Design and planned analyses of an ongoing randomized trial assessing the preventive effect of Helicobacter pylori eradication on occurrence of new gastric carcinomas after endoscopic resection

Background A causal relationship between Helicobacter pylori infection and gastric cancer has been established. A nonrandomized study has shown eradication of H. pylori after endoscopic resection (ER) of early gastric cancer inhibits development of new carcinomas. Eligible subjects are patients with H. pylori infection who are newly diagnosed with early gastric cancer and plan to have ER or who are in the post-resection follow-up phase after ER time of enrollment. Patients are randomly allocated to the eradication or the control arms (no eradication and standard of care). Patients will be evaluated by endoscopy at 0.5, 1, 2, and 3 years after randomization. Diagnosis of a new carcinoma at another site of the stomach is defined as primary endpoint, and recurrence of tumors at the resection site as a secondary endpoint. In addition to intention-to-treat and per-protocol analyses using proportional hazards models, time to recurrence will be compared between treatment and control using multiple logistic regression analyses. In the latter two situations, the models will be adjusted for the factors exerting significant influences on the results. Five hundred and forty-two subjects have been enrolled into the study and are being followed-up. This study will have the statistical power to demonstrate whether H. pylori eradication therapy exerts any clinically relevant inhibitory effects on occurrence or recurrence of gastric cancer. In addition, it will be able to test the hypothesis that H. pylori infection is a promoter in gastric carcinogenesis.