In plain terms

The claim being assessed: Excess adiposity before age 50 causally increases the risk of early-onset colorectal cancer.

Where the evidence stands: Γ — Plausible mechanism. There is a believable biological story and partial support, but the causal step has not been demonstrated.

What this mechanism does: This makes the process worse or keeps it going, but something else starts it.

Strongest evidence for it: The preregistered primary-subclaim rule returns SC2 Γ for the precise age-defined causal claim. This is a bounded assessment of direct early-onset causal identification, not a refutation of the prospective association or the broader adiposity–colorectal-cancer causal literature. The cohort dose-response and genetic-natural-experiment equivalents were evaluated; their age/timing and system scopes do not silently become the exact registered outcome.

What decisive evidence would look like

In plain terms, researchers would change the proposed cause, confirm it changed as intended, and measure a meaningful health outcome while checking other explanations. The result should be able to support or count against the claim.

Relative explanatory fit: Alternative ranked ahead

What this review means

Excess adiposity is a credible contributor to early-onset colorectal cancer, but its precise causal effect before50 and its share of the incidence rise remain less securely identified than the broader colorectal-cancer link.

Implications for health and treatment: The preregistered primary-subclaim rule returns SC2 Γ for the precise age-defined causal claim. This is a bounded assessment of direct early-onset causal identification, not a refutation of the prospective association or the broader adiposity–colorectal-cancer causal literature. The cohort dose-response and genetic-natural-experiment equivalents were evaluated; their age/timing and system scopes do not silently become the exact registered outcome.

Limits for individual care: 58 primary reports with 3 selected partial-body examinations; density, heterogeneous controls and incomplete access set the confidence, not correctness. The grade applies to the exact registered claim; stronger findings for narrower subclaims remain separate. No complete-paper audit or blinded calibration is claimed. This educational review does not establish an individual treatment recommendation.

Medical information notice: This review provides general health information, not medical advice. It cannot diagnose a condition or replace an individual discussion with a qualified clinician.

Source checking used bounded original primary abstract passages/status records and selected original partial-body passages. No complete-abstract or complete-paper audit, or independent scientific peer review, is claimed. Detailed access counts and limitations appear in the review.

A low grade is not an accusation. It means the causal work has not been done yet, which is the ordinary state of most actively studied mechanisms in biology.

At a glance

Sub-claims assessed separately

Full assessment

From here the review is written for a technical reader: caps and thresholds, competing models, the independence basis, and the citation ledger.

Sources: 58 primary reports; claims: 63; verified: 63; unverified: 0. Verification is bounded attribution. Access: 3 partial-body and 55 bounded abstract/status examinations. Literature cutoff: 2026-10-13; source examination UTC date: 2026-10-13. No complete-paper audit or blinded calibration claimed. Adversarial review: Generated by Sol 6.1; adversarially reviewed by Claude Opus 5.5 (Anthropic) on 2026-10-14. 6 corrections applied (see ADVERSARIAL REVIEW AND APPROVAL). Status: Approved as revised.

AT A GLANCE

Excess adiposity is a credible contributor to early-onset colorectal cancer, but its precise causal effect before 50 and its share of the incidence rise remain less securely identified than the broader colorectal-cancer link.

Deciding experiment: A preregistered consortium analysis combining measured repeated adiposity before diagnosis with an age-stratified valid genetic/natural-experiment design, explicitly reporting colorectal cancer diagnosed before 50 and separating adult from childhood exposure, sex, colon/rectum, ancestry, hereditary syndromes and screening/diagnostic contact.

Belief versus evidence: The preregistered primary-subclaim rule returns SC2 Γ for the precise age-defined causal claim. This is a bounded assessment of direct early-onset causal identification, not a refutation of the prospective association or the broader adiposity–colorectal-cancer causal literature. The cohort dose-response and genetic-natural-experiment equivalents were evaluated; their age/timing and system scopes do not silently become the exact registered outcome.

Best-supported rival: Inherited and sporadic molecular pathways differ. The genetically defined CMMRD cohort (PMID38552658) identifies genotype-dependent cancer timing in a known high-risk mechanism, whereas adiposity instruments (PMID33327948) address all-age CRC rather than directly isolating the before 50 endpoint. This separates a defined causal context from an incompletely matched clinical age-specific exposure claim; it does not make inherited disease the majority trend explanation.

COMPETING MECHANISTIC MODELS

Rank 1: Inherited and sporadic molecular pathways differ (Ε). Defined hereditary pathways can cause very early cancers, whereas many sporadic early-onset tumors share common colorectal mechanisms. CMMRD genotype-dependent childhood burden (PMID38552658) and unselected EOCRC germline findings (PMID27978560) support genuine context-specific inherited contributions. Rare syndromes and germline-case prevalence cannot explain most recent sporadic increase; adjusted tumor profiles sometimes resemble older disease (PMID34405229).

Rank 2: Adiposity causally contributes to early-onset colorectal carcinogenesis (Γ). Prediagnostic adiposity may promote metabolic/inflammatory tumor susceptibility and progression before 50. Prospective NHSII (PMID30326010) and Chinese (PMID39036307) risk associations converge with all-age adult-adiposity MR (PMID33327948) and model perturbations (PMID39299402). Age-specific causal effect is not directly identified by all-age MR; site/sex differences, clinical nulls and childhood-MR nulls prevent a uniform timing model.

Rank 3: Multiple diet and microbial exposures contribute independently (Γ). Specific dietary/metabolic and genotoxic microbial routes may produce some young-age cancers independently of adiposity. Isogenic-control organoid colibactin signatures (PMID32106218), mouse adhesin dependence (PMID39506107) and young-age human tumor signatures (PMID40267983) identify a concrete microbial route; diet cohorts supply exposure associations. Diet and antibiotics are not individually established clinical causes; postdiagnosis dysbiosis can be consequence, and overall antibiotic association (PMID35029165) is near-null.

Rank 4: Screening and ascertainment influence observed age trends (Β). Detection, histological coding and case mix alter observed incidence and age distribution. French classification changes (PMID41191353) and tumor-size modeling (PMID40882029) demonstrate or predict local ascertainment contributions. Four-country birth-cohort trends (PMID40848248) and metastatic adenocarcinoma increases are inconsistent with explaining every trend as classification artifact.

The tournament does not alter the nominal level. Compatible pathways and flexible alternatives are not automatically proved by explaining observations.

FULL REVIEW

Registered claim and scope

Excess adiposity before age 50 causally increases the risk of early-onset colorectal cancer.

The precise lead clinical proposition is SC2. Mechanism, competing-cause, generalization and safety companions remain separately reported; they do not replace the primary clinical endpoint.

Selection: position 11 in the registered editorial traffic-priority list; registration timestamp 2026-10-13T16:06:25.313824+00:00. Ranked editorial traffic priority rather than measured global popularity or a preset verdict. Registered claim assessed against original sources; no independent blinded calibration is claimed.

Evidence and causal interpretation

Rubric fit: Cancer rubric02 is the primary fit for initiation; metabolism rubric06 informs adiposity exposure, developmental timing and metabolic-proxy limits. Literal cancer Δ examples concern endogenous model manipulation. The declared population-exposure adaptation also admits the rubric’s explicit cohort dose-response across independent populations and genetic natural experiments. No obesity-randomization requirement is imposed. Qualifying inference must still match cancer before 50, exposure timing and the intended causal adiposity contrast; all-age CRC, adenomas and tumor progression are separate scopes.

The registered proposition concerns colorectal cancer diagnosed before 50, not every early-onset cancer or every adiposity-associated tumor. NHSII women provide a direct prospective risk association with 114 cases; the Chinese cohort and Korean nested analysis supply additional clinical contexts. Recent pooled European findings are informative but define early onset as <55, so their larger event count is not presented as a strictly under 50 result.

Adiposity–CRC causal support extends beyond these cohort associations. Adult BMI/waist-hip-ratio MR uses instruments that reduce some reverse-causation and behavioral-confounding threats, and experimental models identify metabolic tumor-promotion pathways. The original MR Methods, however, define overall/site-specific CRC rather than the registered early-onset endpoint; sensitivity estimators vary in precision and assumptions. This age-scope bridge is the deciding uncertainty, not a demand for randomized obesity exposure.

Early-life exposure is not interchangeable with early diagnosis. Childhood/adolescent cohorts support some later colon associations but differ by sex, site and later weight change; the early-childhood MR paper found little evidence of all-age CRC effects. A childhood instrument null does not negate adult adiposity causality, and a cancer detected after 50 does not become an EOCRC event because weight was measured in adolescence.

Explaining risk is distinct from explaining the rise. GBD BMI-attributed burden estimates assume exposure-risk functions and share a common data source. They do not independently show that most of the incidence increase is due to adiposity. Ecological co-trends are hypothesis-generating. French neuroendocrine reclassification affects local young-age trends, while four-country birth-cohort and metastatic-disease findings argue against a universal ascertainment-only explanation.

The companion exposures have distinct evidence. Ultraprocessed-food and sulfur-microbial-diet studies often measure adenomas rather than cancer, and the large Swedish antibiotic analysis is not conclusively positive overall. Colibactin has a stronger molecule-specific case: isogenic-control organoid mutagenesis, adhesin-dependent mouse carcinogenesis and matching younger human tumor signatures. This does not establish every microbial change, every antibiotic or processed-food category as an independent clinical EOCRC cause.

Inherited pathways are real and must be retained without replacing sporadic exposure questions. CMMRD provides striking genotype-dependent early cancer risk; an unselected EOCRC accrual found pathogenic susceptibility variants in 16 percent of cases. Neither finding makes hereditary mechanisms a complete explanation for recent population trends or turns disease-risk variants into adiposity instruments.

Null and measurement limitations were not discarded. The hospital study found similar obesity prevalence in cases and controls; All of Us found no age-specific BMI interaction rather than proving no BMI effect. The Shanghai screening study has only six actual cancers and a nonsignificant spline overall test. NHANES millions are weighted population estimates. These limits matter to scientific interpretation more than a large citation count.

The selected original publisher Methods/Results for PMID40267983 distinguish historical mutagenic signatures from current bacteria: signature-positive cancers did not reliably have current pks-positive reads. Early clonal imprinting refers to tumor evolution and is not a directly dated childhood infection measurement. This prevents translating the result into a necessary-current-infection claim.

Observed signed-in Consensus Pro synthesis was reconciled against eight original-source aspects. Its male-colon estimate comes from a before 55 pooled cohort, the German timed-weight numbers are case-control odds rather than hazards, and its categorical assertion about most incidence growth exceeds a transportable original decomposition. Provider coverage-matrix counts do not establish absent literature or primary examination totals.

HYPOTHESIS STATE EVALUATION FRAMEWORK (HSEF v5.3)

Subclaim matrix and aggregation

Preregistered aggregation: primary-subclaim. The preregistered primary-subclaim rule returns SC2 Γ for the precise age-defined causal claim. This is a bounded assessment of direct early-onset causal identification, not a refutation of the prospective association or the broader adiposity–colorectal-cancer causal literature. The cohort dose-response and genetic-natural-experiment equivalents were evaluated; their age/timing and system scopes do not silently become the exact registered outcome. Sensitivities: weakest-link Β; primary-subclaim Γ; scope-matched Γ.

