Oral microbial alterations by smoking and metabolic factors in esophageal squamous cell carcinoma

Introduction

Background

Esophageal cancer ranks as the seventh leading cause of cancer-related deaths, often diagnosed at an advanced stage (1). In South Korea, gastric cancer screening with endoscopy has led to a slight rise in early esophageal cancer detection (2), with approximately 30% of cases diagnosed at stage I—higher than that in Western countries. However, advanced esophageal cancer remains predominant (3). Approximately 50% of eligible individuals undergo biannual endoscopic screening (2). With the decreasing prevalence of Helicobacter pylori, research is underway to develop individualized screening methods based on risk factors (4,5). For individuals without H. pylori infection or other risk factors, consideration is being given to extending the interval between endoscopic examinations (4,5). Endoscopic examinations are uncomfortable and invasive, with some deaths occurring during sedated endoscopies annually. Therefore, research into non-invasive methods for screening esophageal cancer is necessary.

Rationale and knowledge gap

The human microbiome coexists symbiotically with its host, and their interactions influence disease states (6). Environmental factors such as smoking can alter oral microbial communities and their secretions (7,8). Dysbiosis in the oral microbiome may be linked to diseases such as periodontitis (9) and oral squamous cell carcinoma (10). Alterations in the oral microbiome of Chinese patients with esophageal cancer, such as reduced diversity and distribution shifts, have been reported (11). The microbiota distribution within healthy populations can vary depending on epidemiologic factors such as age, sex, and metabolic factors (12-15), whereas it has been rarely reported how clinical factors impact microbiota composition in patients with esophageal cancer.

Objective

This study aimed to analyze oral microbiome differences between patients with esophageal squamous cell carcinoma (ESCC) and healthy controls and explore subgroup analyses based on clinical and metabolic factors. We present this article in accordance with the STROBE reporting checklist (available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-432/rc).

Methods

Participants

This study was conducted from December 2021 to May 2023 at five tertiary hospitals and included patients with ESCC (n=21) and health controls (n=20) aged 30 to 85 years (Figure 1A). Participants were excluded if they had a history of digestive system cancer within the past 5 years, recent antibiotic or probiotic use (within 4 weeks), uncontrolled diabetes, alcohol dependence, severe chronic illnesses, or impaired decision-making capacity. Healthy controls were defined as individuals who had undergone upper gastrointestinal endoscopy with no observed upper digestive tract neoplasms, such as esophageal, gastric, or duodenal cancer, or polyps.

Figure 1 Study design and overall microbiome analysis. To investigate oral microbiome differences, patients with esophageal squamous cell carcinoma (n=21) and healthy controls (n=20) were analyzed across two main groupings: (A) by cancer status and (B) by clinical variables. (C) α-diversity analysis (Shannon entropy) showed no significant (n.s) difference between patients and controls. (D) β-diversity clustering revealed no distinct separation between groups. (E) The composition of gut microbiome communities for each sample is shown in bar graphs. BMI, body mass index; LDL, low-density lipoprotein; TG, triglycerides.

Data collection

Comprehensive information about the subjects was collected through face-to-face interviews conducted by trained interviewers and medical records. The questionnaire included information on age, sex, height, weight, body mass index (BMI), presence of chronic diseases, medication history, alcohol consumption, smoking history, and family history. Medical records were used to confirm fasting glucose, cholesterol, triglycerides (TG), low-density lipoprotein (LDL), and high-density lipoprotein levels, basic blood tests, upper gastrointestinal endoscopy findings, and H. pylori infection status. In addition, for patients with ESCC, information on the imaging and pathological findings of esophageal cancer, the location and stage of esophageal cancer, degree of differentiation, and post-participation treatment methods were also confirmed.

This study was approved by the ethics committees of Kosin University Gospel Hospital (No. KUGH 2021-07-004), Kyungpook National University Chilgok Hospital (No. KNUCH 2021-05-039-001), Dankook University Hospital (No. DKUH 2021-06-026), Yeungnam University Hospital (No. 2021-08-050), and Soonchunhyang University Hospital (No. 2021-07-047) and was conducted in accordance with the ethical standards set by the responsible committees on human experimentation, both at the institutional and national levels, and with the Declaration of Helsinki and its subsequent amendments. All subjects provided written informed consent before the study.

