in the expected age-related progression of precancerous lesions over the course of the trial. Thus the trial can be considered a study of the potential inhibitors of the process of gastric carcinogenesis. The trial will also provide an evaluation of therapy for H. pylori, including an assessment of reinfection rates, a critical need for formulating strategies for infection control in China and other developing countries. Participating in the trial are approximately 3400 adults aged 35-70 who were part of an ongoing survey of precancerous gastric lesions in 13 Linqu villages in Shandong Province. Linqu appears to be an ideal setting for the trial since stomach cancer rates are among the highest in the world, precancerous gastric lesions are prevalent, and the population is stable and well characterized. These individuals will be randomly assigned (taking H. pylori positivity into account) into 8 intervention groups according to a 2(3) factorial design. The interventions are: 1) initial treatment of H. pylori infection with omeprazole and amoxicillin followed by 2) daily supplementation with a combination of alpha-tocopherol, vitamin C, and selenium; and 3) daily supplementation with garlic extracts. Compliance was excellent and no serious side effects were seen from any of the interventions.

As part of the ongoing study in Linqu, all participants received an endoscopic exam in the fall of 1994. Repeat gastroscopic exams with biopsies at 7 standard gastric sites conducted during March to May, 1999 and March to April, 2003 to detect early cancers and to evaluate gastric mucosal status. The subjects will be categorized according to the most advanced lesions detected in all biopsies and assigned a severity score. The three major endpoints for analysis will be: 1) prevalence of dysplasia or cancer 2) prevalence of severe chronic atrophic gastritis, intestinal metaplasia, dysplasia, or cancer; and 3) average severity score.

The major endpoint paper was published in JNCI in 2006. Additional analyses of the trial data are underway. A continuation study is proceeding that will allow for follow-up of the approximately 3070 remaining trial participants through 2010 on the long-term effects of the previous treatments on gastric cancer incidence and on cause-specific death rates and allow approximately 364 participants in the previous study with advanced gastric lesions in 2003 an opportunity for annual endoscopic screening for gastric cancer.