Decisive-perturbation inventory and causal tally

[3] Prospective NHSII follow-up of 85256 women documented 114 cancers before 50; obesity and BMI dose increments were associated with greater risk, including analyses without recent endoscopy or family history. Self-reported exposure, few events, residual confounding and female-only enrollment limit causal and broad-scope inference; a linked correction is attached to this source.

[8] Three pooled European cohorts associated higher BMI with CRC diagnosed before 55, especially male colon cancer; effects were weaker for women and rectal cancers. The abstract's early-onset threshold is 55 rather than registered 50; cohort adjustment and sex/site heterogeneity do not prove the precise under 50 causal scope.

[17] All of Us matched case-control analysis included 485 cancers before 50 and found no age-specific BMI or alcohol effects and no common-variant age interaction after multiple testing. No age interaction is not proof that BMI has no causal effect at either age; retrospective exposure, ancestry adjustment and limited rare-effect power constrain inference about trends.

[23] Large sex/site-specific MR analyses linked genetically predicted BMI or waist-hip ratio with all-age CRC, with differing sex effects; measured metabolite classes did not explain the associations. Genetic instruments reflect longer-term adiposity and require exclusion/pleiotropy assumptions; no under 50 endpoint is identified in the abstract, and mediator nonattenuation is not pathway exclusion.

[24] Mouse high-fat-diet exposure over 48 weeks did not increase intestinal-crypt genomic instability without additional activating mutation or chemical exposure. This tests a mutational mechanism rather than all carcinogenic routes; diet-induced obesity and organoid selection do not prove adiposity cannot promote tumors by other mechanisms.

[27] Obese/diabetic KKAy mice retained more chemically induced tumors than wild type despite microbiome changes from antibiotics or cohousing; antibiotic suppression occurred only in wild type. Different genetic backgrounds, diabetes and carcinogen exposure confound adiposity isolation; failure of this microbial intervention does not exclude every organism or human causal pathway.

[35] French registries showed neuroendocrine reclassification materially influenced apparent young-age CRC trends, while metastatic adenocarcinoma incidence also increased. A regional coding natural change demonstrates a local ascertainment contribution, not that all global EOCRC growth is artifact or that adiposity lacks a causal role.

[41] A tertiary-hospital case-control analysis of 269 EOCRC cases found obesity, smoking and diabetes prevalence similar to controls, while family history and other factors were associated. Hospital control selection, contemporaneous measurements and limited cases make an imprecise BMI null; this does not prove adiposity never contributes.

[43] Swedish prescription/pathology-linked analysis of 2557 EOCRC cases found no conclusive overall prior-antibiotic association after adjustment; broad-spectrum exposure had a modest positive estimate. Confounding by infection/indication, short lag and prescription capture remain; the overall near-null does not exclude molecule- or timing-specific effects.

[53] Two-sample MR using BMI from birth through 8 years found little evidence of all-age CRC association in European and East Asian genetic-similarity populations. The null childhood instrument differs from adult/lifelong adiposity and does not negate later-adult BMI causal effects; it also does not directly test cancer diagnosed before 50.

[55] Repeated pks-positive E. coli exposure caused a characteristic mutation signature in human intestinal organoids absent with isogenic pks-mutant controls; matching signatures occurred in human CRC genomes. The experiment establishes organism-specific mutagenesis in an engineered exposure system, not the fraction of EOCRC caused by microbes or proof that antibiotics/processed foods cause cancer.

[56] Adhesin manipulation and FimH inhibition attenuated colibactin genotoxicity and CRC exacerbation in a predisposed transgenic mouse model. Engineered microbiota dependence and one model constrain spontaneous human EOCRC transfer; a linked correction must remain attached rather than a separate report count.

Tally: 0 positive qualifying study units across scoped subclaims; 12 screened discriminating contrast units. These are not independent causal lines or proof of full-claim necessity/sufficiency. Null and negative contrasts are discriminators, not positive threshold crossings. Reports and replications are not independent causal lines. Zero exact age-specific adiposity positive perturbation/natural-experiment study units are claimed; examined model and all-age MR positives retain separate inference scopes. Risk prediction, DNA methylation age, microbial presence and precursors are not cancer-incidence causal crossings. GBD modeled attributable fractions assume exposure-risk relationships and do not independently verify their own causes.

No positive field-valid crossing was verified for the registered causal scope in this bounded audit. Pivotal null or negative tests remain informative; their implications depend on comparator quality, target engagement, power and the registered scope.

Cap audit, precedence and Β/Γ/Δ boundary

A genetic estimate of adult/lifelong adiposity on overall CRC is a genuine causal-inference class under explicit instrument assumptions; it is not demoted to mere GWAS association. Its outcome age scope is nevertheless broader than the registered endpoint. Similarly, endogenous Apc-model interventions can identify a model mechanism while clinical transfer remains constrained. Caps bind offered inferences; no universal association cap is reapplied over a verified same-scope independent causal set, because that set was not established here.

Universal six-cap audit: All six universal caps were considered: cell-only inference, unperturbed association, ectopic expression, biomarker-only intervention, forced transfer/seeding, and inherited-to-sporadic scope transfer. Each binds its offered inference; complementary same-scope causal classes are assessed together rather than re-capped in isolation.

Gamma hard-rule audit: Cell-only microbial mutagenesis is capped for direct organismal inference. Transplanted tumor growth is not spontaneous human initiation. No matched under 50 adiposity genetic/natural perturbation was verified in the examined set. Prospective associations and all-age causal instruments are retained rather than relabeled randomized exposure.

SC1 cap: “Pure association evidence — however large, however replicated — cannot exceed Β.” Prospective under 50 cohorts support a reproducible adiposity association, including NHSII women and a Chinese cohort. The association subclaim remains Β because statistical adjustment, temporal order and a risk gradient do not themselves identify the causal exposure. Age thresholds, sex/site differences and small direct event counts are retained.

SC3 cap: “Pure association evidence — however large, however replicated — cannot exceed Β.” A contribution to individual risk does not establish that adiposity explains most of the recent incidence increase. Ecological co-trends and GBD attributable fractions are association/model-based; a Korean local estimate assigns only a minority BMI-attributable fraction, and classification/detection changes plus birth-cohort differences require explicit decomposition.

SC4 cap: “Pure association evidence — however large, however replicated — cannot exceed Β.” The bundled assertion requires separate evidence for each exposure. Processed-food/diet evidence is observational and often adenoma-only; overall antibiotic association is near-null in a large EOCRC analysis. Colibactin-specific organoid mutagenesis and mouse adhesin perturbations are stronger mechanistically and match younger human tumor signatures, but do not establish all microbial changes, antibiotics and processed foods as individual EOCRC causes. The conjunctive companion is held at its weakest clinical exposure-evidence tier Β.

SC1 field-valid definition: “ - Endogenous genetic manipulation in autochthonous models - Patient-derived system perturbation with matched rescue AND in vivo confirmation - Clonal tracking demonstrating persistence of dependency through selection pressure” Positive full-scope result: no. Prospective under 50 cohorts support a reproducible adiposity association, including NHSII women and a Chinese cohort. The association subclaim remains Β because statistical adjustment, temporal order and a risk gradient do not themselves identify the causal exposure. Age thresholds, sex/site differences and small direct event counts are retained.

SC2 field-valid definition: “ - Endogenous genetic manipulation in autochthonous models - Patient-derived system perturbation with matched rescue AND in vivo confirmation - Clonal tracking demonstrating persistence of dependency through selection pressure” Positive full-scope result: no. Age-specific cohorts, all-age adult-adiposity MR and animal perturbations supply biologically meaningful triangulation. Nevertheless, the reviewed MR outcomes are overall/site-specific CRC rather than cancer before 50, while animal interventions substitute high-fat diet, susceptible genotype or transplanted progression for human timed adiposity. The exact under 50 causal dependency remains unisolated; Γ preserves partial mechanistic support without denying broader adiposity–CRC causality.

SC3 field-valid definition: “ - Endogenous genetic manipulation in autochthonous models - Patient-derived system perturbation with matched rescue AND in vivo confirmation - Clonal tracking demonstrating persistence of dependency through selection pressure” Positive full-scope result: no. A contribution to individual risk does not establish that adiposity explains most of the recent incidence increase. Ecological co-trends and GBD attributable fractions are association/model-based; a Korean local estimate assigns only a minority BMI-attributable fraction, and classification/detection changes plus birth-cohort differences require explicit decomposition.

SC4 field-valid definition: “ - Endogenous genetic manipulation in autochthonous models - Patient-derived system perturbation with matched rescue AND in vivo confirmation - Clonal tracking demonstrating persistence of dependency through selection pressure” Positive full-scope result: no. The bundled assertion requires separate evidence for each exposure. Processed-food/diet evidence is observational and often adenoma-only; overall antibiotic association is near-null in a large EOCRC analysis. Colibactin-specific organoid mutagenesis and mouse adhesin perturbations are stronger mechanistically and match younger human tumor signatures, but do not establish all microbial changes, antibiotics and processed foods as individual EOCRC causes. The conjunctive companion is held at its weakest clinical exposure-evidence tier Β.

SC1 applied rubric adaptation (distinct from the unchanged literal criterion): Apply cancer-initiation rubric02 and its explicit cohort/genetic natural-experiment equivalents to the registered clinical age-specific endpoint; distinguish prospective exposure association, causal instruments, incidence decomposition and precursor outcomes. Metabolic biomarkers alone do not establish cancer causality.

SC2 applied rubric adaptation (distinct from the unchanged literal criterion): Apply cancer-initiation rubric02 and its explicit cohort/genetic natural-experiment equivalents to the registered clinical age-specific endpoint; distinguish prospective exposure association, causal instruments, incidence decomposition and precursor outcomes. Metabolic biomarkers alone do not establish cancer causality.

SC3 applied rubric adaptation (distinct from the unchanged literal criterion): Apply cancer-initiation rubric02 and its explicit cohort/genetic natural-experiment equivalents to the registered clinical age-specific endpoint; distinguish prospective exposure association, causal instruments, incidence decomposition and precursor outcomes. Metabolic biomarkers alone do not establish cancer causality.

SC4 applied rubric adaptation (distinct from the unchanged literal criterion): Apply cancer-initiation rubric02 and its explicit cohort/genetic natural-experiment equivalents to the registered clinical age-specific endpoint; distinguish prospective exposure association, causal instruments, incidence decomposition and precursor outcomes. Metabolic biomarkers alone do not establish cancer causality.

SC1 literal Γ hard rule: “Sphere formation alone” Not a qualifying stand-alone clinical causal class; sphere observations within mixed model experiments are not counted as an under 50 causal crossing.

SC1 literal Γ hard rule: “Marker enrichment alone” Triggered where methylation clocks, transcriptomic/metabolic signatures or microbial abundance are used alone as cause; ceiling Γ for mechanistic inference, with pure association restricted to Β.

