Abstract
Epstein-Barr virus (EBV) and Helicobacter pylori (H. pylori) are two pathogens associated with the development of various human cancers. The coexistence of both microorganisms in gastric cancer specimens has been increasingly reported, suggesting that crosstalk of both pathogens may be implicated in the carcinogenesis process. Considering that chronic inflammation is an initial step in the development of several cancers, including gastric cancer, we conducted a systematic review to comprehensively evaluate publications in which EBV and H. pylori co-infection has been documented in patients with non-malignant gastroduodenal disorders (NMGDs), including gastritis, peptic ulcer disease (PUD), and dyspepsia. We searched the PubMed database up to August 2019, as well as publication references and, among the nine studies that met the inclusion criteria, we identified six studies assessing EBV infection directly in gastric tissues (total 949 patients) and three studies in which EBV infection status was determined by serological methods (total 662 patients). Due to the substantial methodological and clinical heterogeneity among studies identified, we could not conduct a meta-analysis. The overall prevalence of EBV + H. pylori co-infection in NMGDs was 34% (range 1.8% to 60%). A higher co-infection rate (EBV + H. pylori) was reported in studies in which EBV was documented by serological methods in comparison with studies in which EBV infection was directly assessed in gastric specimens. The majority of these studies were conducted in Latin-America and India, with most of them comparing NMGDs with gastric cancer, but there were no studies comparing the co-infection rate in NMGDs with that in asymptomatic individuals. In comparison with gastritis caused by only one of these pathogens, EBV + H. pylori co-infection was associated with increased severity of gastric inflammation. In conclusion, only relatively small studies testing EBV and H. pylori co-infection in NMGDs have been published to date and the variable report results are likely influenced by geographic factors and detection methods.