Oral squamous cell cancer (OSCC) normally occurs in elderly subjects between 50 and 70 years (Sarode et al., 2020) and poses a serious threat to patients’ health and quality of life. However, since the 1990s, the incidence of oral cancer in people younger than 50 years old has been continuously rising (Tota et al., 2017). Clinical studies showed that young oral cancer subjects had higher malignant levels than elderly subjects, and they were more vulnerable to local relapse and lymph node metastasis; therefore, aggressive treatment strategies were conducted (Sarkaria and Harari, 1994; Goldstein and Irish, 2005). Hence, determining the unique pathogenesis of OSCC in young subjects is vital for performing effective interventions and improving survival rate.

Currently, research on the pathogenesis of young OSCC patients is not sufficient to explain the rise in incidence. Studies have shown that classic carcinogens, such as smoking, drinking and HPV infection, are not the main reasons for the rise in incidence as young subjects have short period of time to be exposed to those carcinogens (Sarode et al., 2021). The genomic studies showed that mutations in young and elderly OSCC patients were similar, which were mainly in the genes such as TP53, CDKN2A, NOTCH1, CASP8, FAT1, PIK3CA, and MLL2 (Kandoth et al., 2013; Pickering et al., 2014), but study also reported OSCC patients with younger age appeared to have an attenuated immune response due to lower load of mutations (Maroun et al., 2021), and EGFR was reported to have higher rate of amplification in young patients (Satgunaseelan et al., 2021). However, the mechanism underlying young OSCC incidence elevating cannot be merely explained by genomic alteration. Recent studies showed that the occurrence of OSCC was related to dysbiosis of some microbes other than HPV, since they have various carcinogenic potentials through different mechanisms (Sun et al., 2020; Tuominen and Rautava, 2021). Systematic diseases, including cancers, are not only related to single pathogens but also affected by the microbiota (Kau et al., 2011). A large number of various bacterial species have been found in tumor tissue, mainly in tumor cells and macrophages of the tumor microenvironment (Nejman et al., 2020). Pathogenic bacteria in the oral cavity provoke or aggravate inflammation, promote cell proliferation and inhibit cell apoptosis through metabolic activities (Hooper et al., 2009). Studies have also shown that compared to healthy subjects, the number of anaerobes were significantly increased in the microbiota of the OSCC oral cavity, including Porphyromonas gingivalis, and Fusobacterium, while Streptococcus and Rothia were sharply decreased (Hooper et al., 2006). The dysbiosis of microbiota is firmly related to the development of OSCC.

There is a dynamic change in microbiota composition as age increases, with an increase in alpha diversity and a drop in beta diversity, which lead to differences in immune status, hormone and metabolism levels. The current study focused on the compositional and functional differences between the microbiota of young and elderly OSCC subjects and aimed to explain the role of distinct microbiota in cancer. The findings further explain the mechanism of increased incidence of OSCC in young subjects and provide potential targets for clinical research.

A total of 5,181,923 raw reads were obtained from the sequencing, 4,657,023 (116,425.6 in average) filtered reads were remained for further analysis, after removing the merged and chimeric reads, a total of 4,261,032 high-quality sequences (representing 82% of the total sequences) were obtained from the oral cancer samples, with an average of 106,525.8 sequences for each subject. The average length of the sequences was 422 bp, with the maximum length reaching 441 bp and the shortest length being 97 bp. Clustering of all high-quality sequences at 97% identity resulted in 11,414 taxonomic units; on average, 8.6 phyla, 20.17 classes, 34.28 orders, 55.5 families, and 83.52 genera were distributed into each subject (Supplementary Figure 1). The composition of OTU across different taxonomic levels were shown in Supplementary Figure 2.