SC1 literal Γ hard rule: “Xenograft tumor-initiation without lineage-resolved endogenous evidence” Triggered for the transplanted/orthotopic tumor-support class, not imposed on prospective human cohorts or genetic instruments.

SC1 literal Γ hard rule: “In vitro-only mechanistic studies, however elegant” Triggered for organoid/cell-only support offered as direct human causal identification; related mouse and human-genome triangulation is assessed separately rather than erased.

SC2 literal Γ hard rule: “Sphere formation alone” Not a qualifying stand-alone clinical causal class; sphere observations within mixed model experiments are not counted as an under 50 causal crossing.

SC2 literal Γ hard rule: “Marker enrichment alone” Triggered where methylation clocks, transcriptomic/metabolic signatures or microbial abundance are used alone as cause; ceiling Γ for mechanistic inference, with pure association restricted to Β.

SC2 literal Γ hard rule: “Xenograft tumor-initiation without lineage-resolved endogenous evidence” Triggered for the transplanted/orthotopic tumor-support class, not imposed on prospective human cohorts or genetic instruments.

SC2 literal Γ hard rule: “In vitro-only mechanistic studies, however elegant” Triggered for organoid/cell-only support offered as direct human causal identification; related mouse and human-genome triangulation is assessed separately rather than erased.

SC3 literal Γ hard rule: “Sphere formation alone” Not a qualifying stand-alone clinical causal class; sphere observations within mixed model experiments are not counted as an under 50 causal crossing.

SC3 literal Γ hard rule: “Marker enrichment alone” Triggered where methylation clocks, transcriptomic/metabolic signatures or microbial abundance are used alone as cause; ceiling Γ for mechanistic inference, with pure association restricted to Β.

SC3 literal Γ hard rule: “Xenograft tumor-initiation without lineage-resolved endogenous evidence” Triggered for the transplanted/orthotopic tumor-support class, not imposed on prospective human cohorts or genetic instruments.

SC3 literal Γ hard rule: “In vitro-only mechanistic studies, however elegant” Triggered for organoid/cell-only support offered as direct human causal identification; related mouse and human-genome triangulation is assessed separately rather than erased.

SC4 literal Γ hard rule: “Sphere formation alone” Not a qualifying stand-alone clinical causal class; sphere observations within mixed model experiments are not counted as an under 50 causal crossing.

SC4 literal Γ hard rule: “Marker enrichment alone” Triggered where methylation clocks, transcriptomic/metabolic signatures or microbial abundance are used alone as cause; ceiling Γ for mechanistic inference, with pure association restricted to Β.

SC4 literal Γ hard rule: “Xenograft tumor-initiation without lineage-resolved endogenous evidence” Triggered for the transplanted/orthotopic tumor-support class, not imposed on prospective human cohorts or genetic instruments.

SC4 literal Γ hard rule: “In vitro-only mechanistic studies, however elegant” Triggered for organoid/cell-only support offered as direct human causal identification; related mouse and human-genome triangulation is assessed separately rather than erased.

Upward threshold ladder

SC1 Δ: “Requires at least one meaningful causal perturbation showing the proposed mechanism contributes to initiation, maintenance, or relapse.” Satisfied: no. A prespecified same-scope age-specific causal-instrument or natural-experiment analysis with corroborating independent exposure evidence and negative controls; the majority-trend claim additionally requires transportable incidence decomposition, and each diet/antibiotic/microbial exposure must be separately tested.

SC2 Δ: “Requires at least one meaningful causal perturbation showing the proposed mechanism contributes to initiation, maintenance, or relapse.” Satisfied: no. A prespecified same-scope age-specific causal-instrument or natural-experiment analysis with corroborating independent exposure evidence and negative controls; the majority-trend claim additionally requires transportable incidence decomposition, and each diet/antibiotic/microbial exposure must be separately tested.

SC3 Δ: “Requires at least one meaningful causal perturbation showing the proposed mechanism contributes to initiation, maintenance, or relapse.” Satisfied: no. A prespecified same-scope age-specific causal-instrument or natural-experiment analysis with corroborating independent exposure evidence and negative controls; the majority-trend claim additionally requires transportable incidence decomposition, and each diet/antibiotic/microbial exposure must be separately tested.

SC4 Δ: “Requires at least one meaningful causal perturbation showing the proposed mechanism contributes to initiation, maintenance, or relapse.” Satisfied: no. A prespecified same-scope age-specific causal-instrument or natural-experiment analysis with corroborating independent exposure evidence and negative controls; the majority-trend claim additionally requires transportable incidence decomposition, and each diet/antibiotic/microbial exposure must be separately tested.

Upper ladder: Cohort dose-response across independent populations, adult/early-life genetic instruments, familial natural contexts, pathway perturbation and ascertainment changes were all considered. Adult MR supports the broader CRC link but does not estimate a timed pre-50 exposure on a strictly under 50 endpoint. Cohort-specific positive and null gradients do not alone remove shared behavioral confounding. Upper levels remain reachable through matched natural experiments and convergent classes; no randomized-obesity requirement is imposed.

Independence: No model or scored subclaim reaches Ζ in this draft. Cohort, genetic and animal classes fail through different mechanisms, but their endpoint scopes are not automatically identical. Repeated MR using overlapping consortia, NHSII reports, GBD models and Israeli follow-ups are not counted as new independent causal lines solely because publications differ.

Falsification and pitfalls

Association dressed as mechanism. Adjusted cohort odds or hazard ratios are described as isolated adiposity causal effects.

Selective generalization. All-age CRC or <55 outcomes are treated as exact <50 disease; early-life exposure is confused with early-onset diagnosis.

Confirmation by biomarker. Methylation age, microbial abundance and metabolic signatures are treated as causes.

Mechanism conflation. Processed foods, antibiotic exposure and organism-specific genotoxin mechanisms are bundled.

A universal cap re-applied to a convergent evidence set. Prevented by explicitly retaining MR and perturbation classes with their actual scope instead of denying broad causal evidence.

Scope fitted to the result. Sex/site nulls and childhood-MR nulls remain in the matrix; groups are not defined by response.

Replication confused with reproduction. NHSII reports, Israeli recruitment follow-ups and GBD reanalyses are grouped or marked overlapping.

Adversarial steelman: above, then below

One rung above (Δ): multiple prospective age-specific risk gradients, adult-adiposity MR, familial carrier/noncarrier findings and in-vivo metabolic perturbations supply a strong triangulation case. Cancer rubric02 explicitly permits cohort dose-response and genetic natural experiments, so an RCT is unnecessary. A reasonable reviewer could admit this combination as the first population-level causal crossing, particularly for colon cancer.

One rung below (Β): the directly under 50 clinical studies remain observational, and an adiposity-correlated dietary/behavioral factor could confound their risk gradient. Strict age matching of every causal class would leave only association within the exact endpoint.

Δ is not assigned in this draft because the causal-instrument outcome and most experimental systems do not directly match the under 50 cancer endpoint, while several apparently young-age cohorts use <55 or all-age outcomes. The blocker is scope and unresolved shared exposure confounding, not absence of an unethical RCT. Β is rejected for the assembled hypothesis because endogenous-model and adult-adiposity genetic support provide more than co-occurrence, though they do not settle the exact early-onset causal magnitude. Coordinator should explicitly adjudicate the scope bridge rather than silently treating this judgement as a mechanical rule.

Final score, causal direction and M/S/D

The preregistered primary-subclaim rule returns SC2 Γ for the precise age-defined causal claim. This is a bounded assessment of direct early-onset causal identification, not a refutation of the prospective association or the broader adiposity–colorectal-cancer causal literature. The cohort dose-response and genetic-natural-experiment equivalents were evaluated; their age/timing and system scopes do not silently become the exact registered outcome.

Model interventions and adult-adiposity causal instruments support promotion/amplification of colorectal carcinogenic processes rather than adiposity being required or sufficient alone. Direct human before 50 mediation remains uncertain. The role is a mechanism interpretation informed by the broader experimental evidence, not proof that adiposity initiates most EOCRC cases.

Momentum: Mechanistic only. Verified active causal work includes DGAT perturbation (PMID39299402), microbiota/adhesin manipulation (PMID39506107) and adult/childhood adiposity MR (PMID33327948)/39587448. New under 50 cohort and genomic output is active, but no verified named pending age-specific adiposity intervention or instrument result was identified.

Stability: Stable. A prespecified age-stratified adiposity natural experiment or prevention analysis could clarify the causal scope and raise confidence/level; no named pending result with a directional effect on this grade was verified.

Divergence: Contested. No manufactured field consensus or arithmetic gap. These dimensions do not change level or confidence.

PROPOSED DECISIVE EXPERIMENTS

A preregistered consortium analysis combining measured repeated adiposity before diagnosis with an age-stratified valid genetic/natural-experiment design, explicitly reporting colorectal cancer diagnosed before 50 and separating adult from childhood exposure, sex, colon/rectum, ancestry, hereditary syndromes and screening/diagnostic contact.

Readout: Incident pathologically verified CRC before 50 with prespecified absolute/relative effects; adenomas and diagnosis after 50 analyzed separately; attributable-trend claims require population exposure and ascertainment decomposition. Controls: Instrument strength/pleiotropy diagnostics, independent replication without reused outcome cohorts, exposure lag, negative-control outcomes/exposures, detection-contact adjustment with sensitivity analyses and fixed age/site strata; weight-loss prevention contrasts must handle selection and concurrent interventions. Supporting: A concordant under 50 causal estimate survives instrument and ascertainment checks and aligns with prospective exposure dose/timing; mechanism classes reproduce same-context predictions. Falsifying: Well-powered matched analyses show a clinically important near-null or confounding explanation despite reliable exposure measurement, or adult-versus-childhood/site effects contradict the proposed transportability.

KNOWLEDGE GAPS

A causal-instrument estimate matching strictly under 50 CRC and timed prediagnostic adiposity.

Absolute risk and effect heterogeneity by sex, site, ancestry and inherited background without outcome-defined scopes.

Distinguishing childhood adiposity effects from persistent adult exposure and correlated diet/activity.

A valid population decomposition of how much adiposity contributes to secular incidence, separating detection and histological classification.

Independent diet, antibiotic-class/timing and specific microbial exposure causal tests; precursor outcomes and diagnostic signatures remain separate.

Mechanism-specific prevention consequences in humans, without transferring susceptible-mouse or transplanted progression results to spontaneous young-age initiation.

REVIEWER NOTES

Original complete abstracts/status records and selected Methods/Results were examined and receipted. This is a bounded critical review, not an exhaustive systematic review, full-paper audit, independent scientific peer review or blinded calibration. Final coordinator adjudication must explicitly review the same-scope cohort/genetic-equivalent threshold decision.

Search leads, reviews, protocols and correction notices do not pad the primary count. Downloaded XML does not earn examination credit. Linked notices remain attached; unexamined correction details limit inference. This is a coordinator source/claim consistency assessment, not independent blinded calibration.