Sample collection and DNA extraction

All participants were asked to maintain an empty stomach and refrain from oral hygiene procedures (such as brushing teeth and using dental floss) on the morning of the sampling. The sampling was performed between 7:00 and 8:00 AM for all participants. Participants underwent swab sample collection from the oral mucosa. Samples were stored at −80 ℃ until use. DNA was extracted using a DNeasyPowerSoil Pro Kit (Qiagen, Hilden, Germany) following the manufacturer’s instructions. The extracted DNA was quantified using Quant-IT PicoGreen (Invitrogen, Waltham, Massachusetts, USA). Detailed information is provided in Appendix 1.

Sequence data analysis

The amplicon sequence variants (ASVs) table obtained from Macrogen was analyzed using quantitative insights into microbial ecology 2 (QIIME2, RRID:SCR_021258) (16). The α- and β-diversity analyses were performed using the diversity analysis function in QIIME2 for the control and esophageal cancer groups as well as the subgroups with additional clinical variables. The subsampling depth for diversity analysis was set to 19,500. We used Shannon, evenness, and faith pd as indices of α-diversity to examine the diversity of species in the sample. For β-diversity, we performed principal coordinate analysis (PCoA) using the “phyloseq” and “microbiome” packages of QIIME2 and R (version 4.3.2, RRID:SCR_001905) based on the ASVs table to check the clustering of the groups. Two-dimensional PCoA plots are presented for comparison of the control and esophageal cancer groups and to identify clustering among subgroups. For TG and LDL, when subdividing groups based on clinical variables, samples without such information were excluded from the analysis. For group comparisons subdivided by esophageal cancer location, only samples from patients with esophageal cancer were used, excluding control samples. Details of the sequence data analysis were confirmed with the processing institution.

Taxa abundance analysis

To investigate the differences in taxa abundance between groups, the ASVs table was matched (annotated) with the greengenes (version 13_8, RRID:SCR_002830) database as a reference using the vsearch function in QIIME2 with similarity set to 99%. With the annotated taxa table, statistical analyses of metagenomic profiles (STAMP, RRID:SCR_018887) (17) were used to identify taxa that differed by more than 5% between groups or taxa that were statistically significantly different. STAMP is a tool that performs statistical hypothesis testing of differences between metagenome profiles for a defined set of groups and visualizes the results according to the differences.

Subgroup analysis by clinical variables

We used six clinical variables (age, BMI, smoking status, malignant disease history, TG, and LDL) to subdivide the control and patient groups (Figure 1B). For each clinical variable, age was subdivided into <65 and ≥65 years, BMI was subdivided into ≥25 and <25 kg/m², TG was subdivided into ≥150 and <150 mg/dL, and LDL was subdivided into >100 and ≤100 mg/dL. Samples without the corresponding clinical variables were excluded and analyses were performed on those with information. Furthermore, the patient group was subdivided according to esophageal cancer location. Diversity and taxa abundance analyses were again performed in each subgroup.

Statistical analysis and bioinformatics

Statistical methods (independent sample T-test and Chi-squared test) were used to compare differences in age, sex, BMI, smoking, alcohol consumption, family history, and frequency of chronic diseases between the two groups. α-diversity between groups was calculated using the Kruskal-Wallis test. PCoA analysis was used to estimate the similarity and clustering between samples in β-diversity. Additional analyses were conducted if necessary. In the taxa abundance analysis, statistical tests of two groups were performed by T-test, and multiple groups were performed by analysis of variance.

Results

Diversity and taxa proportion analyses between the control and esophageal cancer groups

In this study, there were 20 controls and 21 patients, of which 37 (37/41) were men (Figure 1A). The baseline characteristics of the groups are shown in Table 1. There were no significant statistical differences between the groups except for drinking and smoking status. We analyzed the differences in oral microbiome among the four groups by dividing both control and patient groups into two subgroups based on clinical variables (Figure 1B). When examining α-diversity (Figure 1C) and β-diversity (Figure 1D) of the oral microbiota between control and esophageal cancer groups, no statistically significant differences were observed.

At the phylum level, Bacillota accounted for the largest proportion in both groups, followed by Pseudomonadota and unclassified Bacteria (Figure 1E and Table 2). Of the eight identified phyla, Bacillota, unclassified Bacteria, Bacteroidota, and Mycoplasmatota were more prevalent in the patient group than in the normal group (Table 2). The mean percentages of the two groups were used to determine statistical differences, but there were no significant differences (Table 2).

