CoChat

High-resolution bacterial 16S rRNA gene profile meta-analysis and biofilm status reveal common colorectal cancer consortia

Open access

The DOI matches its Crossref record; 73 of the 74 references it cites were found in a registry or an index.

Abstract

Colorectal cancer (CRC) remains the third most common cancer worldwide, with a growing incidence among young adults. Multiple studies have presented associations between the gut microbiome and CRC, suggesting a link with cancer risk. Although CRC microbiome studies continue to profile larger patient cohorts with increasingly economical and rapid DNA sequencing platforms, few common associations with CRC have been identified, in part due to limitations in taxonomic resolution and differences in analysis methodologies. Complementing these taxonomic studies is the newly recognized phenomenon that bacterial organization into biofilm structures in the mucus layer of the gut is a consistent feature of right-sided (proximal), but not left-sided (distal) colorectal cancer. In the present study, we performed 16S rRNA gene amplicon sequencing and biofilm quantification in a new cohort of patients from Malaysia, followed by a meta-analysis of eleven additional publicly available data sets on stool and tissue-based CRC microbiota using Resphera Insight, a high-resolution analytical tool for species-level characterization. Results from the Malaysian cohort and the expanded meta-analysis confirm that CRC tissues are enriched for invasive biofilms (particularly on right-sided tumors), a symbiont with capacity for tumorigenesis ( Bacteroides fragilis ), and oral pathogens including Fusobacterium nucleatum , Parvimonas micra , and Peptostreptococcus stomatis . Considered in aggregate, species from the Human Oral Microbiome Database are highly enriched in CRC. Although no detected microbial feature was universally present, their substantial overlap and combined prevalence supports a role for the gut microbiota in a significant percentage (>80%) of CRC cases.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1038/s41522-017-0040-3