Consensus and original-source reconciliation

Consensus cites the pooled European cohort for a 55 percent male colon-risk increase per five BMI units while discussing a before 50 proposition. Original Methods explicitly define early onset as before 55 to increase power; Results confirm HR 1.55 for male colon cancer, female BMI HR 1.03, and weaker rectal associations. The number is correctly attributable to this paper, but it does not directly verify the registered before 50 endpoint or a proximal-only causal mechanism. Original PMID41422343.

Consensus describes timed BMI-at-20 and BMI-at-30 estimates of 2.56 and 2.06 as hazard-level elevations. The original German study is a retrospective population-based case-control analysis using logistic regression and odds ratios, with cases younger than 55 and recalled earlier weight. These estimates are neither prospective hazards nor strictly before 50 cancer outcomes. Original PMID34914944.

Consensus recognizes that traditional BMI instruments describe lifelong exposure, but its opening asserts that prospective and genetic analyses demonstrate the before 50 proposition. Original MR Methods use overall and site-specific CRC outcome consortia without a before 50 estimate; IVW and weighted-median positives coexist with heterogeneity and imprecise alternative estimators. This strengthens the broader causal link under assumptions while leaving age and timed-exposure transfer explicit. Original PMID33327948.

Consensus presents obesity associations around 1.5-1.9 as consistent prospective evidence. The original NHSII report provides a directly before 50 endpoint: 114 cancers in 85256 women, obesity RR 1.93 versus a lean-reference range and a BMI risk gradient. It is an observational female cohort with a linked disclosure correction, not a genetic or randomized exposure experiment and not proof of uniform effects across sex and site. Original PMID30326010.

Consensus categorically states adiposity does not explain most incidence growth, largely citing reviews and site patterns. The Korean original computes local BMI-attributable fractions around ten percent under causal assumptions, while its outcome and exposure prevalence are specific to that cohort. This is informative but does not establish a global decomposition or disprove a majority contribution everywhere. The review therefore says a majority-growth claim is unestablished, rather than presenting its negation as proved. Original PMID40971243.

Consensus emphasizes a consistent adiposity relationship but did not visibly incorporate the requested All of Us counterpoint. That original found no age-specific BMI effect in a retrospective matched sample. Lack of age interaction is not a zero overall causal effect and does not by itself explain incidence trends; it is retained as a scope and precision boundary. Original PMID42610385.

Consensus gives a broad inherited-syndrome range, mentioning Lynch syndrome. The unselected original finds pathogenic susceptibility variants in 16 percent of 450 cases, spanning Lynch, APC, MUTYH and other genes. These categories must not be collapsed into a Lynch-only percentage or treated as a population attributable fraction for incidence growth. Original PMID27978560.

Consensus describes microbial/antibiotic exposures as investigational but gives no visible original colibactin source detail in its response. The examined original publisher Methods/Results distinguish historical mutation signatures from current pks-positive bacteria and infer early clonal timing from tumor evolution. This supports a specific organism-genotoxin mechanism more strongly than generic dysbiosis, without establishing universal microbial causation or a directly measured childhood infection date. Original PMID40267983.

Consensus was used to identify literature and cross-check specific claims about design, endpoints, scope and research gaps. Those aspects were checked against original primary abstracts/status records and selected original Methods/Results passages. Consensus output was not treated as original full text. Provider search omissions are not evidence that a study does not exist. Source access and examination receipts are retained. Scite use or monthly unavailability is recorded per case; provider citation counts do not change causal grades.

Research timeline and gaps: Adiposity and Early-Onset Colorectal Cancer

Selected publication milestones, not an exhaustive history or discovery priority claim.

2011 — Measured adolescent BMI linked prospectively to later colon risk; exposure age and diagnosis age remain distinct. Original PMID22056504.

2011 — Early-adult BMI associations occur in mismatch-repair carriers and noncarriers; hereditary ascertainment remains separate. Original PMID21559014.

2015 — Early-life fatness associations differ by sex in two cohorts and mainly later-life outcomes. Original PMID25777804.

2016 — Measured adolescent BMI and inflammatory ESR predict later CRC in Swedish men. Original PMID25986947.

2016 — Adult-adiposity MR supports a broader CRC causal link rather than an age-specific intervention effect. Original PMID27427428.

2017 — Unselected EOCRC germline testing finds a real susceptibility subgroup without explaining a secular trend. Original PMID27978560.

2017 — Measured childhood BMI is associated with adult colon, but not rectal, cancer. Original PMID28803329.

2019 — NHSII directly observes a prospective adiposity gradient for CRC before 50 in women. Original PMID30326010.

2020 — Isogenic-control human organoid exposure identifies a colibactin mutation signature matching human tumors. Original PMID32106218.

2020 — Sex/site-specific adult-adiposity MR supports overall CRC causality under instrument assumptions. Original PMID33327948.

2021 — Claims-based metabolic syndrome associates with EOCRC but does not isolate adiposity. Original PMID33037055.

2021 — A sulfur-microbial dietary score associates with early adenomas, not a direct clinical cancer outcome. Original PMID34273347.

2022 — Large Swedish antibiotic analysis finds no conclusive overall EOCRC association. Original PMID35029165.

2022 — German early-adult obesity associations use a <55 outcome threshold. Original PMID34914944.

2023 — Age-specific susceptibility GWAS does not constitute MR of adiposity. Original PMID36869385.

2024 — Inherited repair deficiency establishes a distinct high-risk early-cancer context. Original PMID38552658.

2024 — Adhesin perturbation narrows a colibactin carcinogenic mechanism in a predisposed mouse model. Original PMID39506107.

2025 — Human genomic signatures link historical colibactin mutagenesis with younger tumors, distinct from current bacteria. Original PMID40267983.

2025 — Histological reclassification contributes materially to some local young-age incidence trends. Original PMID41191353.

2025 — Early-childhood BMI MR is near-null for all-age CRC, distinguishing timing from adult adiposity. Original PMID39587448.

2026 — Pooled European cohort establishes sex/site heterogeneity but defines early onset as <55. Original PMID41422343.

2026 — All of Us shows no age-specific BMI interaction; this is not proof of a zero overall effect. Original PMID42610385.

2026 — UPF association concerns early conventional adenomas and not serrated lesions. Original PMID41231486.

Research gaps

A causal-instrument estimate matching strictly under 50 CRC and timed prediagnostic adiposity.

Absolute risk and effect heterogeneity by sex, site, ancestry and inherited background without outcome-defined scopes.

Distinguishing childhood adiposity effects from persistent adult exposure and correlated diet/activity.

A valid population decomposition of how much adiposity contributes to secular incidence, separating detection and histological classification.

Independent diet, antibiotic-class/timing and specific microbial exposure causal tests; precursor outcomes and diagnostic signatures remain separate.

Mechanism-specific prevention consequences in humans, without transferring susceptible-mouse or transplanted progression results to spontaneous young-age initiation.

Boundary experiment

A preregistered consortium analysis combining measured repeated adiposity before diagnosis with an age-stratified valid genetic/natural-experiment design, explicitly reporting colorectal cancer diagnosed before 50 and separating adult from childhood exposure, sex, colon/rectum, ancestry, hereditary syndromes and screening/diagnostic contact.

ADVERSARIAL REVIEW AND APPROVAL

Authoring model: Sol 6.1. Adversarial reviewer: Claude Opus 5.5 (Anthropic). Date: 2026-10-14.

Scope of checks. All 58 reference PMIDs verified against PubMed (58/58 resolve to the stated title, first author, journal and year; no retractions or expressions of concern; no empty DOI fields; no dropped collective authors). All 23 research-timeline years checked against the cited records. Ten load-bearing claims spot-checked against source abstracts (PMID 30326010, 41422343, 34914944, 42610385, 35029165, 33327948, 39587448, 32106218, 39506107, 27978560). Scoring logic re-derived against hsef-hypothesis-review-v5/SKILL.md Step 3 and references/hsef-framework.md. The score card was checked for the same rendering-defect family as the prose and for internal agreement with it. No level or confidence change.

Rejected findings (checked, no change). 1. Headline Γ and the primary-subclaim rule. SC2 is named as the primary subclaim before scoring ("The precise lead clinical proposition is SC2"), the alternatives are reported (weakest-link Β, scope-matched Γ), and the derivation wording matches the arithmetic. The scope-matched rule would also return Γ here, so no rule-shopping advantage exists. No change. 2. "Best-supported rival" renderer artifact. Checked the tournament ordering: rank 1 is "Inherited and sporadic molecular pathways differ", which is not this review's nominal model (the nominal "Adiposity causally contributes to early-onset colorectal carcinogenesis" ranks 2). The prose is correct as written. 3. Spliced earlier-approval sentence. Not present in this review; no approval is claimed. Checked. 4. Momentum "Mechanistic only". The required artifact names specific active work (DGAT perturbation, microbiota/adhesin manipulation, adult and childhood adiposity MR) and the review explicitly states no verified named pending age-specific result was identified. Consistent with lead policy that completed work cannot support "Decisive work underway". 5. REVIEWER NOTES versus the Consensus section. The notes do not claim that no Consensus work was performed; a substantive eight-aspect Consensus reconciliation section is present and consistent with them. No change. 6. Pooled European female BMI HR 1.03. The abstract reports only "weaker associations with BMI … for women"; the specific 1.03 figure is attributed to the original Results, to which the review claims partial-body access. Not contradicted by the abstract and directionally consistent. Not changed, but see residual limitations. 7. Causal tally count — resolved. The score card recorded 13 screened discriminating contrasts against 12 printed in the inventory. The score card's own evidence_inventory settles it: exactly 12 entries carry the decisive-perturbation flag, matching the 12 printed. Both artifacts are now 12 (finding A-5). The headline tally of 0 positive qualifying study units is unchanged. 8. v5.claim_form "phenomenon" warning. The registered claim states a causal relation plus consequence; rejected per the standing convention.

Residual limitations. Three of 58 sources had partial-body examination; the rest rest on abstracts and linked-status metadata, so figure-level and subgroup-level numbers taken from original Results (notably the female BMI HR in PMID 41422343) could not be independently reconfirmed here. The package has no lay_summary, so the validator fails that v5.4 check — a pre-existing gap, not introduced by this pass. This pass is a source, claim and scoring-logic audit, not a blinded recalibration.

Approval: Following adversarial review and the corrections listed above, the revised document is approved. — Claude Opus 5.5 (Anthropic), 2026-10-14.

REFERENCES

1. Ayoub NL, Francoeur AA, Chang J et al. Population-Level Trends in Lifestyle Factors and Early-Onset Breast, Colorectal, and Uterine Cancers. Cancers. 2026. PMID 41514674. DOI: 10.3390/cancers18010167. Access: abstract_metadata. Study: PMID41514674.

2. Barot SV, Sangwan N, Nair KG et al. Distinct intratumoral microbiome of young-onset and average-onset colorectal cancer. EBioMedicine. 2024. PMID 38306898. DOI: 10.1016/j.ebiom.2024.104980. Access: abstract_metadata. Study: PMID38306898.