Subgroup analysis: diversity and taxa abundance

Age

In the age-based subgroup analysis, the age <65 years patient group showed statistical differences compared to the other groups (Figure 2A). Clustering analysis also confirmed clustering between the age <65 years patient group and the other groups in one-to-one comparisons (Figure 2B-2D). Taxa analysis revealed differences in abundance among several genus- or species-level taxa between groups (Figure 2E). Veillonella dispar showed a higher mean proportion in the patients ≥65 years compared to patients <65 years, whereas Haemophilus parainfluenzae and Streptococcus infantis showed significantly lower mean proportions in the patients <65 years comparing to controls <65 years (Figure 2E and Table 3).

Figure 2 Age-stratified analysis of oral microbiome diversity in patients with esophageal squamous cell carcinoma and controls. (A) Faith PD index revealed significant α-diversity differences, particularly between the age <65 years patient and both the age <65 years control and age ≥65 years patient groups. β-diversity using two-dimensional principal coordinate analysis plots demonstrated distinct clustering patterns when comparing (B) age ≥65 years patient vs. age <65 years patient, (C) age <65 years patient vs. age <65 years control, and (D) age <65 years patient vs. age ≥65 years control groups. (E) Taxonomic analysis. The number above the dot indicates the confidence interval. *, P value <0.05. Faith PD index, faith’s phylogenetic diversity index.

Smoking

In the smoking-based subgroup analysis, the “never-smoking patient” (NSP) group showed a higher diversity than the “current or past smoking patient” (CSP) group (α-diversity by Shannon, Figure 3A). β-diversity analysis showed a clear clustering between the NSP and CSP groups (Figure 3B) and between the NSP and “never-smoking control” (NSC) groups (Figure 3C). Taxa abundance analysis revealed differences in several taxa (Figure 3D). Unclassified Clostridiales was higher in NSP than in NSC, whereas unclassified Filifactor, S. infantis, and Porphyromonas endodontalis were higher in NSP than in CSP (Figure 3D and Table 4).

Figure 3 Impact of smoking on oral microbiome diversity in patients with esophageal squamous cell carcinoma and controls. (A) Faith PD index revealed significant α-diversity differences between CSP and NSP. β-Diversity using two-dimensional principal coordinate analysis plots showed distinct microbial community clustering between (B) CSP vs. NSP and (C) NSP vs. NSC. (D) Taxonomic analysis (forest plot). The number above the dot indicates the confidence interval. *, P value <0.05. CSC, current smoking control; CSP, current smoking patient; faith PD index, faith’s phylogenetic diversity index; NSC, never-smoking control; NSP, never-smoking patient.

BMI

In the BMI-based subgroup analysis, patients with ESCC and BMI ≥25 kg/m² showed a higher diversity than those with BMI <25 kg/m² (Figure 4A,4B). Clustering comparisons showed a clear clustering between the BMI ≥25 kg/m² and BMI <25 kg/m² patient groups (Figure 4C) and between the BMI ≥25 kg/m² patient and BMI ≥25 kg/m² control groups (Figure 4D). Taxa abundance analysis showed that unclassified Neisseria, Prevotella pallens, and Rothia mucilaginosa showed lower mean proportions in in the BMI ≥25 kg/m² patient group than the BMI <25 kg/m² control group (Figure 4E and Table 5).

Figure 4 Association between BMI and oral microbiome diversity in patients with esophageal squamous cell carcinoma and controls. α-diversity analysis revealed significant differences in (A) Shannon entropy and (B) evenness between patient BMI subgroups. β-diversity two-dimensional principal coordinate analysis plots demonstrated distinct clustering between (C) the BMI ≥25 kg/m² vs. BMI <25 kg/m² patient groups and (D) the BMI ≥25 kg/m² patients vs. BMI ≥25 kg/m² control groups. (E) Taxonomic analysis. The number above the dot indicates the confidence interval. *, P value <0.05. BMI, body mass index.

TG

The TG <150 mg/dL patient group showed higher diversity than the TG ≥150 mg/dL control group (Figure 5A), and β-diversity analysis showed a clear clustering between these groups (Figure 5B). Clear clustering was also observed between the TG <150 mg/dL and TG ≥150 mg/dL patient groups, which showed marginally significant differences in α-diversity (Figure 5C). Taxa abundance analysis revealed that unclassified Campylobacter, unclassified Corynebacterium, and unclassified Leptotrichia showed significant abundance differences between the TG <150 mg/dL patient and TG ≥150 mg/dL control groups, whereas unclassified Streptococcus showed significant differences between the TG <150 mg/dL and TG ≥150 mg/dL patient groups (Figure 5D and Table 6).