At the phylum level, the composition of the younger and elderly OSCC microbiomes was similar; they were mainly composed of Proteobacteria, Bacteroidetes, Firmicutes, and Fusobacteria (Figure 1A). The relative abundance of Proteobacteria in the younger and elderly groups occupied the dominant position, with proportions of 65.5 and 57.1%, respectively. The microbiota at the genus level were more differentiated than the comparison at phylum level. Leptothrix was the most dominant genus in both groups, but in tissue from younger OSCC patients, the relative abundances of Ochrobactrum, Prevotella, and Ralstonia were significantly higher, while in the elderly group, Pedobacter was more abundant. Fusobacterium had a similar position in both groups, with a relative abundance of approximately 5% (Figure 1B). The relative analysis showed that, based on the abundance of unfiltered (Figure 2A) and filtered (Figure 2B) OTU data, the genera were divided into two clusters spontaneously by their age groups, taxa under Proteobacteria were most shown.

It is a global issue that the cancer incidence in young people (under 50 years old) is continuously increasing (Siegel et al., 2020). Adolescent and young adult (AYA) cancers represent a unique spectrum of malignancies (Gupta et al., 2020), and the rapidly growing numbers of subjects have resulted in an epidemiological problem. In the current study, we focused on the increasing incidence of OSCC in young adults. The demographics showed that the gender distribution in the two groups is unbalanced, with more female in the younger group and more male in elder group. Since the differences in hormones and metabolisms may lead to shift of the microbiota, the confounding effects need to be considered. The underlying mechanism of young OSCC incidence involves a complex combination of changes in environmental and societal factors. When genomic studies failed to reveal the characteristics of young OSCC patients, we put our efforts into studying oral microbiota. The oral cavity is a relatively exposed environment and the starting point of the digestive tract, and it contains approximately 1,000 species of commensal microbes with crucial roles related to the health status of individuals (Sun et al., 2020). Therefore, by investigating the composition of microbiota in young and elderly OSCC tissues, we hoped to explain the differences between them and the consequences of those differences.

Humans are considered meta organisms that consist of themselves as well as various microbial communities (Greer et al., 2016). Up to trillions of individual bacterial cells have colonized the surface of the mouth, skin, airways, vagina, genitourinary tract, and gastrointestinal tract. The oral microbiota mainly belongs to the phyla Firmicutes, Actinobacteria, Bacteroidetes, Proteobacteria, Fusobacteria, and Spirochaetes (Verma et al., 2018), which was proven by our results. At the genus level, the most common genus was Streptococcus, followed by Haemophilus, Actinomyces, and Prevotella (Gao et al., 2018). Although not as well studied as the gut microbiome, some evidence has shown that there are links between increasing age and microbiota shifting, and changes in the immune system may play a role, leading to chronic inflammation in the oral cavity and increasing an individual’s susceptibility to various oral diseases (Feres et al., 2016; Belibasakis, 2018). Studies have reported that in healthy subjects, the difference in oral microbiota was not significant between young and elderly individuals, but in individuals with oral diseases, some taxa were observed to be increased in young and elderly individuals (Feres et al., 2016).

The compositional analysis of the young and elderly OSCC microbiota showed that Proteobacteria was most predominant in both groups; Proteobacteria is normally seen in the microbiota of OSCC tissue, as the acidic and anaerobic environment is suitable for the growth of members under this phylum (Ringel et al., 2015). However, at the genus level, we found that the relative abundances of Ochrobactrum, Prevotella, and Ralstonia were significantly higher in younger group. Species of Ochrobactrum are seen in prostate cancer and bladder cancer (Apisarnthanarak et al., 2005; Yu et al., 2015). They are usually observed in patients with immunological suppression (Qasimyar et al., 2014), and some members of this genus may increase resistance to antibiotics (Higgins et al., 2001). Prevotella has been well studied in the oral microbiota; it is the second most abundant oral genus in the mouth, exceeding 10% of the whole microbiome (Segata et al., 2012). Prevotella is related to poor oral hygiene and the cause of oral diseases, such as prevalent gingivitis and periodontitis (Teles et al., 2012; Larsen, 2017). The anaerobic species of Prevotella are responsible for chronic inflammatory processes, and the inflammatory mediators produced by them lead to disorders of fibroblasts, epithelial cells and extracellular matrix components, increasing the risk of tumorigenesis (Hoare et al., 2019). Species of Ralstonia are opportunistic pathogens that are related to various human diseases; they are normally seen in infections of the respiratory system (Block et al., 2013), and their high abundance is associated with persistent inflammation and the development of cancer (Sung et al., 2020). The high abundance of these genera indicates a poor status of the oral cavity microenvironment.