3. Liu PH, Wu K, Ng K et al. Association of Obesity With Risk of Early-Onset Colorectal Cancer Among Women. JAMA oncology. 2019. PMID 30326010. DOI: 10.1001/jamaoncol.2018.4280. Access: abstract_metadata. Study: NHSII.

4. Juo YY, Gibbons MAM, Dutson E et al. Obesity Is Associated with Early Onset of Gastrointestinal Cancers in California. Journal of obesity. 2018. PMID 30327727. DOI: 10.1155/2018/7014073. Access: abstract_metadata. Study: PMID30327727.

5. Wang C, Du M, Kim H et al. Ultraprocessed Food Consumption and Risk of Early-Onset Colorectal Cancer Precursors Among Women. JAMA oncology. 2026. PMID 41231486. DOI: 10.1001/jamaoncol.2025.4777. Access: abstract_metadata. Study: NHSII.

6. Yue B, Lu Z, Zong D et al. Global, regional, and national burden of early-onset and late-onset colorectal cancer attributable to high body-mass index from 1990 to 2021: a trend analysis and forecasts up to 2040 based on the global burden of disease study 2021. BMC gastroenterology. 2025. PMID 41257562. DOI: 10.1186/s12876-025-04432-7. Access: abstract_metadata. Study: GBD2021-CRC.

7. Zhang Y, Han F, Qu J. Global, regional, and national burden of early-onset colorectal cancer attributable to high BMI (1990-2021). Frontiers in medicine. 2025. PMID 41041441. DOI: 10.3389/fmed.2025.1631392. Access: abstract_metadata. Study: GBD2021-CRC.

8. Laskar RS, Murphy N, Ferrari P et al. A prospective investigation of early-onset colorectal cancer risk factors-pooled analysis of three large-scale European cohorts. British journal of cancer. 2026. PMID 41422343. DOI: 10.1038/s41416-025-03303-y. Access: full_text_partial. Study: PMID41422343.

9. Imperiale TF, Myers LJ, Barker BC et al. Risk Factors for Early-onset Sporadic Colorectal Cancer in Male Veterans. Cancer prevention research (Philadelphia, Pa.). 2023. PMID 37079701. DOI: 10.1158/1940-6207.CAPR-22-0506. Access: abstract_metadata. Study: PMID37079701.

10. Pan Z, Huang J, Huang M et al. Risk factors for early-onset colorectal cancer: A large-scale Chinese cohort study. Journal of the National Cancer Center. 2023. PMID 39036307. DOI: 10.1016/j.jncc.2023.01.001. Access: abstract_metadata. Study: PMID39036307.

11. Shen J, Wu Y, Mo M et al. Risk Factors Associated With Early-Onset Colorectal Neoplasm in Chinese Youth: A Prospective Population-Based Study. Frontiers in oncology. 2021. PMID 34692479. DOI: 10.3389/fonc.2021.702322. Access: abstract_metadata. Study: PMID34692479.

12. Puzzono M, Mannucci A, Di Leo M et al. Diet and Lifestyle Habits in Early-Onset Colorectal Cancer: A Pilot Case-Control Study. Digestive diseases (Basel, Switzerland). 2022. PMID 35086089. DOI: 10.1159/000521932. Access: abstract_metadata. Study: PMID35086089.

13. Sala I, Botteri E, Berstad P et al. Association between lifestyle and risk of early-onset cancer: Evidence from the European Prospective Investigation into Cancer and Nutrition and the UK Biobank. European journal of cancer (Oxford, England : 1990). 2026. PMID 42269246. DOI: 10.1016/j.ejca.2026.116869. Access: abstract_metadata. Study: PMID42269246.

14. Bendor CD, Bardugo A, Tsur AM et al. Adolescent to adulthood weight trajectories and the risk of obesity-related cancers, overall and early-onset: a population-based cohort study. EClinicalMedicine. 2026. PMID 41552005. DOI: 10.1016/j.eclinm.2025.103711. Access: abstract_metadata. Study: PMID41552005.

15. Daca-Alvarez M, Perea J, Corchete L et al. Regional patterns of early-onset colorectal cancer from the GEOCODE (Global Early-Onset COlorectal Cancer DatabasE)-European consortium: retrospective cohort study. BJS open. 2025. PMID 40103402. DOI: 10.1093/bjsopen/zraf024. Access: abstract_metadata. Study: PMID40103402.

16. Zheng R, Du M, Zhang B et al. Body mass index (BMI) trajectories and risk of colorectal cancer in the PLCO cohort. British journal of cancer. 2018. PMID 29872147. DOI: 10.1038/s41416-018-0121-y. Access: abstract_metadata. Study: PMID29872147.

17. Nigam A, Onongaya C, Meshram P et al. Traditional Risk Factors Are Not Associated with the Rise of Early-Onset Colorectal Cancer: An All of Us Study. Diseases of the colon and rectum. 2026. PMID 42610385. DOI: 10.1097/DCR.0000000000004351. Access: abstract_metadata. Study: PMID42610385.

18. Hussan H, Patel A, Ma J et al. Historical Obesity and Early-Onset Cancers: A Propensity-Weighted Analysis of the National Health and Nutrition Examination Survey. Digestive diseases and sciences. 2024. PMID 38030832. DOI: 10.1007/s10620-023-08194-8. Access: abstract_metadata. Study: PMID38030832.

19. Baek JY, Jeong SY, Shin A. Risk Factors for Early-Onset Colorectal Cancer: A Nested Case‒Control Study within the Korean National Health Insurance Service‒Health Screening Cohort. Cancer epidemiology, biomarkers & prevention : a publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology. 2026. PMID 40971243. DOI: 10.1158/1055-9965.EPI-25-0400. Access: abstract_metadata. Study: PMID40971243.

20. Ghimire J, Collins ME, Snarski P et al. Obesity-Facilitated Colon Cancer Progression Is Mediated by Increased Diacylglycerol O-Acyltransferases 1 and 2 Levels. Gastroenterology. 2025. PMID 39299402. DOI: 10.1053/j.gastro.2024.09.011. Access: abstract_metadata. Study: PMID39299402.

21. Paragomi P, Zhang Z, Abe SK et al. Body Mass Index and Risk of Colorectal Cancer Incidence and Mortality in Asia. JAMA network open. 2024. PMID 39196559. DOI: 10.1001/jamanetworkopen.2024.29494. Access: abstract_metadata. Study: PMID39196559.

22. Nimri L, Saadi J, Peri I et al. Mechanisms linking obesity to altered metabolism in mice colon carcinogenesis. Oncotarget. 2015. PMID 26472027. DOI: 10.18632/oncotarget.5561. Access: abstract_metadata. Study: PMID26472027.

23. Bull CJ, Bell JA, Murphy N et al. Adiposity, metabolites, and colorectal cancer risk: Mendelian randomization study. BMC medicine. 2020. PMID 33327948. DOI: 10.1186/s12916-020-01855-9. Access: full_text_partial. Study: PMID33327948.

24. Meyenberg M, Hakobyan A, Papac-Milicevic N et al. Mutational landscape of intestinal crypt cells after long-term in vivo exposure to high fat diet. Scientific reports. 2023. PMID 37633982. DOI: 10.1038/s41598-023-41123-3. Access: abstract_metadata. Study: PMID37633982.

25. Glenny EM, Lin T, Bandera VM et al. Obesity promotes conserved inflammatory and metabolic transcriptional programs in colon tumors: evidence from mouse models and the ColoCare Study Patient Cohort. The American journal of clinical nutrition. 2025. PMID 40998019. DOI: 10.1016/j.ajcnut.2025.09.031. Access: abstract_metadata. Study: PMID40998019.

26. Saxena A, Tammali R, Ramana KV et al. Aldose Reductase Inhibitor, Fidarestat Prevents High-fat Diet-induced Intestinal Polyps in ApcMin/+ Mice. Current cancer drug targets. 2018. PMID 28786349. DOI: 10.2174/1568009617666170808105633. Access: abstract_metadata. Study: PMID28786349.

27. Iwama N, Matsuda M, Tsuruta M et al. Relationship between obesity-related colorectal tumors and the intestinal microbiome: an animal-based trial. Journal of cancer research and clinical oncology. 2023. PMID 36399198. DOI: 10.1007/s00432-022-04477-1. Access: abstract_metadata. Study: PMID36399198.

28. Kong C, Liang L, Liu G et al. Integrated metagenomic and metabolomic analysis reveals distinct gut-microbiome-derived phenotypes in early-onset colorectal cancer. Gut. 2023. PMID 35953094. DOI: 10.1136/gutjnl-2022-327156. Access: abstract_metadata. Study: PMID35953094.

29. Cercek A, Chatila WK, Yaeger R et al. A Comprehensive Comparison of Early-Onset and Average-Onset Colorectal Cancers. Journal of the National Cancer Institute. 2021. PMID 34405229. DOI: 10.1093/jnci/djab124. Access: abstract_metadata. Study: PMID34405229.

30. Downham L, Laversanne M, Perdomo S et al. Increase of early-onset colorectal cancer: a cohort effect. Journal of the National Cancer Institute. 2026. PMID 40848248. DOI: 10.1093/jnci/djaf238. Access: abstract_metadata. Study: PMID40848248.

31. Li J, Pan Y, Guo F et al. Patterns in genomic mutations among patients with early-onset colorectal cancer: an international, multicohort, observational study. The Lancet. Oncology. 2025. PMID 40617240. DOI: 10.1016/S1470-2045(25)00239-6. Access: abstract_metadata. Study: PMID40617240.

32. Ercan AB, Aronson M, Fernandez NR et al. Clinical and biological landscape of constitutional mismatch-repair deficiency syndrome: an International Replication Repair Deficiency Consortium cohort study. The Lancet. Oncology. 2024. PMID 38552658. DOI: 10.1016/S1470-2045(24)00026-3. Access: abstract_metadata. Study: PMID38552658.

33. Mohammad Mirzaei N, Hur C, Terry MB et al. Modeling Early-Onset Cancer Kinetics Reveals Changes in Underlying Risk and the Impact of Population Screening. Cancer research. 2025. PMID 40882029. DOI: 10.1158/0008-5472.CAN-25-0186. Access: abstract_metadata. Study: PMID40882029.

34. Wang H, Cai Y, Jin M et al. Identification of specific susceptibility loci for the early-onset colorectal cancer. Genome medicine. 2023. PMID 36869385. DOI: 10.1186/s13073-023-01163-w. Access: abstract_metadata. Study: PMID36869385.

35. Jooste V, Nousbaum JB, Alves A et al. Epidemiological Classification Changes and Incidence of Early-Onset Colorectal Cancer. JAMA network open. 2025. PMID 41191353. DOI: 10.1001/jamanetworkopen.2025.41732. Access: abstract_metadata. Study: PMID41191353.

36. Jin N, Hoyd R, Yilmaz AS et al. Epigenetic Modulation, Intratumoral Microbiome, and Immunity in Early-Onset Colorectal Cancer. Cancer research communications. 2025. PMID 41134679. DOI: 10.1158/2767-9764.CRC-25-0177. Access: abstract_metadata. Study: PMID41134679.

