Oral Microbiota Community Dynamics Associated With Oral Squamous Cell Carcinoma Staging

Abstract

Oral squamous cell carcinoma (OSCC) is a highly aggressive cancer and the fourth leading malignancy among males in Taiwan. Some pathogenic bacteria are associated with periodontitis and oral cancer. However, the comprehensive profile of the oral microbiome during the cancer's progression from the early stage to the late stage is still unclear. We profiled the oral microbiota and identified bacteria biomarkers associated with OSCC. The microbiota of an oral rinse from 51 healthy individuals and 197 OSCC patients at different stages were investigated using 16S rRNA V3V4 amplicon sequencing, followed by bioinformatics and statistical analyses. The oral microbiota communities from stage 4 patients showed significantly higher complexity than those from healthy controls. The populations also dynamically changed with the cancer's progression from stage 1 to stage 4. The predominant phyla in the oral samples showed variation in the relative abundance of Fusobacteria, Bacteroidetes, and Actinobacteria. The abundance of Fusobacteria increased significantly with the progression of oral cancer from the healthy controls (2.98%) to OSCC stage 1 (4.35%) through stage 4 (7.92%). At the genus level, the abundance of Fusobacterium increased, while the number of Streptococcus, Haemophilus, Porphyromonas, and Actinomyces decreased with cancer progression. Fusobacterium periodonticum, Parvimonas micra, Streptococcus constellatus, Haemophilus influenza, and Filifactor alocis were associated with OSCC, and they progressively increased in abundance from stage 1 to stage 4. The abundances of Streptococcus mitis, Haemophilus parainfluenzae, and Porphyromonas pasteri were inversely associated with OSCC progression. We selected a bacterial marker panel of three bacteria (upregulated F. periodonticum, down-regulated S. mitis, and P. pasteri), which had an AUC of 0.956 (95% CI = 0.925-0.986) in discriminating OSCC stage 4 from the healthy controls. Furthermore, the functional prediction of oral bacterial communities showed that genes involved in carbohydrate-related metabolism, such as methane metabolism, and energy-metabolism-related parameters, such as oxidative phosphorylation and carbon fixation in photosynthetic organisms, were enriched in late-stage OSCC, while those responsible for amino acid metabolism, such as folate biosynthesis and valine, leucine, and isoleucine biosynthesis, were significantly associated with the healthy controls. In conclusion, our results provided evidence of oral bacteria community changes during oral cancer progression and suggested the possibility of using bacteria as OSCC diagnostic markers.

Keywords: 16S rRNA sequencing; cancer progression; community dysbiosis; complexity; oral microbiome.

Figure 1

Diversity estimate calculations of bacterial taxa between healthy and OSCC patients. The Shannon index, Chao index, observed otus, and PD whole tree of the oral microbiome from healthy controls and OSCC patients. Horizontal lines represent mean values. p < 0.05 indicates statistical significance using a non-parametric Mann-Whitney U test.

Figure 2

Principal component analysis (PCoA) and clustering analysis in healthy and OSCC patients. The oral microbiome compositions of healthy individuals and OSCC patients were analyzed using PCoA analysis and clustering analysis. (A) The individual samples are color coded to indicate healthy controls, stage 1, stages 2 and 3, and stage 4. The plot was analyzed and generated using the Bray-Curtis distances. (B) Hierarchical clustering was done using the healthy and OSCC stage 4 groups.

Figure 3

Relative abundance of different bacteria phyla in healthy and OSCC patients. Bacteria taxonomic profiling at the phylum level of oral microbiome from healthy controls and OSCC patients. (A) The five most abundant bacteria phylum in the oral microbiome. (B) Box plots show the relative abundance of Fusobacteria, Bacteroidetes, and Actinobacteria in healthy and OSCC patients. Horizontal lines represent mean values. p < 0.05 indicates statistical significance using a nonparametric Mann-Whitney U test.

Figure 4

Relative abundance of the most prevalent genera in healthy and OSCC patients. Bacteria taxonomic profiling at the genus level of the oral microbiome from healthy controls and OSCC patients. (A) The 11 bacteria at the genus level in the oral microbiome. The box plots show the relative abundance of Fusobacterium (B) and Streptococcus, Haemophilus, Porphyromonas, and Actinomyces (C) in healthy controls and OSCC patients. Horizontal lines represent mean values. p < 0.05 indicates statistical significance using a nonparametric Mann-Whitney U test.

Figure 5

The differentially enriched bacteria in OSCC patients when compared to healthy controls. (A) Linear discriminant analysis effect size (LEfSe) analysis showing bacteria at the species level that were altered between the OSCC stage 4 and healthy controls. The relative abundance of the upregulated bacteria (B) and downregulated bacteria (C) at the species level in OSCC.

Figure 6

Differentially enriched functions between healthy controls and OSCC patients by PICRUSt analysis. (A) Comparison of the relative abundance of the PICRUSt-generated functional profile of the saliva microbiome in OSCC patients and healthy controls. (B) The upregulated pathways in OSCC stage 4 patients. (C) The down-regulated pathways in OSCC stage 4 patients.