Cites

73 of 73 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 Meta-analysis of biofilm prevalence in CRC. a Microbial biofilms in Carnoy’s-fixed tissue were detected by FISH with the universal 16S rRNA gene probe Eub338 (stained in red) and the nucleic acid stain DAPI (blue). White brackets denote the mucus layer, visible via autofluorescence of the tissue. Left panel: a biofilm-positive paired normal tissue from the right colon of a CRC patient with abundant bacteria adjacent to the epithelium (arrow). Right panel: a biofilm-negative paired normal tissue from the left colon of a CRC patient with minimal bacteria (arrow) present only in the outer edge of the mucus. Scale bars represent 100 μm. b The locations of tumors from USA, MAL1, and MAL2 are overlaid on a diagram of the colon. c Percentage of biofilm-positive right and left tumors for the three cohorts. All statistics shown are Fisher’s exact tests. d Total percentage of biofilm-positive right- and left-sided tumors from all three cohorts is depicted with a Fisher’s exact test p-value. e Left-sided and right-sided tumors separated according to tumor stage and biofilm status. Multivariate logistic regression analysis was performed after controlling for tumor side. f, g Biofilm-positive samples were stained with DAPI (blue) and probes against four bacterial membership groups: Fusobacterium (yellow), Bacteroidetes (green), Lachnospiraceae (red), and Proteobacteria (magenta). Scale bars in large images represent 100 μm; scale bars in smaller, inset images represent 5 μm. f Representative polymicrobial biofilm from a tumor (left panel) and its paired normal tissue (right panel), with blooms of Fusobacterium in yellow visible only in the tumor. g Representative Proteobacteria-dominant tumor (left panel) and its paired normal tissue (right panel)
Fig. 1 Meta-analysis of biofilm prevalence in CRC. a Microbial biofilms in Carnoy’s-fixed tissue were detected by FISH with the universal 16S rRNA gene probe Eub338 (stained in red) and the nucleic acid stain DAPI (blue). White brackets denote the mucus layer, visible via autofluorescence of the tissue. Left panel: a biofilm-positive paired normal tissue from the right colon of a CRC patient with abundant bacteria adjacent to the epithelium (arrow). Right panel: a biofilm-negative paired normal tissue from the left colon of a CRC patient with minimal bacteria (arrow) present only in the outer edge of the mucus. Scale bars represent 100 μm. b The locations of tumors from USA, MAL1, and MAL2 are overlaid on a diagram of the colon. c Percentage of biofilm-positive right and left tumors for the three cohorts. All statistics shown are Fisher’s exact tests. d Total percentage of biofilm-positive right- and left-sided tumors from all three cohorts is depicted with a Fisher’s exact test p-value. e Left-sided and right-sided tumors separated according to tumor stage and biofilm status. Multivariate logistic regression analysis was performed after controlling for tumor side. f, g Biofilm-positive samples were stained with DAPI (blue) and probes against four bacterial membership groups: Fusobacterium (yellow), Bacteroidetes (green), Lachnospiraceae (red), and Proteobacteria (magenta). Scale bars in large images represent 100 μm; scale bars in smaller, inset images represent 5 μm. f Representative polymicrobial biofilm from a tumor (left panel) and its paired normal tissue (right panel), with blooms of Fusobacterium in yellow visible only in the tumor. g Representative Proteobacteria-dominant tumor (left panel) and its paired normal tissue (right panel)CC BY · PubMed Central
Fig. 2 Microbial associations with biofilm status. Frozen tissue from USA, MAL1, and MAL2 cohorts were characterized by 16S rRNA amplicon sequencing. a Top panel: biofilm positive CRC and normal flanking tissues demonstrated several functional shifts in the bacterial composition, including increases in gene content associated with cytoskeletal proteins, peptidoglycan biosynthesis, sporulation, and flagellar assembly based on PICRUSt analysis. Bottom panel: the bacterial families Veillonellaceae, Coriobacteriaceae, and Lachnospiraceae were enriched in biofilm-positive samples, while Sphingomonadaceae was enriched in biofilm-negative samples. Random-effects models with 95% CI above or below 0 (red diamonds) were considered statistically significant. Hedge’s g difference statistic is shown on the X axes. The fixed effects model assumes there exists a single effect size shared by all included studies, while the random effects model allows for variation in the effect size from study to study. Heterogeneity analysis includes estimates of I2 (percentage of variation reflecting true heterogeneity), τ2 (random-effects between study variance), and p-value from Cochran’s Q test for heterogeneity. b The functional alterations due to biofilm status were linked to multiple, differentially abundant families (f) and genera (g) in the biofilms
Fig. 2 Microbial associations with biofilm status. Frozen tissue from USA, MAL1, and MAL2 cohorts were characterized by 16S rRNA amplicon sequencing. a Top panel: biofilm positive CRC and normal flanking tissues demonstrated several functional shifts in the bacterial composition, including increases in gene content associated with cytoskeletal proteins, peptidoglycan biosynthesis, sporulation, and flagellar assembly based on PICRUSt analysis. Bottom panel: the bacterial families Veillonellaceae, Coriobacteriaceae, and Lachnospiraceae were enriched in biofilm-positive samples, while Sphingomonadaceae was enriched in biofilm-negative samples. Random-effects models with 95% CI above or below 0 (red diamonds) were considered statistically significant. Hedge’s g difference statistic is shown on the X axes. The fixed effects model assumes there exists a single effect size shared by all included studies, while the random effects model allows for variation in the effect size from study to study. Heterogeneity analysis includes estimates of I2 (percentage of variation reflecting true heterogeneity), τ2 (random-effects between study variance), and p-value from Cochran’s Q test for heterogeneity. b The functional alterations due to biofilm status were linked to multiple, differentially abundant families (f) and genera (g) in the biofilmsCC BY · PubMed Central
Fig. 3 Associations of B. fragilis and human oral microbiota with CRC status for USA, MAL1, and MAL2 cohorts. Microbiome profiling with 16S rRNA gene sequencing was applied to tumor (CRC) and paired normal tissues (Normal) from CRC patients for all three cohorts, as well as healthy biopsies (Healthy Bx) for USA and MAL1. a Bar charts of microbial sequence relative abundance for B. fragilis and bacteria from the Human Oral Microbiome Database (HOMD) from the three 16S rRNA amplicon data sets: USA (top panel), MAL1 (middle panel), and MAL2 (bottom panel). Each vertical bar represents an individual patient. b B. fragilis as well as the oral pathogens F. nucleatum, P. micra, P. stomatis, G. morbillorum, and the overall consortia associated with the Human Oral Microbiome Database (HOMD) were all found to be significantly enriched in tumor specimens compared to biopsies from healthy patients without cancer (top panel), and compared to normal flanking tissue (bottom panel). Random-effects models with 95% CI above or below 0 (red diamonds) were considered statistically significant. Hedge’s g difference statistic is shown on the X axes. The fixed effects model assumes there exists a single effect size shared by all included studies, while the random effects model allows for variation in the effect size from study to study. Heterogeneity analysis includes estimates of I2 (percentage of variation reflecting true heterogeneity), τ2 (random-effects between study variance), and p-value from Cochran’s Q test for heterogeneity