On the other hand, the LEfSe results indicated that Betaproteobacteria, Burkholderiales, Ralstonia, Burkholderiaceae, and Rhizobiales were potential biomarkers for the young OSCC microbiota, while Enterobacteriaceae, Enterobacterales, Sphingobacteriia, Sphingobacteriales, and Pedobacter were potential biomarkers for the elderly OSCC microbiota. The results of random forest analysis showed that Caulobacter, Pedobacter, Ralstonia, Escherichia, Flavitalea, and Ochrobactrum were effective genera in differentiating the two groups by the increase in MSE and increase in Node Purity. We examined the relationship between the biomarker and clinical information, and the results showed that the top taxa in the young OSCC microbiota were more enriched in early tumor stages, while the biomarkers for the elderly OSCC microbiota were more related to late tumor stages. Based on the compositional analysis and clinical correlation analysis of the microbiota between the two groups, we assumed that the microbiota of younger OSCC is more associated with inflammation, while the elder OSCC microbiota was more likely to help the tumorigenesis to modify the oral microenvironment.

To study the functional variance that may be caused by the compositional difference, PICRUSt2 was performed. From the comparison of the detected taxa and the functional predictions between the two groups, several metabolic pathways were found more active in the younger OSCC group. Vitamin B6 serves as an antioxidant and anti-inflammatory molecule, and its degradation is related to dysbiosis of immunity and gene expression (Bird, 2018). 2-Aminophenol has been proven to exhibit antitumor effects (Pujante-Galián et al., 2020), and the degradation of these components promotes tumorigenesis in young subjects. The super pathway of oxidation of C₁ compounds to CO₂ and formaldehyde assimilation I are normally seen in pathogenic bacteria (Buzolyova and Somov, 1999). KEGG analysis showed that the microbiota in OSCC tissue of young subjects may work differently in the pathways of antigen processing and presentation, base excision repair, the HIF-1 signaling pathway, the calcium signaling pathway and Fcγ R-mediated phagocytosis.

Taken together, the current study revealed compositional and functional differences between the microbiota of OSCC tissues from young and elderly subjects. The results showed that Ochrobactrum, Prevotella, and Ralstonia were enriched in the young OSCC microbiota and may promote tumorigenesis by causing inflammation and opportunistic infection. From the view of functional prediction, the microbiota of young OSCC patients was more enriched in the degradation of antitumor components and increased oxidative stress. By analyzing the microbiome from the two groups of subjects, the current study explained the difference between the tumorigenesis of young and elderly OSCC subjects from a new point of view. There were several limitations of the current research: (1) the study had a limited sample size, and the findings from the analysis of the microbiota were not linked to pathogenesis factors; (2) 16S rDNA amplicon analysis was not accurate enough to study at optimal resolution and thus, did not allow for precisely functional predictions; and (3) proper negative control need to be set during the sample collection procedure to better avoid contamination. In future studies, whole genome shotgun strategies and cell biological experiments will be performed to research the relationship between microbiota shifts and their effects on the tumorigenesis of young subjects.

The data presented in the study are deposited in the SRA repository, accession number PRJNA803155.

The studies involving human participants were reviewed and approved by the Institutional Review Board of Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine. The patients/participants provided their written informed consent to participate in this study.

WC designed the project. QF and ML provided suggestions for the project. QX, XP, and ZL collected the clinical samples. ZZ and ML performed the experiments, data statistics, and bioinformatics analysis. ZZ wrote the manuscript. WC and ZZ revised the manuscript. All authors read and approved the final manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.