Periodontal Pathogens Are A Risk Factor For Oral Cavity Squamous Cell Carcinoma, Independent Of Tobacco Use, Human Papillomavirus, And Heavy Alcohol Drinking

Introduction: The prevalence of oral cavity squamous cell carcinoma (OC-SCC) has declined only moderately despite a drastic reduction in the prevalence of cigarette smoking in USA since 1975. New OC-SCC cases often lack risks related to smoking and human papillomavirus (HPV). Thus, a significant proportion of newly diagnosed oral cancer cases have no known risk factor. A new paradigm is needed to guide our understanding of the unidentified risk factors responsible for OC-SCC. Very recently, several studies reported alteration of microbiome in oral cancer. However, the significance of these studies is uncertain for OC-SCC, due to the lack of statistical correction for false discoveries associated with multiple comparisons, heterogeneous patient population with mixed oropharyngeal and OC-SCC, and no control for important confounding risk factors for OC-SCC.

Material and method: To explore new risk factors of OC-SCC, we compared the oral microbiome among HPV-negative nonsmokers with OC-SCC, the precancerous lesion oral leukoplakia, and normal oral health, using 16S rRNA gene sequencing, and we controlled for heavy drinking in the comparisons.

Results and discussion: The microbiome was globally altered in diseases compared with healthy controls. Three periodontal pathogens Fusobacterium, Prevotella, Alloprevotella were enriched and commensal Streptococcus depleted in OC-SCC independent of heavy alcohol consumption. Based on the four genera plus a marker genus Veillonella for leukoplakia, the oral microbiome can be classified into two types of microbiome, periodontal pathogen-low and periodontal pathogen-high, correlated with the diseases and negative controls, respectively, with 81.58% of accuracy in the prediction of disease states. Network analysis revealed collaborative relationships among the pathogens and among commensals but inhibitive relationships between pathogens and commensals. Importantly, Capnocytophaga, a putative periodontal pathogen, was significantly more abundant in patients with recurrence of OC-SCC. To explore potential mechanisms, we deduced microbial gene and pathway contents based on 16S rRNA genes found in the samples. Overall, the microbiome showed a progressive enrichment of proinflammatory genes encoding HSP90A and ligands for TLR1 (lipoproteins) and enzymes for synthesis of ligands for TLR2 (peptidoglycans) and TLR4 (LPS) along the controls→leukoplakia→OC-SCC sequence. Periodontal pathogens were the major source for the enrichment of these genes.

Conclusion: Our findings suggest periodontal pathogens are an independent risk factor in subjects lacking major risks of OC-SCC. Microbiome-mediated inflammation might contribute to the development of OC-SCC. Further studies are needed to determine whether bacterial HSP90 or TLR ligands contribute to the development of leukoplakia and OC-SCC.