Fig. 3 Associations of B. fragilis and human oral microbiota with CRC status for USA, MAL1, and MAL2 cohorts. Microbiome profiling with 16S rRNA gene sequencing was applied to tumor (CRC) and paired normal tissues (Normal) from CRC patients for all three cohorts, as well as healthy biopsies (Healthy Bx) for USA and MAL1. a Bar charts of microbial sequence relative abundance for B. fragilis and bacteria from the Human Oral Microbiome Database (HOMD) from the three 16S rRNA amplicon data sets: USA (top panel), MAL1 (middle panel), and MAL2 (bottom panel). Each vertical bar represents an individual patient. b B. fragilis as well as the oral pathogens F. nucleatum, P. micra, P. stomatis, G. morbillorum, and the overall consortia associated with the Human Oral Microbiome Database (HOMD) were all found to be significantly enriched in tumor specimens compared to biopsies from healthy patients without cancer (top panel), and compared to normal flanking tissue (bottom panel). Random-effects models with 95% CI above or below 0 (red diamonds) were considered statistically significant. Hedge’s g difference statistic is shown on the X axes. The fixed effects model assumes there exists a single effect size shared by all included studies, while the random effects model allows for variation in the effect size from study to study. Heterogeneity analysis includes estimates of I2 (percentage of variation reflecting true heterogeneity), τ2 (random-effects between study variance), and p-value from Cochran’s Q test for heterogeneityCC BY · PubMed Central
Fig. 4 Expanded differential relative abundance meta-analysis supports species-level findings. A meta-analysis of a CRC tissue or stool vs. healthy biopsy or stool and b CRC vs. flanking normal tissues supports the enrichment of B. fragilis and oral microbes in CRC. Random-effects models with 95% CI above or below 0 (red diamonds) were considered statistically significant. Hedge’s g difference statistic is shown on the X axes. The fixed effects model assumes there exists a single effect size shared by all included studies, while the random effects model allows for variation in the effect size from study to study. Heterogeneity analysis includes estimates of I2 (percentage of variation reflecting true heterogeneity), τ2 (random-effects between study variance), and p-value from Cochran’s Q test for heterogeneity
Fig. 4 Expanded differential relative abundance meta-analysis supports species-level findings. A meta-analysis of a CRC tissue or stool vs. healthy biopsy or stool and b CRC vs. flanking normal tissues supports the enrichment of B. fragilis and oral microbes in CRC. Random-effects models with 95% CI above or below 0 (red diamonds) were considered statistically significant. Hedge’s g difference statistic is shown on the X axes. The fixed effects model assumes there exists a single effect size shared by all included studies, while the random effects model allows for variation in the effect size from study to study. Heterogeneity analysis includes estimates of I2 (percentage of variation reflecting true heterogeneity), τ2 (random-effects between study variance), and p-value from Cochran’s Q test for heterogeneityCC BY · PubMed Central
Fig. 5 Not all fusobacterial species are associated with CRC. a In addition to F. nucleatum, we detected significant enrichment of F. necrophorum, F. periodonticum, and L. trevansanii in CRC tissue or stool compared to healthy biopsies or stool. Hedge’s g difference statistic is shown on the X axes. The F. periodonticum forest plot includes ambiguous assignments (see Methods). b Other fusobacterial species were not differentially abundant between CRC and healthy patients. The fixed effects model assumes there exists a single effect size shared by all included studies, while the random effects model allows for variation in the effect size from study to study. Heterogeneity analysis includes estimates of I2 (percentage of variation reflecting true heterogeneity), τ2 (random-effects between study variance), and p-value from Cochran’s Q test for heterogeneity. c Bar chart of all detected species within the fusobacterial phylum and their total sequence relative abundances in the MAL1 cohort. Each vertical bar represents an individual patient. Several healthy Malaysian patients in the MAL1 cohort harbored high levels of fusobacterial species not associated with CRC
Fig. 5 Not all fusobacterial species are associated with CRC. a In addition to F. nucleatum, we detected significant enrichment of F. necrophorum, F. periodonticum, and L. trevansanii in CRC tissue or stool compared to healthy biopsies or stool. Hedge’s g difference statistic is shown on the X axes. The F. periodonticum forest plot includes ambiguous assignments (see Methods). b Other fusobacterial species were not differentially abundant between CRC and healthy patients. The fixed effects model assumes there exists a single effect size shared by all included studies, while the random effects model allows for variation in the effect size from study to study. Heterogeneity analysis includes estimates of I2 (percentage of variation reflecting true heterogeneity), τ2 (random-effects between study variance), and p-value from Cochran’s Q test for heterogeneity. c Bar chart of all detected species within the fusobacterial phylum and their total sequence relative abundances in the MAL1 cohort. Each vertical bar represents an individual patient. Several healthy Malaysian patients in the MAL1 cohort harbored high levels of fusobacterial species not associated with CRCCC BY · PubMed Central
Fig. 6 Combined prevalence of CRC-associated microbial features in USA, MAL1, and MAL2 cohorts. Positive status for B. fragilis and the human oral microbes from the HOMD database were defined as a relative abundance >2 and >10% of all 16S rRNA gene sequences, respectively. a Percentage of healthy biopsies, paired normal, and tumor tissues that were positive for each microbial feature (biofilms, HOMD, and B. fragilis). Statistics shown are Fisher’s exact test. *p < 0.01, ***p < 0.001. b Percentage and overlap of healthy biopsies, paired normal, and tumor tissues harboring one, two, three or no microbial features
Fig. 6 Combined prevalence of CRC-associated microbial features in USA, MAL1, and MAL2 cohorts. Positive status for B. fragilis and the human oral microbes from the HOMD database were defined as a relative abundance >2 and >10% of all 16S rRNA gene sequences, respectively. a Percentage of healthy biopsies, paired normal, and tumor tissues that were positive for each microbial feature (biofilms, HOMD, and B. fragilis). Statistics shown are Fisher’s exact test. *p < 0.01, ***p < 0.001. b Percentage and overlap of healthy biopsies, paired normal, and tumor tissues harboring one, two, three or no microbial featuresCC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 73 references