Figure 5 Influence of triglyceride levels on oral microbiome composition in patients with esophageal squamous cell carcinoma and controls. (A) α-diversity (faith PD index) revealed significant differences between the TG <150 mg/dL patient and TG ≥150 mg/dL control groups, with a near-significant trend (P=0.07) compared to the TG ≥150 mg/dL patient group. β-diversity 2D PCoA plots visualized clustering between (B) the TG <150 mg/dL patient vs. TG ≥150 mg/dL control groups and (C) the TG <150 vs. TG ≥150 mg/dL patient groups. (D) Taxonomic analysis (forest plot). The number above the dot indicates the confidence interval. *, P value <0.05. Faith PD index, faith’s phylogenetic diversity index; PCoA, principal coordinate analysis; TG, triglycerides.

LDL

The LDL ≤100 mg/dL patient group showed higher diversity than the LDL ≤100 mg/dL control and LDL >100 mg/dL patient groups, although this was marginally significant (Figure 6A). However, β-diversity analysis showed a clear clustering (Figure 6B,6C). Analysis of taxa abundance showed that H. parainfluenzae showed lower abundance in the LDL ≤100 mg/dL patient group than in the LDL ≤100 mg/dL control group (Figure 6D). Additionally, V. dispar showed a higher abundance in the LDL >100 mg/dL patient group than in the LDL ≤100 mg/dL patient group (Figure 6D and Table 7).

Figure 6 Impact of LDL cholesterol on oral microbiome in patients with esophageal squamous cell carcinoma and controls. (A) α-diversity analysis in 4 groups. β-diversity (two-dimensional principal coordinate analysis plots) revealed compositional differences between (B) the LDL ≤100 mg/dL patient vs. LDL ≤100 mg/dL control groups, (C) and the LDL ≤100 vs. LDL >100 mg/dL patient groups. (D) Taxonomic analysis. The number above the dot indicates the confidence interval. *, P value <0.05. LDL, low-density lipoprotein.

Malignant disease

In the subgroup analysis based on the presence of malignant disease history, the “with malignant history patient” (WMP) group showed significantly higher diversity than the “non-malignant history patient” (NMP) group (Figure S1A). Clustering analysis also showed a clear clustering between the WMP and NMP groups (Figure S1B). Taxa analysis revealed that the abundance of unclassified Oribacterium was significantly higher in the WMP group than in the NMP group (Figure S1C and Table S1).

Tumor location

When analyzed by location [lower, middle (mid), and upper], the upper group showed a tendency toward lower diversity compared to other groups, although this was not statistically significant (Figure S2A,S2B). However, β-diversity analysis showed a clear clustering between the lower and middle groups and between the upper group and other groups (Figure S2C-S2F). Analysis of taxa abundance differences between groups revealed that Rothia dentocariosa, Bulleidia moorei, and P. pallens had a higher abundance in the lower group than in the middle group (Figure S2G). For P. pallens, the lower group also showed a higher abundance than the upper group (Figure S2G). Unclassified Granulicatella showed a significantly higher abundance in both the lower and middle groups than in the upper group (Figure S2G and Table S2).

Discussion

Key findings

This study explored the oral microbiome in patients with ESCC and healthy controls, focusing on clinical and metabolic factors. Although no significant differences in α- or β-diversity were observed between groups, taxa abundance analysis revealed that both groups were dominated by Bacillota, with a higher proportion of Bacteroidota in the patient group. Subgroup analyses highlighted distinct microbial patterns linked to clinical and metabolic profiles. Notably, Haemophilus parainfluenzae was lower abundant in ESCC than control among younger patients and those with low LDL levels, consistent with its role as a common commensal more abundant in healthy oral environments (18). Veillonella dispar was more abundant in older patients (vs. young control) and those with high LDL levels (vs. low LDL patients), suggesting a potential association with inflammatory or metabolically altered conditions commonly found in aging or dyslipidemia (19). Streptococcus infantis was less abundant in ESCC patients <65 years compared to controls <65 years, and in smoking patients compared to non-smoking patients. These findings suggest that reductions in this species may reflect early microbial imbalance associated with age- and smoking-related dysbiosis in ESCC. Together, these results highlight the influence of specific clinical factors on oral microbial composition in ESCC and support their potential as microbial biomarkers.

Strengths and limitations

This study has several limitations. First, the relatively small sample size may limit statistical power and the generalizability, as borderline associations in abundance analyses may not have reached statistical significance. Additionally, environmental factors, including lifestyle, oral hygiene, and dietary patterns, were not fully controlled during the selection of healthy controls and may have influenced the results. The cross-sectional study design further restricted the ability to assess dynamic changes in microbial communities over time or establish causal relationships. Furthermore, some taxa were not annotated to the species or strain level, reducing the precision and interpretability.

