Skip to content
PaperFren

Microbiome

Mendelian randomisation associates specific gut taxa with several cancers

Long Y · BMC medicine · 2023

Metadata + PaperFren explanation · cc by · source: Europe PMC

Genetic instruments for gut microbiota taxa showed causal associations with breast, lung, colorectal, prostate, gastric, and head/neck cancers—sometimes in opposing directions for the same genus.

Study at a glance

Design
Mendelian randomisation — Two-sample MR of MiBioGen gut taxa instruments against eight cancer GWAS summary sets
N
211 taxa instruments; cancer GWAS sample sizes vary by malignancy — no single primary N
Population
Gut microbiome and cancer GWAS summary statistics (IEU OpenGWAS/consortia)
Outcome
Causal microbiome–cancer associations (stringent and sensitivity analyses)

Structured fields used in claim comparison tables when every cited study has a complete layer.

Key findings

Eleven stringent microbiome-to-cancer causal relationships emerged, including Actinobacteria/Bifidobacteriales with higher breast cancer odds, Tyzzerella3 with higher lung adenocarcinoma but lower colorectal cancer risk, Ruminococcustorques group with lower prostate cancer risk, and Peptostreptococcaceae with higher gastric cancer risk. Reverse MR supported bidirectional Tyzzerella3–lung adenocarcinoma effects. Additional associations appeared at relaxed thresholds across datasets.

Methodology

Authors harmonized MiBioGen gut microbiome GWAS instruments (211 taxa) with cancer GWAS summary statistics for eight malignancies from IEU Open GWAS and consortia. They ran two-sample MR (IVW, MR-Egger, weighted median, MR-PRESSO) with Bonferroni thresholds, sensitivity analyses, and reverse MR to test directionality.

Limitations

MR assumes valid genetic instruments and European-ancestry GWAS limits generalizability; 16S-based taxa lack strain resolution, and mechanistic pathways from taxa to cancer remain unproven experimentally.

How this study connects

Role on claims

Each row is a claim on a concept or method page where this paper supports, challenges, or qualifies the statement. Roles are hand-checked — not a model guess.

  • SupportsMicrobiomeconcept

    Genetic instruments for gut taxa link a small set of microbiome–cancer associations under Mendelian randomisation.

    Using Mendelian randomisation on gut-taxa genetic instruments, eleven stringent microbiome-to-cancer associations were reported (for example higher breast-cancer odds with Actinobacteria/Bifidobacteriales signals; Tyzzerella3 linked with higher lung adenocarcinoma but lower colorectal cancer risk).

    Evidence for the claim as stated.

  • QualifiesMicrobiomeconcept

    Before type 1 diabetes autoimmunity, Bacteroides dorei can dominate children’s gut communities.

    In children who later developed type 1 diabetes autoimmunity, Bacteroides dorei became dominant in the gut community before persistent autoantibodies (mean diagnosis age about 16.8 months).

    Scope note — different disease and method — cancer MR in adults, not pediatric T1D timing

    Limits the claim's scope: a different population, assay, or outcome.

  • QualifiesMicrobiomeconcept

    In middle-aged men, gut composition associates with metabolites and metabolic-syndrome traits.

    In middle-aged men, gut microbiota composition was associated with plasma metabolites and metabolic-syndrome–related traits, placing the microbiome alongside genetics and lifestyle as a correlated host factor.

    Scope note — different outcome — cancer odds via MR, not metabolic metabolites

    Limits the claim's scope: a different population, assay, or outcome.

  • SupportsMicrobiomeconcept

    Cancer MR, pediatric T1D timing, and adult metabolic-trait associations all involve gut microbiota, but they answer different disease and design questions. Treating them as one interchangeable “gut microbiome disease effect” collapses distinct estimands.

    Evidence for the claim as stated.

  • SupportsMicrobiomeconcept

    Human observational and MR studies do not automatically generalise to bee antibiotic/pesticide disruptions or to enzyme-focused colonization experiments. Animal and insect systems test different host–microbe questions; they limit how far a human association travels rather than falsifying it.

    Evidence for the claim as stated.

  • QualifiesGWASconcept

    Polygenic score accuracy can differ a lot among groups that share a broad ancestry label.

    Polygenic score accuracy can differ substantially among groups that share a broad ancestry label; performance tracks ascertainment and SES-related structure, not ancestry labels alone.

    Scope note — different question — MR disease associations, not PGS prediction accuracy

    Limits the claim's scope: a different population, assay, or outcome.