Cited by

See all 335

Metadata

Authors
Drewes, J. L., White, J. R., Dejea, C. M., Fathi, P., Iyadorai, T., Vadivelu, J., Roslani, A. C., Wick, E. C., Mongodin, E. F., Loke, M. F., Thulasi, K., Gan, H. M., Goh, K. L., Chong, H. Y., Kumar, S., Wanyiri, J. W., & Sears, C. L.
Affiliations
  • Johns Hopkins University (US)
  • Resphera Biosciences (US)
  • University of Malaya (MY)
  • University of Maryland, Baltimore (US)
  • Monash University Malaysia (MY)
Corresponding author
Sears, Cynthia L.
Venue
npj Biofilms and Microbiomes
Published
2017-11-29
Volume
3
Issue
1
Pages
34-34
ISSN
2055-5008 (ISSN-L)
OpenAlex
W2770639073
Cited by
335
References
74
Citations by year
2026: 33 · 2025: 42 · 2024: 34 · 2023: 44 · 2022: 35 · 2021: 54 · 2020: 41 · 2019: 31 · 2018: 20 · 2017: 1
FWCI
10.39
Type
journal article
Language
en
Open access
Licence
Locations
  • npj Biofilms and Microbiomes (publishedVersion, cc-by)
  • Europe PMC (PubMed Central) (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
  • Own your potential (DEAKIN) (submittedVersion)
  • Deakin Research Online (Deakin University) (submittedVersion)
  • eScholarship (California Digital Library) (submittedVersion, cc-by)
  • Figshare (submittedVersion, cc-by)
Funders
  • Cancer Research Institute (CA006973)
  • Universiti Malaya (RP016A-13HTM)
  • National Institutes of Health (1s10od016374-01, CA006973, CA062924, DK089502)
  • Sidney Kimmel Comprehensive Cancer Center (P30 CA006973, CA006973)
  • National Cancer Institute (R01CA196845, R01CA151393, P50CA062924, P30CA006973)
  • National Institute of Diabetes and Digestive and Kidney Diseases (P30DK089502)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Gut microbiota and health
  • Health Sciences › Dentistry › Periodontics › Oral microbiology and periodontitis research
  • Health Sciences › Medicine › Infectious Diseases › Clostridium difficile and Clostridium perfringens research
Keywords
colorectal cancer, gut microbiota, bacterial biofilms, 16S rRNA gene profiling, Bacteroides fragilis, Fusobacterium nucleatum, Parvimonas micra, Peptostreptococcus stomatis, oral pathogens, right-sided colorectal cancer, mucus layer

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: High-resolution bacterial 16S rRNA gene profile meta-analysis and biofilm status reveal common colorectal cancer consortia. https://app.cochat.ai/papers/10.1038/s41522-017-0040-3
The paper
Drewes, J. L., White, J. R., Dejea, C. M., Fathi, P., Iyadorai, T., Vadivelu, J., Roslani, A. C., Wick, E. C., Mongodin, E. F., Loke, M. F., Thulasi, K., Gan, H. M., Goh, K. L., Chong, H. Y., Kumar, S., Wanyiri, J. W., & Sears, C. L. (2017). High-resolution bacterial 16S rRNA gene profile meta-analysis and biofilm status reveal common colorectal cancer consortia. npj Biofilms and Microbiomes. https://doi.org/10.1038/s41522-017-0040-3

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.