37. Nguyen LH, Cao Y, Hur J et al. The Sulfur Microbial Diet Is Associated With Increased Risk of Early-Onset Colorectal Cancer Precursors. Gastroenterology. 2021. PMID 34273347. DOI: 10.1053/j.gastro.2021.07.008. Access: abstract_metadata. Study: NHSII.

38. Gilad O, Dosunmu GT, Gao G et al. Somatic Mutation Profiles of Colorectal Cancer by Birth Cohort. JCO precision oncology. 2025. PMID 41075301. DOI: 10.1200/PO-25-00717. Access: abstract_metadata. Study: PMID41075301.

39. Tran V, Spezia-Lindner N, Millien VO et al. Epidemiology of Early-onset Colorectal Cancer in a Safety-net Hospital System in Texas. Journal of clinical gastroenterology. 2025. PMID 40910875. DOI: 10.1097/MCG.0000000000002092. Access: abstract_metadata. Study: PMID40910875.

40. Chen H, Zheng X, Zong X et al. Metabolic syndrome, metabolic comorbid conditions and risk of early-onset colorectal cancer. Gut. 2021. PMID 33037055. DOI: 10.1136/gutjnl-2020-321661. Access: abstract_metadata. Study: PMID33037055.

41. Gausman V, Dornblaser D, Anand S et al. Risk Factors Associated With Early-Onset Colorectal Cancer. Clinical gastroenterology and hepatology : the official clinical practice journal of the American Gastroenterological Association. 2020. PMID 31622737. DOI: 10.1016/j.cgh.2019.10.009. Access: abstract_metadata. Study: PMID31622737.

42. Chang VC, Cotterchio M, De P et al. Risk factors for early-onset colorectal cancer: a population-based case-control study in Ontario, Canada. Cancer causes & control : CCC. 2021. PMID 34120288. DOI: 10.1007/s10552-021-01456-8. Access: abstract_metadata. Study: PMID34120288.

43. Nguyen LH, Cao Y, Batyrbekova N et al. Antibiotic Therapy and Risk of Early-Onset Colorectal Cancer: A National Case-Control Study. Clinical and translational gastroenterology. 2022. PMID 35029165. DOI: 10.14309/ctg.0000000000000437. Access: abstract_metadata. Study: PMID35029165.

44. Sanford NN, Giovannucci EL, Ahn C et al. Obesity and younger versus older onset colorectal cancer in the United States, 1998-2017. Journal of gastrointestinal oncology. 2020. PMID 32175114. DOI: 10.21037/jgo.2019.12.07. Access: abstract_metadata. Study: PMID32175114.

45. Li H, Boakye D, Chen X et al. Associations of Body Mass Index at Different Ages With Early-Onset Colorectal Cancer. Gastroenterology. 2022. PMID 34914944. DOI: 10.1053/j.gastro.2021.12.239. Access: abstract_metadata. Study: PMID34914944.

46. Jensen BW, Gamborg M, Gögenur I et al. Childhood body mass index and height in relation to site-specific risks of colorectal cancers in adult life. European journal of epidemiology. 2017. PMID 28803329. DOI: 10.1007/s10654-017-0289-0. Access: abstract_metadata. Study: PMID28803329.

47. Kantor ED, Udumyan R, Signorello LB et al. Adolescent body mass index and erythrocyte sedimentation rate in relation to colorectal cancer risk. Gut. 2016. PMID 25986947. DOI: 10.1136/gutjnl-2014-309007. Access: abstract_metadata. Study: PMID25986947.

48. Zhang X, Wu K, Giovannucci EL et al. Early life body fatness and risk of colorectal cancer in u.s. Women and men-results from two large cohort studies. Cancer epidemiology, biomarkers & prevention : a publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology. 2015. PMID 25777804. DOI: 10.1158/1055-9965.EPI-14-0909-T. Access: abstract_metadata. Study: PMID25777804.

49. Win AK, Dowty JG, English DR et al. Body mass index in early adulthood and colorectal cancer risk for carriers and non-carriers of germline mutations in DNA mismatch repair genes. British journal of cancer. 2011. PMID 21559014. DOI: 10.1038/bjc.2011.172. Access: abstract_metadata. Study: PMID21559014.

50. Célind J, Ohlsson C, Bygdell M et al. Childhood Body Mass Index Is Associated with Risk of Adult Colon Cancer in Men: An Association Modulated by Pubertal Change in Body Mass Index. Cancer epidemiology, biomarkers & prevention : a publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology. 2019. PMID 30842130. DOI: 10.1158/1055-9965.EPI-18-1077. Access: abstract_metadata. Study: PMID30842130.

51. Levi Z, Kark JD, Katz LH et al. Adolescent body mass index and risk of colon and rectal cancer in a cohort of 1.79 million Israeli men and women: A population-based study. Cancer. 2017. PMID 28736986. DOI: 10.1002/cncr.30819. Access: abstract_metadata. Study: Israeli-adolescent-recruitment-cohort.

52. Levi Z, Kark JD, Barchana M et al. Measured body mass index in adolescence and the incidence of colorectal cancer in a cohort of 1.1 million males. Cancer epidemiology, biomarkers & prevention : a publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology. 2011. PMID 22056504. DOI: 10.1158/1055-9965.EPI-11-0531. Access: abstract_metadata. Study: Israeli-adolescent-recruitment-cohort.

53. Papadimitriou N, Murphy N, Jenab M et al. Body mass index at birth and early life and colorectal cancer: A two-sample Mendelian randomization analysis in European and East Asian genetic similarity populations. Pediatric obesity. 2025. PMID 39587448. DOI: 10.1111/ijpo.13186. Access: abstract_metadata. Study: PMID39587448.

54. Gao C, Patel CJ, Michailidou K et al. Mendelian randomization study of adiposity-related traits and risk of breast, ovarian, prostate, lung and colorectal cancer. International journal of epidemiology. 2016. PMID 27427428. DOI: 10.1093/ije/dyw129. Access: abstract_metadata. Study: PMID27427428.

55. Pleguezuelos-Manzano C, Puschhof J, Rosendahl Huber A et al. Mutational signature in colorectal cancer caused by genotoxic pks+ E. coli. Nature. 2020. PMID 32106218. DOI: 10.1038/s41586-020-2080-8. Access: abstract_metadata. Study: PMID32106218.

56. Jans M, Kolata M, Blancke G et al. Colibactin-driven colon cancer requires adhesin-mediated epithelial binding. Nature. 2024. PMID 39506107. DOI: 10.1038/s41586-024-08135-z. Access: abstract_metadata. Study: PMID39506107.

57. Díaz-Gay M, Dos Santos W, Moody S et al. Geographic and age variations in mutational processes in colorectal cancer. Nature. 2025. PMID 40267983. DOI: 10.1038/s41586-025-09025-8. Access: full_text_partial. Study: PMID40267983.

58. Pearlman R, Frankel WL, Swanson B et al. Prevalence and Spectrum of Germline Cancer Susceptibility Gene Mutations Among Patients With Early-Onset Colorectal Cancer. JAMA oncology. 2017. PMID 27978560. DOI: 10.1001/jamaoncol.2016.5194. Access: abstract_metadata. Study: PMID27978560.

Source-specific evidence cards

[1] Population-Level Trends in Lifestyle Factors and Early-Onset Breast, Colorectal, and Uterine Cancers.

Shows: A US ecological analysis found rising obesity prevalence correlated with early-onset colorectal and other cancer incidence among women.

Does not show: Population-level temporal correlation cannot identify individual causal exposure, distinguish shared calendar trends or attribute most incidence growth to adiposity.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC3.

[2] Distinct intratumoral microbiome of young-onset and average-onset colorectal cancer.

Shows: Tumor and adjacent-tissue sequencing in 136 young-onset and 140 older-onset cases identified differing microbial profiles related to site, stage and obesity.

Does not show: Samples collected after cancer exists cannot establish preceding microbial causation; anatomy, stage, contamination and host response may explain differences.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

[3] Association of Obesity With Risk of Early-Onset Colorectal Cancer Among Women.

Shows: Prospective NHSII follow-up of 85256 women documented 114 cancers before 50; obesity and BMI dose increments were associated with greater risk, including analyses without recent endoscopy or family history.

Does not show: Self-reported exposure, few events, residual confounding and female-only enrollment limit causal and broad-scope inference; a linked correction is attached to this source.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. ErratumIn: PMID30844032. Subclaims: SC1, SC2.

[4] Obesity Is Associated with Early Onset of Gastrointestinal Cancers in California.

Shows: California linked cancer-discharge data associated coded obesity and morbid obesity with younger diagnosis age, including colorectal cancer.

Does not show: A cross-sectional cancer-case analysis with coded obesity and adjustment does not establish prediagnostic exposure or a population incidence causal effect.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2.

[5] Ultraprocessed Food Consumption and Risk of Early-Onset Colorectal Cancer Precursors Among Women.

Shows: NHSII prospective endoscopy-based analysis associated higher ultraprocessed-food intake with conventional adenomas before 50, but not serrated lesions, after multiple adjustments.

Does not show: Adenomas are precursors rather than incident cancer; diet measurement, endoscopy selection and residual confounding do not individually establish processed-food causality.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

[6] Global, regional, and national burden of early-onset and late-onset colorectal cancer attributable to high body-mass index from 1990 to 2021: a trend analysis and forecasts up to 2040 based on the global burden of disease study 2021.

Shows: A GBD 2021 attributable-risk model estimated increasing high-BMI-attributed early-onset CRC deaths and disability burden across time and geography.

Does not show: The attribution assumes an exposure-risk model rather than independently testing causation; counts and demographic decomposition do not establish that adiposity explains most incident-case increase.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3.

[7] Global, regional, and national burden of early-onset colorectal cancer attributable to high BMI (1990-2021).

Shows: A separate GBD 2021 model estimated increasing BMI-attributed early-onset CRC disability burden and sex/geographic disparities.

Does not show: Shared GBD data and population-attributable-fraction assumptions do not create an independent causal experiment or measure most of the underlying incidence trend.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3.

[8] A prospective investigation of early-onset colorectal cancer risk factors-pooled analysis of three large-scale European cohorts.

Shows: Three pooled European cohorts associated higher BMI with CRC diagnosed before 55, especially male colon cancer; effects were weaker for women and rectal cancers.

Does not show: The abstract's early-onset threshold is 55 rather than registered 50; cohort adjustment and sex/site heterogeneity do not prove the precise under 50 causal scope.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. Selected body sections: 1:Methods:paragraph0, 1:Methods:paragraph2, 1:Methods:paragraph4, 1:Methods:paragraph5, 1:Methods:paragraph6, 2:Results:paragraph0, 2:Results:paragraph3, 2:Results:paragraph4. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2, SC3.

[9] Risk Factors for Early-onset Sporadic Colorectal Cancer in Male Veterans.