  • QualifiesGWASconcept

    Liver eQTL maps link common variants to expression of disease-relevant metabolic genes.

    Human liver eQTL mapping links common variants to expression of disease-relevant metabolic pathway genes, showing one route from GWAS-scale variation to molecular intermediates.

    Scope note — different endpoint — cancer odds via MR, not liver expression QTLs

    Limits the claim's scope: a different population, assay, or outcome.

  • SupportsGWASconcept

    Mendelian randomisation on gut taxa reports a limited set of microbiome–cancer links.

    Using genetic instruments for gut microbiota taxa, Mendelian randomisation reported eleven stringent microbiome-to-cancer associations (directions sometimes opposing across cancers for related taxa)—an MR application that depends on GWAS-quality instruments and their assumptions.

    Evidence for the claim as stated.

  • SupportsGWASconcept

    Human PGS accuracy, liver eQTLs, rare-variant synthetic-signal theory, livestock/plant GWAS, and microbiome–cancer MR all live under a GWAS umbrella, but they answer different estimands. Pooling them as one “GWAS finding” erases species, trait, and method limits.

    Evidence for the claim as stated.

  • SupportsGWASconcept

    A genome-wide significant hit is not the same claim as a portable polygenic score or an MR causal estimate. Synthetic-signal work limits naive readings of peaks; PGS work limits naive portability; MR adds instrument assumptions on top of association.

    Evidence for the claim as stated.

Open questions

Tensions this paper is part of

From concept pages' “where studies disagree.” Disagreement means the same question; scope means different assays, populations, or outcomes.

History

When this study was placed

Dated entries from the concept change log — when this paper was added or removed as support, challenge, or qualifier on a claim.

  1. 2026-09-14

    Placed as a scope qualifier on GWAS

    Human liver eQTL mapping links common variants to expression of disease-relevant metabolic pathway genes, showing one route from GWAS-scale variation to molecular intermediates.

  2. 2026-09-14

    Placed as a scope qualifier on GWAS

    Polygenic score accuracy can differ substantially among groups that share a broad ancestry label; performance tracks ascertainment and SES-related structure, not ancestry labels alone.

  3. 2026-09-14

    Placed as supporting evidence on GWAS

    Using genetic instruments for gut microbiota taxa, Mendelian randomisation reported eleven stringent microbiome-to-cancer associations (directions sometimes opposing across cancers for related taxa)—an MR application that depends on GWAS-quality instruments and their assumptions.

  4. 2026-09-14

    Placed as supporting evidence on GWAS

    Human PGS accuracy, liver eQTLs, rare-variant synthetic-signal theory, livestock/plant GWAS, and microbiome–cancer MR all live under a GWAS umbrella, but they answer different estimands. Pooling them as one “GWAS finding” erases species, trait, and method limits.

  5. 2026-09-14

    Placed as supporting evidence on GWAS

    A genome-wide significant hit is not the same claim as a portable polygenic score or an MR causal estimate. Synthetic-signal work limits naive readings of peaks; PGS work limits naive portability; MR adds instrument assumptions on top of association.

  6. 2026-09-14

    Placed as supporting evidence on Microbiome

    Using Mendelian randomisation on gut-taxa genetic instruments, eleven stringent microbiome-to-cancer associations were reported (for example higher breast-cancer odds with Actinobacteria/Bifidobacteriales signals; Tyzzerella3 linked with higher lung adenocarcinoma but lower colorectal cancer risk).

  7. 2026-09-14

    Placed as a scope qualifier on Microbiome

    In children who later developed type 1 diabetes autoimmunity, Bacteroides dorei became dominant in the gut community before persistent autoantibodies (mean diagnosis age about 16.8 months).

  8. 2026-09-14

    Placed as a scope qualifier on Microbiome

    In middle-aged men, gut microbiota composition was associated with plasma metabolites and metabolic-syndrome–related traits, placing the microbiome alongside genetics and lifestyle as a correlated host factor.

  9. 2026-09-14

    Placed as supporting evidence on Microbiome

    Cancer MR, pediatric T1D timing, and adult metabolic-trait associations all involve gut microbiota, but they answer different disease and design questions. Treating them as one interchangeable “gut microbiome disease effect” collapses distinct estimands.

  10. 2026-09-14

    Placed as supporting evidence on Microbiome

    Human observational and MR studies do not automatically generalise to bee antibiotic/pesticide disruptions or to enzyme-focused colonization experiments. Animal and insect systems test different host–microbe questions; they limit how far a human association travels rather than falsifying it.

Related papers in this topic

Same topic cluster — not a recommendation engine.