Comparison with similar research

Microbiome diversity showed no difference between healthy controls and ESCC patients aged over 65 years but increased in ECCC patients under 65 years. Generally, microbiome diversity declines in disease states (20). However, certain diseases exhibit increased diversity (21), reflecting a disruption of the normal microbial community and colonization by atypical microbes. Higher diversity does not necessarily indicate a healthy state but may instead point to community imbalance and the proliferation of pathogenic species (22). In our study, the elevated diversity in younger ESCC patients was accompanied by the depletion of commensal taxa such as Haemophilus parainfluenzae and Streptococcus infantis, suggesting that early microbial imbalance, rather than the presence of a healthy microbiota, may play a role in carcinogenesis in this subgroup. Meanwhile, diversity was notably lower in older ESCC patients, which may reflect age-related immune alterations that limit the growth of certain microbial taxa, or differences in microbiota composition driven by disease progression and host factors such as age and host genetics (23,24). These findings highlight the complex interplay between age, disease state, and microbial diversity in ESCC, underscoring the need for further investigation into the underlying mechanisms driving these differences.

Explanations of findings

ESCC patients with a history of smoking exhibited marked reductions in diversity, consistent with prior studies that show smoking impacts both malignancy and the microbiome (25-27). Carcinogenic substances from smoking, including specific nitrosamines, can directly affect the oral and esophageal mucosa. Additionally, smoking induces changes in the oral microbiome, resulting in significant alterations in microbial communities. This dual effect of smoking creates a pro-inflammatory microenvironment, potentially leading to malignant lesions and oral carcinogenesis (28). Thus, the observed diversity changes owing to smoking appear to be more pronounced in patients with esophageal cancer. Taxa-level analyses revealed significant differences in S. infantis, P. endodontalis, and unclassified Filifactor between smokers and non-smokers, with S. infantis being less abundant in the smoking patient group. These findings warrant further investigation into the roles of these taxa in ESCC.

ESCC patients with higher BMI were associated with increased microbiome diversity. Previous studies have reported decreased or unchanged α-diversity in groups with higher BMI (29,30). Because higher BMI in patients with esophageal cancer is also associated with comorbidities such as hypertension, diabetes, and coronary artery disease (31), further studies are needed to clarify the connections between BMI, esophageal cancer, and microbiome diversity. At the genus and species levels, Rothia was identified as a bacterium with altered abundance in the BMI subgroups. In adolescents, Rothia is commonly found in saliva and increases with higher BMI (32). It has also been identified in neonatal gut microbiota (33). In adults, a healthy esophageal microbiome can be categorized into three esotypes, one characterized by an increased abundance of Haemophilus and Rothia (23,34). Our findings are consistent with this, as Rothia abundance was reduced in ESCC patients with BMI ≥25 kg/m² compared to healthy controls of both BMI categories. Its decreased abundance in high-BMI ESCC patients suggests a potential role in disease progression, warranting study.

In our study, groups with higher TG levels generally showed reduced diversity, consistent with previous findings (35). A significant difference in diversity was observed between the TG ≥150 mg/dL control group and the TG <150 mg/dL esophageal cancer group. Unclassified Campylobacter exhibited higher abundance in the TG <150 mg/dL esophageal cancer group than in TG ≥150 mg/dL control group. Campylobacter is associated with inflammatory diseases such as moyamoya disease, cardiovascular disease, and inflammatory bowel disease (36-38). TG levels are linked to cardiovascular risk (39), the observed increase in unclassified Campylobacter abundance in the TG <150 esophageal cancer group compared to that in the TG ≥150 control group suggests a potential association between TG levels and this bacterium. In ESCC patients, unclassified Streptococcus was more abundant in those with TG ≥150 mg/dL than in those with TG <150 mg/dL, possibly reflecting cancer-related metabolic changes, and showing a pattern similar to previous observations in high-altitude populations where hypoxia-associated microbial fermentation was linked to lipid metabolism (40).

