Background: Although oral hygiene and health have long been reported to be associated with increased risk of gastric cancer (GC), the direct relationship of oral microbes with the risk of GC have not been evaluated fully. We aimed to test whether tongue coating microbiome was associated with GC risk. Methods: Pyrosequencing of 16S rRNA gene of tongue coating microbiome was used in 57 newly diagnosed gastric adenocarcinomas and 80 healthy controls. Benjamini-Hochberg (BH) was applied for multiple comparison correction. Co-abundance group (CAGs) analysis was adopted. Results: We found that higher relative abundance of Firmicutes, and lower of Bacteroidetes were associated with increased risk of GC. In genus level, Streptococcus trended with a higher risk of GC, the four other genera (Neisseria, Prevotella, Prevotella7, and Porphyromonas) were found to have a decreased risk of GC. Different from overall GC and non-cardia cancer, Alloprevotella and Veillonella trended with the higher risk of cardia cancer. Finally, we analyzed the microbiota by determining CAGs and six clusters were identified. Except the Cluster 2 (mainly Streptococcus and Abiotrophia), the other clusters had an inverse association with GC. Of them, the Cluster 6 (mainly Prevotella and Prevotella7 etc) had a relatively good classification power with 0.76 of AUC. Conclusion: Microbiome in tongue coating may have potential guiding value for early detection and prevention of GC.
Keywords: Biomarkers; Case-control; Gastric cancer; Microbiota; Tongue coating.