Shows: A male-veteran case-control analysis of 600 sporadic cancers at 35-49 identified BMI and other factors in validated multivariable risk models using clinic and colonoscopy controls.

Does not show: Risk-model discrimination is not causal identification; measurements only 6-18 months before diagnosis remain vulnerable to occult-disease change and selection.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2, SC3.

[10] Risk factors for early-onset colorectal cancer: A large-scale Chinese cohort study.

Shows: A prospective Chinese cohort of adults initially 30-50 recorded 222 early-onset CRC cases and associations with BMI, alcohol, hypertension, diabetes and family history.

Does not show: Covariate adjustment does not isolate adiposity from metabolic or behavioral exposures; event-age definition and borderline BMI precision need retained limits.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2, SC4.

[11] Risk Factors Associated With Early-Onset Colorectal Neoplasm in Chinese Youth: A Prospective Population-Based Study.

Shows: A prospective Shanghai screening cohort under 50 identified 124 neoplasms, including only six CRCs; BMI was associated in multivariable analysis.

Does not show: Most endpoints were adenomas, and spline overall p=0.19 does not show a significant BMI dose-response; six cancers cannot establish a precise cancer-incidence causal effect.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2.

[12] Diet and Lifestyle Habits in Early-Onset Colorectal Cancer: A Pilot Case-Control Study.

Shows: A small Italian case-control study associated meat, dairy and smoking with EOCRC but found no significant difference in other examined variables.

Does not show: Retrospective diet over five years,47 cases, hospital controls and multiple comparisons limit causality and make BMI null evidence imprecise.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC4.

[13] Association between lifestyle and risk of early-onset cancer: Evidence from the European Prospective Investigation into Cancer and Nutrition and the UK Biobank.

Shows: EPIC/UKBiobank analyses associated selected smoking, BMI and activity scores with pooled lifestyle-related cancers before 50, with different sex and breast-cancer results.

Does not show: Pooled multi-cancer endpoints do not supply a colorectal-specific BMI effect; cohort overlap and score construction do not isolate adiposity causality.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC3, SC4.

[14] Adolescent to adulthood weight trajectories and the risk of obesity-related cancers, overall and early-onset: a population-based cohort study.

Shows: Measured adolescent-to-adult BMI trajectories in 800024 participants associated persistent or adult high BMI with obesity-related cancers, including pooled early-onset outcomes; colorectal risk rose with weight gain.

Does not show: The abstract separates pooled early-onset estimates from all-age colorectal estimates; observational weight-loss categories are not randomized prevention and can reflect illness or selection.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2, SC3.

[15] Regional patterns of early-onset colorectal cancer from the GEOCODE (Global Early-Onset COlorectal Cancer DatabasE)-European consortium: retrospective cohort study.

Shows: A six-country series of 851 nonhereditary EOCRC patients showed geographical differences in BMI, age, stage and family history.

Does not show: A case-only series has no healthy population denominator and cannot estimate adiposity risk or explain population incidence increases.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC3.

[16] Body mass index (BMI) trajectories and risk of colorectal cancer in the PLCO cohort.

Shows: PLCO cohort analysis associated recalled earlier-life overweight and adult weight gain with subsequent colorectal cancer risk.

Does not show: These are primarily later-diagnosed cancers in a screening cohort; early exposure does not equal cancer diagnosis before 50, and recall/confounding remain.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2.

[17] Traditional Risk Factors Are Not Associated with the Rise of Early-Onset Colorectal Cancer: An All of Us Study.

Shows: All of Us matched case-control analysis included 485 cancers before 50 and found no age-specific BMI or alcohol effects and no common-variant age interaction after multiple testing.

Does not show: No age interaction is not proof that BMI has no causal effect at either age; retrospective exposure, ancestry adjustment and limited rare-effect power constrain inference about trends.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2, SC3.

[18] Historical Obesity and Early-Onset Cancers: A Propensity-Weighted Analysis of the National Health and Nutrition Examination Survey.

Shows: NHANES historical-BMI comparisons found more prior obesity among weighted early-onset than later-onset CRC cases.

Does not show: The millions are survey-weighted population estimates, not millions of examined patients; survivor recall, case-only age comparison and propensity weighting do not identify a causal incidence effect.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2, SC3.

[19] Risk Factors for Early-Onset Colorectal Cancer: A Nested Case‒Control Study within the Korean National Health Insurance Service‒Health Screening Cohort.

Shows: A Korean nested case-control study of 285 cancers at 40-49 associated high BMI with risk and estimated modest sex-specific BMI attributable fractions.

Does not show: Calculated fractions assume causal validity and local prevalence; observational estimates do not show that BMI accounts for most global growth or isolated obesity prevention benefit.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2, SC3.

[20] Obesity-Facilitated Colon Cancer Progression Is Mediated by Increased Diacylglycerol O-Acyltransferases 1 and 2 Levels.

Shows: DGAT inhibition reduced colon tumors in high-fat-fed ApcMin mice and growth in colon-cancer systems, linking lipid metabolism to obesity-associated tumor biology.

Does not show: Genetic susceptibility, diet composition, inhibitors and tumor-progression models do not isolate excess adiposity as the cause of sporadic human cancer before 50.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[21] Body Mass Index and Risk of Colorectal Cancer Incidence and Mortality in Asia.

Shows: Seventeen pooled Asian cohorts showed BMI dose-related CRC incidence and mortality associations, stronger for colon and male mortality.

Does not show: Most participants and outcomes were older; all-age association and sex/site heterogeneity do not directly identify under 50 adiposity causality.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2.

[22] Mechanisms linking obesity to altered metabolism in mice colon carcinogenesis.

Shows: High-fat-fed mice bearing injected colon-cancer cells developed larger tumors; adipose-conditioned media and leptin altered cell metabolism and signaling.

Does not show: Transplanted tumor growth and ex-vivo crosstalk do not establish endogenous cancer initiation or isolate adiposity from dietary exposure in EOCRC.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[23] Adiposity, metabolites, and colorectal cancer risk: Mendelian randomization study.

Shows: Large sex/site-specific MR analyses linked genetically predicted BMI or waist-hip ratio with all-age CRC, with differing sex effects; measured metabolite classes did not explain the associations.

Does not show: Genetic instruments reflect longer-term adiposity and require exclusion/pleiotropy assumptions; no under 50 endpoint is identified in the abstract, and mediator nonattenuation is not pathway exclusion.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. Selected body sections: 1:Methods:paragraph0, 1:Methods:paragraph1, 1:Methods:paragraph2, 1:Methods:paragraph4, 1:Methods:paragraph5, 1:Methods:paragraph7, 2:Results:paragraph0, 2:Results:paragraph2, 2:Results:paragraph8. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[24] Mutational landscape of intestinal crypt cells after long-term in vivo exposure to high fat diet.

Shows: Mouse high-fat-diet exposure over 48 weeks did not increase intestinal-crypt genomic instability without additional activating mutation or chemical exposure.

Does not show: This tests a mutational mechanism rather than all carcinogenic routes; diet-induced obesity and organoid selection do not prove adiposity cannot promote tumors by other mechanisms.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[25] Obesity promotes conserved inflammatory and metabolic transcriptional programs in colon tumors: evidence from mouse models and the ColoCare Study Patient Cohort.

Shows: Diet-induced obesity reduced survival in an orthotopic transplanted mouse colon-tumor model and showed conserved inflammatory/metabolic transcriptional patterns with older human tumors/adipose samples.

Does not show: Transplantation, progression and gene signatures do not establish spontaneous early-onset human initiation or distinguish diet composition from adiposity.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. UpdateOf: PMID40667217. Subclaims: SC2.

[26] Aldose Reductase Inhibitor, Fidarestat Prevents High-fat Diet-induced Intestinal Polyps in ApcMin/+ Mice.

Shows: Fidarestat reduced high-fat-diet-associated intestinal polyp number and size in genetically predisposed ApcMin mice.

Does not show: Drug specificity, genotype and many small-intestinal polyps limit translation; prevention in this model is not adiposity reduction preventing sporadic under 50 human CRC.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[27] Relationship between obesity-related colorectal tumors and the intestinal microbiome: an animal-based trial.

Shows: Obese/diabetic KKAy mice retained more chemically induced tumors than wild type despite microbiome changes from antibiotics or cohousing; antibiotic suppression occurred only in wild type.

Does not show: Different genetic backgrounds, diabetes and carcinogen exposure confound adiposity isolation; failure of this microbial intervention does not exclude every organism or human causal pathway.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC4.

[28] Integrated metagenomic and metabolomic analysis reveals distinct gut-microbiome-derived phenotypes in early-onset colorectal cancer.

Shows: EOCRC/late-onset case-control multiomics with validation identified distinct microbial/metabolite patterns and diagnostic classifiers.

Does not show: Postdiagnosis dysbiosis and classifier accuracy do not establish temporally preceding microbial or meat-induced EOCRC causality.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

[29] A Comprehensive Comparison of Early-Onset and Average-Onset Colorectal Cancers.

Shows: Comparative clinical/genomic analysis found young tumors more often left-sided; apparent pathway differences diminished after sidedness adjustment, while youngest cases had more pathogenic germline variants.

Does not show: Case-series genomic similarity or disparity does not identify environmental causal exposure or establish that heredity explains the incidence trend.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3, SC4.

[30] Increase of early-onset colorectal cancer: a cohort effect.

Shows: Long-term registry analyses in four countries found increasing EOCRC in successive birth cohorts and particularly young ages.

Does not show: Age-period-cohort patterns show trends but do not identify adiposity or another specific cause; shared detection and cohort exposures remain alternatives.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3.

[31] Patterns in genomic mutations among patients with early-onset colorectal cancer: an international, multicohort, observational study.

Shows: International tumor sequencing across seven cohorts found age-related mutation patterns depending on hypermutation group and ancestry/site composition.

Does not show: Tumor mutations after diagnosis are not exposure instruments or causes of changing population incidence; cohort overlap and tumor selection limit independent etiologic inference.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3, SC4.

[32] Clinical and biological landscape of constitutional mismatch-repair deficiency syndrome: an International Replication Repair Deficiency Consortium cohort study.

Shows: A constitutional mismatch-repair-deficiency cohort showed very high childhood cancer burden, genotype-dependent onset and characteristic mutation signatures.

Does not show: A rare inherited syndrome provides a real alternative cause but cannot be transferred to sporadic adult EOCRC or explain most secular incidence growth.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3.

[33] Modeling Early-Onset Cancer Kinetics Reveals Changes in Underlying Risk and the Impact of Population Screening.

Shows: A multistage model incorporating tumor size at diagnosis supported both detection changes and accelerated underlying early-onset carcinogenic processes across cohorts.

Does not show: Model identifiability and fit do not experimentally identify adiposity or partition its specific contribution; model results are not new independent patients or causal perturbations.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. UpdateOf: PMID39677417. Subclaims: SC3.

[34] Identification of specific susceptibility loci for the early-onset colorectal cancer.

Shows: Parallel CRC/EOCRC GWAS identified susceptibility loci and a validated early-onset polygenic risk score, with allele-specific POLA 2 expression analysis.