Previous study suggested a negative association between LDL levels and gut microbial diversity (35). Higher LDL levels in individuals with periodontal disease (41) and specific taxa such as Porphyromonas gingivalis and Fusobacterium are linked to LDL metabolism (42,43). H. parainfluenzae, a common oral commensal, is present in higher abundance in healthy individuals (44). In our study, H. parainfluenzae was more abundant in the LDL ≤100 mg/dL control group than in the LDL ≤100 mg/dL ESCC patient group, consistent with its association with healthier oral conditions. Additionally, Veillonella dispar was less abundant in the LDL ≤100 mg/dL ESCC group than in those with LDL >100 mg/dL, suggesting that lipid metabolism may influence the distribution of specific oral taxa. Although microbial diversity was only marginally higher in the LDL ≤100 mg/dL ESCC group, this trend may reflect compensatory shifts in microbial composition in response to cancer-related metabolic alterations.

Several solid cancers exhibit significant changes in the oral microbiome (45). Similarly, our study observed differences in the oral microbiome between patients with ESCC and a history of other cancers and those without a history of other cancers. These findings provide epidemiological evidence suggesting that microbial communities may be associated with the development of various cancers, regardless of their specific type. The proposed mechanisms include the creation of a pro-inflammatory microenvironment and the suppression of immune responses by the microbial community (46). Additionally, in our results, tumor location influenced microbiome composition, with taxa such as Rothia showing significant differences between tumor sites. β-diversity analysis revealed distinct clustering between the lower and middle groups, as well as between the upper group and the others. Among the taxa discussed in other sections, the abundance of Rothia was significantly higher in the lower group than in the middle group. Although the association between the location of esophageal cancer and the oral microbiome has not been well studied, these results highlight the potential role of Rothia, warranting further investigation.

Implications and actions needed

To overcome the limitations described above, future studies should include larger sample sizes, advanced microbial annotation techniques, and rigorous control of environmental variables. Despite these limitations, this study is a foundational investigation into the association between esophageal cancer and the oral microbiome as well as differences based on specific clinical factors, paving the way for further research to refine our understanding of microbial contributions to ESCC.

Conclusions

In this study, altered abundance of several oral microbiomes are associated with clinical and metabolic factors in esophageal cancer patients. Higher abundance of V. dispar was associated with higher LDL level and old age in esophageal cancer patients. Furthermore, lower abundance of H. parainfluenzae and S. infantis may be a potential biomarker of esophageal cancer among individuals younger than 65 years. Through larger-scale prospective studies and more refined analytical approaches, the mechanisms of esophageal cancer development and the role of oral microbiomes in early diagnosis could be elucidated.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-432/rc

Data Sharing Statement: Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-432/dss