Does not show: Disease susceptibility GWAS is not MR of adiposity; associated variants and functional expression do not identify a weight-mediated causal effect.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

[35] Epidemiological Classification Changes and Incidence of Early-Onset Colorectal Cancer.

Shows: French registries showed neuroendocrine reclassification materially influenced apparent young-age CRC trends, while metastatic adenocarcinoma incidence also increased.

Does not show: A regional coding natural change demonstrates a local ascertainment contribution, not that all global EOCRC growth is artifact or that adiposity lacks a causal role.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3.

[36] Epigenetic Modulation, Intratumoral Microbiome, and Immunity in Early-Onset Colorectal Cancer.

Shows: TCGA/ORIEN analyses found age-related methylation and immune-microbial correlation patterns, with no consistent young-versus-older microbial difference.

Does not show: Epigenetic clocks and postdiagnosis deconvolution are biomarkers rather than causal aging or microbial exposure evidence; platform and contamination biases remain.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. UpdateOf: PMID40236026. Subclaims: SC4.

[37] The Sulfur Microbial Diet Is Associated With Increased Risk of Early-Onset Colorectal Cancer Precursors.

Shows: NHSII sulfur-microbial diet scores were associated with adenomas before 50, particularly higher-risk histology, but not serrated lesions or clearly older-age adenomas.

Does not show: Dietary index and precursor association do not measure microbial carcinogen exposure directly or establish an individual processed-food/microbe causal claim.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

[38] Somatic Mutation Profiles of Colorectal Cancer by Birth Cohort.

Shows: A 369-patient somatic sequencing series found no significant mutation/pathway differences by birth cohort after adjustment.

Does not show: Small tumor series and selected sequencing cannot exclude rare or epigenetic mechanisms; absence of genomic differences does not identify adiposity as the secular cause.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3, SC4.

[39] Epidemiology of Early-onset Colorectal Cancer in a Safety-net Hospital System in Texas.

Shows: A Texas safety-net EOCRC case series found predominantly advanced-stage diagnoses, often within three months of reported symptom onset.

Does not show: Selection into one hospital system and stage among existing cases do not estimate population incidence, obesity causality or the benefit of screening.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3.

[40] Metabolic syndrome, metabolic comorbid conditions and risk of early-onset colorectal cancer.

Shows: MarketScan nested case-control analysis associated metabolic syndrome and accumulated metabolic conditions with cancers before 50, more clearly for colon than rectum.

Does not show: Claims-coded metabolic clusters do not isolate adiposity from diabetes or other components, and adjusted odds do not explain the proportion of incidence growth.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2, SC3.

[41] Risk Factors Associated With Early-Onset Colorectal Cancer.

Shows: A tertiary-hospital case-control analysis of 269 EOCRC cases found obesity, smoking and diabetes prevalence similar to controls, while family history and other factors were associated.

Does not show: Hospital control selection, contemporaneous measurements and limited cases make an imprecise BMI null; this does not prove adiposity never contributes.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2, SC3.

[42] Risk factors for early-onset colorectal cancer: a population-based case-control study in Ontario, Canada.

Shows: Ontario population-based case-control analysis associated sugary drinks, Western diet and sedentary time with EOCRC among 175 cases and 253 controls.

Does not show: Recall, correlated diet/adiposity exposures, participation and multiple estimates limit individual-exposure causal interpretation.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

[43] Antibiotic Therapy and Risk of Early-Onset Colorectal Cancer: A National Case-Control Study.

Shows: Swedish prescription/pathology-linked analysis of 2557 EOCRC cases found no conclusive overall prior-antibiotic association after adjustment; broad-spectrum exposure had a modest positive estimate.

Does not show: Confounding by infection/indication, short lag and prescription capture remain; the overall near-null does not exclude molecule- or timing-specific effects.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

[44] Obesity and younger versus older onset colorectal cancer in the United States, 1998-2017.

Shows: US survey analysis associated obesity with CRC diagnosis in 18-49-year-olds, with a significant age/BMI interaction.

Does not show: Survey prevalence and measured BMI around reported diagnosis do not establish temporally prior incidence or randomized exposure; survivor and reporting bias remain.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2.

[45] Associations of Body Mass Index at Different Ages With Early-Onset Colorectal Cancer.

Shows: German population-based case-control data associated recalled obesity at 20,30 and roughly ten years before diagnosis with CRC diagnosed before 55, particularly without prior colonoscopy.

Does not show: The endpoint uses 55 rather than 50, recalled weight and colonoscopy selection limit causality, and no adiposity intervention was tested.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2.

[46] Childhood body mass index and height in relation to site-specific risks of colorectal cancers in adult life.

Shows: Measured childhood BMI in 257623 Danish children was associated with subsequent adult colon cancer but not rectal cancer.

Does not show: Early-life exposure is prospectively established, but adult outcomes are not restricted to diagnosis before 50; growth, later BMI and correlated childhood factors may mediate or confound.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2.

[47] Adolescent body mass index and erythrocyte sedimentation rate in relation to colorectal cancer risk.

Shows: Measured adolescent BMI and inflammatory ESR in 239658 Swedish men predicted subsequent colorectal cancer over decades.

Does not show: Male conscription selection, long follow-up into later ages and inflammation/weight correlations constrain exact EOCRC and isolated-adiposity causality.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2.

[48] Early life body fatness and risk of colorectal cancer in u.s. Women and men-results from two large cohort studies.

Shows: Two US cohorts associated recalled early-life body fatness more strongly with later CRC in women than men after adult-BMI adjustment.

Does not show: Childhood recall, sex heterogeneity, overadjustment for a potential mediator and mostly later-life endpoints limit direct pre 50 causal inference.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2.

[49] Body mass index in early adulthood and colorectal cancer risk for carriers and non-carriers of germline mutations in DNA mismatch repair genes.

Shows: ColonCancerFamilyRegistry analysis associated BMI at 20 with subsequent CRC among mismatch-repair mutation carriers and noncarriers.

Does not show: Recalled weight, weighted familial ascertainment and limited noncarrier events do not establish sporadic EOCRC causality or transfer hereditary risk as an adiposity instrument.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2, SC3.

[50] Childhood Body Mass Index Is Associated with Risk of Adult Colon Cancer in Men: An Association Modulated by Pubertal Change in Body Mass Index.

Shows: Swedish measured-BMI follow-up associated childhood BMI with adult colon risk primarily when followed by above-median pubertal gain; rectal findings were null.

Does not show: A measured antecedent interaction is a useful boundary but remains observational, and adult cancer endpoints are not restricted to before 50.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2.

[51] Adolescent body mass index and risk of colon and rectal cancer in a cohort of 1.79 million Israeli men and women: A population-based study.

Shows: Israeli pre-recruitment cohort of 1.79 million men and women associated adolescent overweight with colon cancer and higher obesity with selected rectal outcomes.

Does not show: Registry-linked measured exposure strengthens timing, but later attained ages, sex/site differences and overlapping earlier cohort reports do not independently isolate causality.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2.

[52] Measured body mass index in adolescence and the incidence of colorectal cancer in a cohort of 1.1 million males.

Shows: An earlier Israeli male cohort associated measured adolescent overweight with subsequent colon but not rectal cancer in young-to-middle adulthood.

Does not show: It overlaps the later larger Israeli report; follow-up into middle age and residual socioeconomic/dietary confounding prevent counting a separate causal line.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2.

[53] Body mass index at birth and early life and colorectal cancer: A two-sample Mendelian randomization analysis in European and East Asian genetic similarity populations.

Shows: Two-sample MR using BMI from birth through 8 years found little evidence of all-age CRC association in European and East Asian genetic-similarity populations.

Does not show: The null childhood instrument differs from adult/lifelong adiposity and does not negate later-adult BMI causal effects; it also does not directly test cancer diagnosed before 50.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[54] Mendelian randomization study of adiposity-related traits and risk of breast, ovarian, prostate, lung and colorectal cancer.

Shows: Cancer-consortium MR associated genetically predicted adult BMI with higher all-age colorectal cancer odds.

Does not show: Instrument relevance, pleiotropy and lifelong exposure assumptions remain, and the outcome is not specifically EOCRC or a timed weight-loss intervention.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[55] Mutational signature in colorectal cancer caused by genotoxic pks+ E. coli.

Shows: Repeated pks-positive E. coli exposure caused a characteristic mutation signature in human intestinal organoids absent with isogenic pks-mutant controls; matching signatures occurred in human CRC genomes.

Does not show: The experiment establishes organism-specific mutagenesis in an engineered exposure system, not the fraction of EOCRC caused by microbes or proof that antibiotics/processed foods cause cancer.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

[56] Colibactin-driven colon cancer requires adhesin-mediated epithelial binding.

Shows: Adhesin manipulation and FimH inhibition attenuated colibactin genotoxicity and CRC exacerbation in a predisposed transgenic mouse model.

Does not show: Engineered microbiota dependence and one model constrain spontaneous human EOCRC transfer; a linked correction must remain attached rather than a separate report count.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. ErratumIn: PMID41652773. Subclaims: SC4.

[57] Geographic and age variations in mutational processes in colorectal cancer.

Shows: Sequencing 981 international CRC genomes found colibactin signatures enriched in younger cancers and imprinted early, including associated APC driver indels.

Does not show: Mutation timing is inferred from tumor evolution, not a directly measured childhood exposure; signatures were not associated with current pks-positive bacteria, and case-only genomes do not estimate population causal incidence or most growth.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. Selected body sections: Methods: Recruitment; Regressions; Evolutionary analysis, Results: Study design; Molecular classification; Enrichment of colibactin signatures; Colibactin is an early mutagenic event. No complete-paper or supplement audit claimed. UpdateOf: PMID40034755. Subclaims: SC3, SC4.

[58] Prevalence and Spectrum of Germline Cancer Susceptibility Gene Mutations Among Patients With Early-Onset Colorectal Cancer.

Shows: Prospective accrual of 450 unselected cancers before 50 identified pathogenic germline susceptibility mutations in 16 percent of cases, including distinct syndrome groups.

Does not show: A germline prevalence among cases is not an adiposity instrument, general-population attributable fraction or explanation of the secular increase.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3.

Documentary record

Every grade on this site is reproducible from its record. The files below are the posted edition exactly as assessed.

• ADVERSARIAL_REVIEW.md
• CONSENSUS_CROSSCHECK.html
• CONSENSUS_CROSSCHECK.md
• RESEARCH_TIMELINE.html
• RESEARCH_TIMELINE.md
• SHA256.txt
• adiposity-early-colorectal-cancer.review.md
• adiposity-early-colorectal-cancer.scorecard.json
• adversarial-findings.json
• codex-correction-provenance.json
• consensus-validation.json
• edition-approval-audit.json
• reader-summary.json
• research-timeline.json
• validation-report.json
• validation-round2-2026-10-14.json
• validation.txt
• video-insert.md
• web-page.md

Sources: 58 · load-bearing claims: 63 · verified: 63 · unverified: 0 · contradicted: 0 · schema 5.3