Peer Review File: Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-432/prf

Funding: This study was funded by grants from the Korean Society of Gastrointestinal Cancer Research (Research Award 2021 to S.Y.N.), the National Research Foundation by Republic of Korea (NRF-2022R1A2C2013044 to S.Y.N.) and the National Research Council of Science & Technology (NST) by the Korea government (MSIT) (No. GTL24022-000 to S.N.).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-432/coif). S.N. reports the funding from the National Research Council of Science & Technology (NST) by the Korea government (MSIT) (No. GTL24022-000). S.Y.N. reports the funding from the Korean Society of Gastrointestinal Cancer Research (Research Award 2021) and the National Research Foundation by Republic of Korea (No. NRF-2022R1A2C2013044). The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are account for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study was approved by the ethics committees of Kosin University Gospel Hospital (No. KUGH 2021-07-004), Kyungpook National University Chilgok Hospital (No. KNUCH 2021-05-039-001), Dankook University Hospital (No. DKUH 2021-06-026), Yeungnam University Hospital (No. 2021-08-050), and Soonchunhyang University Hospital (No. 2021-07-047). All subjects provided written informed consent before the study.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Sung H, Ferlay J, Siegel RL, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin 2021;71:209-49. [Crossref] [PubMed]
2. Kim JH, Han KD, Lee JK, et al. Association between the National Cancer Screening Programme (NSCP) for gastric cancer and oesophageal cancer mortality. Br J Cancer 2020;123:480-6. [Crossref] [PubMed]
3. Park SY, Kim DJ. Esophageal Cancer in Korea: Epidemiology and Treatment Patterns. J Chest Surg 2021;54:454-9. [Crossref] [PubMed]
4. Lee SY. Comparison of Gastric Cancer Screening Strategies between Helicobacter pylori-infected, -eradicated, and -naive Individuals. Korean J Helicobacter Up Gastrointest Res 2022;22:313-6.
5. Kim YI, Choi IJ. Current Evidence for a Paradigm Shift in Gastric Cancer Prevention From Endoscopic Screening to Helicobacter pylori Eradication in Korea. J Gastric Cancer 2022;22:169-83. [Crossref] [PubMed]
6. Cullin N, Azevedo Antunes C, Straussman R, et al. Microbiome and cancer. Cancer Cell 2021;39:1317-41. [Crossref] [PubMed]
7. Xu X, He J, Xue J, et al. Oral cavity contains distinct niches with dynamic microbial communities. Environ Microbiol 2015;17:699-710. [Crossref] [PubMed]
8. Mukherjee C, Moyer CO, Steinkamp HM, et al. Acquisition of oral microbiota is driven by environment, not host genetics. Microbiome 2021;9:54. [Crossref] [PubMed]
9. Kolenbrander PE, Palmer RJ Jr, Periasamy S, et al. Oral multispecies biofilm development and the key role of cell-cell distance. Nat Rev Microbiol 2010;8:471-80. [Crossref] [PubMed]
10. Yang SF, Huang HD, Fan WL, et al. Compositional and functional variations of oral microbiota associated with the mutational changes in oral cancer. Oral Oncol 2018;77:1-8. [Crossref] [PubMed]
11. Zhao Q, Yang T, Yan Y, et al. Alterations of Oral Microbiota in Chinese Patients With Esophageal Cancer. Front Cell Infect Microbiol 2020;10:541144. [Crossref] [PubMed]
12. Gupta VK, Paul S, Dutta C. Geography, Ethnicity or Subsistence-Specific Variations in Human Microbiome Composition and Diversity. Front Microbiol 2017;8:1162. [Crossref] [PubMed]
13. Jia G, Zhi A, Lai PFH, et al. The oral microbiota - a mechanistic role for systemic diseases. Br Dent J 2018;224:447-55. [Crossref] [PubMed]
14. Burcham ZM, Garneau NL, Comstock SS, et al. Patterns of Oral Microbiota Diversity in Adults and Children: A Crowdsourced Population Study. Sci Rep 2020;10:2133. [Crossref] [PubMed]
15. Lee YH, Chung SW, Auh QS, et al. Progress in Oral Microbiome Related to Oral and Systemic Diseases: An Update. Diagnostics (Basel) 2021;11:1283. [Crossref] [PubMed]
16. Bolyen E, Rideout JR, Dillon MR, et al. Reproducible, interactive, scalable and extensible microbiome data science using QIIME 2. Nat Biotechnol 2019;37:852-7. [Crossref] [PubMed]
17. Parks DH, Beiko RG. Identifying biologically relevant differences between metagenomic communities. Bioinformatics 2010;26:715-21. [Crossref] [PubMed]
18. Perera D, McLean A, Morillo-López V, et al. Mechanisms underlying interactions between two abundant oral commensal bacteria. ISME J 2022;16:948-57. [Crossref] [PubMed]
19. Luangphiphat W, Prombutara P, Muangsillapasart V, et al. Exploring of gut microbiota features in dyslipidemia and chronic coronary syndrome patients undergoing coronary angiography. Front Microbiol 2024;15:1384146. [Crossref] [PubMed]
20. Pickard JM, Zeng MY, Caruso R, et al. Gut microbiota: Role in pathogen colonization, immune responses, and inflammatory disease. Immunol Rev 2017;279:70-89. [Crossref] [PubMed]
21. Zhang L, Liu Y, Zheng HJ, et al. The Oral Microbiota May Have Influence on Oral Cancer. Front Cell Infect Microbiol 2019;9:476. [Crossref] [PubMed]
22. Su SC, Chang LC, Huang HD, et al. Oral microbial dysbiosis and its performance in predicting oral cancer. Carcinogenesis 2021;42:127-35. [Crossref] [PubMed]
23. Deshpande NP, Riordan SM, Castaño-Rodríguez N, et al. Signatures within the esophageal microbiome are associated with host genetics, age, and disease. Microbiome 2018;6:227. [Crossref] [PubMed]
24. Yang W, Chen CH, Jia M, et al. Tumor-Associated Microbiota in Esophageal Squamous Cell Carcinoma. Front Cell Dev Biol 2021;9:641270. [Crossref] [PubMed]
25. Thomas AM, Gleber-Netto FO, Fernandes GR, et al. Alcohol and tobacco consumption affects bacterial richness in oral cavity mucosa biofilms. BMC Microbiol 2014;14:250. [Crossref] [PubMed]
26. Yang Y, Zheng W, Cai QY, et al. Cigarette smoking and oral microbiota in low-income and African-American populations. J Epidemiol Community Health 2019;73:1108-15. [Crossref] [PubMed]
27. Li Z, Liu Y, Dou L, et al. The effects of smoking and drinking on the oral and esophageal microbiota of healthy people. Ann Transl Med 2021;9:1244. [Crossref] [PubMed]
28. Kadam S, Vandana M, Patwardhan S, et al. Looking beyond the smokescreen: can the oral microbiome be a tool or target in the management of tobacco-associated oral cancer? Ecancermedicalscience 2021;15:1179. [Crossref] [PubMed]
29. Yang Y, Cai Q, Zheng W, et al. Oral microbiome and obesity in a large study of low-income and African-American populations. J Oral Microbiol 2019;11:1650597. [Crossref] [PubMed]
30. Shaalan A, Lee S, Feart C, et al. Alterations in the Oral Microbiome Associated With Diabetes, Overweight, and Dietary Components. Front Nutr 2022;9:914715. [Crossref] [PubMed]
31. Ren C, Cai XY, Qiu MZ, et al. Impact of body mass index on survival of esophageal squamous carcinoma patients in southern China. J Thorac Dis 2015;7:337-45. [Crossref] [PubMed]
32. Sharara SH, Cleaver LM, Saloom H, et al. Salivary bacterial community profile in normal-weight and obese adolescent patients prior to orthodontic treatment with fixed appliances. Orthod Craniofac Res 2022;25:569-75. [Crossref] [PubMed]
33. Geng J, Ni Q, Sun W, et al. The links between gut microbiota and obesity and obesity related diseases. Biomed Pharmacother 2022;147:112678. [Crossref] [PubMed]
34. Nesteruk K, Spaander MCW, Leeuwenburgh I, et al. Achalasia and associated esophageal cancer risk: What lessons can we learn from the molecular analysis of Barrett's-associated adenocarcinoma? Biochim Biophys Acta Rev Cancer 2019;1872:188291. [Crossref] [PubMed]
35. Matey-Hernandez ML, Williams FMK, Potter T, et al. Genetic and microbiome influence on lipid metabolism and dyslipidemia. Physiol Genomics 2018;50:117-26. [Crossref] [PubMed]
36. Zhang L. Oral Campylobacter species: Initiators of a subgroup of inflammatory bowel disease? World J Gastroenterol 2015;21:9239-44. [Crossref] [PubMed]
37. Schulz S, Hofmann B, Grollmitz J, et al. Campylobacter Species of the Oral Microbiota as Prognostic Factor for Cardiovascular Outcome after Coronary Artery Bypass Grafting Surgery. Biomedicines 2022;10:1801. [Crossref] [PubMed]
38. Takayanagi K, Kanamori F, Ishii K, et al. Higher abundance of Campylobacter in the oral microbiome of Japanese patients with moyamoya disease. Sci Rep 2023;13:18545. [Crossref] [PubMed]
39. Crea F. An update on triglyceride-rich lipoproteins and their remnants in atherosclerotic cardiovascular disease. Eur Heart J 2021;42:4777-80. [Crossref] [PubMed]
40. Zhao L, Wang H, Gao Y, et al. Characteristics of oral microbiota in plateau and plain youth-positive correlations between blood lipid level, metabolism and specific microflora in the plateau group. Front Cell Infect Microbiol 2022;12:952579. [Crossref] [PubMed]
41. Nepomuceno R, Pigossi SC, Finoti LS, et al. Serum lipid levels in patients with periodontal disease: A meta-analysis and meta-regression. J Clin Periodontol 2017;44:1192-207. [Crossref] [PubMed]
42. Bengtsson T, Karlsson H, Gunnarsson P, et al. The periodontal pathogen Porphyromonas gingivalis cleaves apoB-100 and increases the expression of apoM in LDL in whole blood leading to cell proliferation. J Intern Med 2008;263:558-71. [Crossref] [PubMed]
43. Koren O, Spor A, Felin J, et al. Human oral, gut, and plaque microbiota in patients with atherosclerosis. Proc Natl Acad Sci U S A 2011;108:4592-8. [Crossref] [PubMed]
44. Bik EM, Long CD, Armitage GC, et al. Bacterial diversity in the oral cavity of 10 healthy individuals. ISME J 2010;4:962-74. [Crossref] [PubMed]
45. Feng K, Ren F, Wang X. Association between oral microbiome and seven types of cancers in East Asian population: a two-sample Mendelian randomization analysis. Front Mol Biosci 2023;10:1327893. [Crossref] [PubMed]
46. Irfan M, Delgado RZR, Frias-Lopez J. The Oral Microbiome and Cancer. Front Immunol 2020;11:591088. [Crossref] [PubMed]