In plain terms

The claim being assessed: Aerobic glycolysis is a primary causal driver of malignant transformation and sustained tumour growth rather than a metabolic consequence of upstream oncogenic signalling.

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: Primary-subclaim SC3 returns Γ. Cell-level feedback and strong context-specific metabolic contributions make the hypothesis more than association, but no examined evidence set establishes general upstream metabolic primacy across tumour classes. SC2 remains Γ for the registered broad state-necessity claim; named endogenous dependencies reach Ε as a competing model. Weakest-link sensitivity is Β due to descriptive SC1; primary and scope-matched sensitivities are Γ. The real Warburg observation is not labelled refuted.

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 for Warburg Effect

Claim being assessed: Aerobic glycolysis is a primary causal driver of malignant transformation and sustained tumour growth rather than a metabolic consequence of upstream oncogenic signalling.. This review asks whether the proposed factor causes the health outcome—not just whether the two appear together. For the wording assessed here, the evidence is rated Plausible mechanism (HSEF Γ; Moderate confidence). There is a believable biological story and partial support, but the causal step has not been demonstrated. That rating applies to this claim and scope; it does not mean the biological process is irrelevant in every setting.

Implications for health and treatment: This is an assessment of a causal explanation, not proof that a particular supplement, diet, test, or treatment improves health. A treatment claim needs studies that test the intervention in people and measure meaningful benefits and harms. These findings can help frame research and informed clinical questions, but they do not identify what is right for one person.

Limits for individual care: Evidence from cells, animals, associations, or a plausible mechanism may not predict what happens in people. The grade is limited to the exact claim, outcomes, and sources reviewed, and it is not a diagnosis or an individual forecast. Do not start, stop, or change care based on this page; discuss personal decisions with a qualified clinician.

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.

This assessment rests on 68 sources and 72 individually checked claims. 72 of those claims were verified against the original paper.

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

ID	Sub-claim	Type	Level	Pivotal card
SC1	Many cancer cells exhibit preferential aerobic glycolysis.	existence	Β	—
SC2	The aerobic glycolytic state is necessary for tumour growth.	necessity	Γ	Card ↗
SC3	Aerobic glycolysis is a primary upstream driver rather than a consequence of oncogenic signalling.	primacy	Γ	Card ↗

Visual Evidence Cards · Pivotal Subclaims

Pivotal subclaim illustrations

Dark-theme simulated-micrograph renders and concise causal evidence summaries for the pivotal subclaims that govern this review's grade. Click any card to inspect the master 2400×2400 render.

Master 2400×2400 render ↗

Master 2400×2400 render ↗

Figures

Explanatory figures summarizing this review. They carry no evidential weight of their own; where a figure and the review differ, the review governs.

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: 68 primary reports; claims: 72; verified: 72; unverified: 0. Verification is bounded attribution. Access: 7 partial-body and 61 complete abstract/status examinations. Cutoff: 2026-10-05. No complete-paper audit or blinded calibration claimed. Adversarial review: Generated by Sol 6.1; adversarially reviewed by Claude Opus 5.5 (Anthropic) on 2026-10-06. 1 correction applied (see ADVERSARIAL REVIEW AND APPROVAL). Status: Approved as revised.

AT A GLANCE

Aerobic glycolysis is a real, heterogeneous tumour phenotype and can support specific cancer dependencies. The evidence does not establish it as the general primary event upstream of oncogenic signalling.

Deciding experiment: In prespecified autochthonous KRAS, APC and mammary tumour models, independently manipulate cytosolic NADH recycling and lactate fermentation at physiological exposure while preserving glucose uptake, PPP/hexosamine carbon routing and enzyme scaffolding. Apply perturbations before initiation and separately after tumour establishment, with orthogonal flux rescue, matched oncogene signalling and lineage-resolved metastasis/relapse follow-up.

Belief versus evidence: Primary-subclaim SC3 returns Γ. Cell-level feedback and strong context-specific metabolic contributions make the hypothesis more than association, but no examined evidence set establishes general upstream metabolic primacy across tumour classes. SC2 remains Γ for the registered broad state-necessity claim; named endogenous dependencies reach Ε as a competing model. Weakest-link sensitivity is Β due to descriptive SC1; primary and scope-matched sensitivities are Γ. The real Warburg observation is not labelled refuted.

Best-supported rival: Oncogene-driven reprogramming with feedback. The strongest separator is inducible KRAS maintenance control [57] versus GLUT3 culture sufficiency [37]. The endogenous HFSC SCC LDHA null [67], with reduced metabolic and FDG signals but preserved tumours, tests the nominated fermentation state more directly than PKM2 isoform deletion alone.

COMPETING MECHANISTIC MODELS

Rank 1: Oncogene-driven reprogramming with feedback (Ε). Oncogenic lesions induce metabolic programs that subsequently support tumour behaviour. Inducible KRAS dependence, KRAS allele-dose progression, direct KRAS4A-HK1 interaction and KRAS-NIX/p38gamma effectors support directionality [31,35,39–41,57]. Oncogenes are pleiotropic and metabolic feedback can also activate signalling; this is not an exclusively one-way model.

Rank 2: Context-specific respiration and biosynthetic outputs (Ε). Mitochondrial redox and aspartate/asparagine output can constrain proliferation and persistence. TFAM loss, aspartate rescue, engineered bypass and selected relapse-state dependencies narrow outputs and contexts [21–24,28,30,38]. Respiration can be bypassed in cells or by environmental nutrients; aspartate was not limiting in every tumour [28].

Rank 3: Metabolic symbiosis and flexible fuel use (Ε). Tumour compartments use lactate and other fuels according to oxygen, stage and nutrient access. Human lactate tracing and mouse uptake loss identify a fuel route; endogenous SCC loss and double loss map compensation [49–51,58,66–67]. Tracing proves use, not necessity; transporter inhibition has alternative effects and full human lineage-resolved dependence remains incomplete.

Rank 4: Proliferation-related redox constraint (Γ). NAD regeneration and shuttle capacity can cause lactate fermentation without a unique cancer-initiating role. PDH/redox rescue and NADH-shuttle modulation separate fermentation rate from proliferation [63–64]. The strongest direct tests are cultured cells; autochthonous long-term stage-specific generalization remains open.

Rank 5: Glycolysis as primary upstream driver (Γ). Aerobic fermentation starts and generally maintains malignant state before oncogenic signalling. GLUT3 culture transformation, YAP feedback and named glycolytic dependencies are its strongest support [10–11,37,44,46]. Overexpression, enzyme multifunctionality, endogenous fermentation nulls and oncogene-to-flux direction block full-scope primacy [45–48,57,64–67].

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

Aerobic glycolysis is a primary causal driver of malignant transformation and sustained tumour growth rather than a metabolic consequence of upstream oncogenic signalling.

The index designates primary-subclaim aggregation with SC3 gating primacy. The real metabolic observation SC1 cannot be labelled refuted because the stronger causal-driver claim fails.

Selection: public-interest rank 8 from the editorial ranking compiled 2026-09-23; categorical prior verdicts were visible. Private fresh review; earlier documents are preserved.

Evidence and causal interpretation

Aerobic glycolysis means lactate fermentation despite available oxygen. It does not mean that oxygen is absent, mitochondria are inactive or every tumour obtains most ATP from fermentation. The registered hypothesis adds primacy to that observation. These burdens must be separated, as must initiation, maintenance, metastasis and relapse.

Human brain and lung tracing confirms complex carbon fates: glucose enters lactate, TCA intermediates, anaplerosis and biosynthetic products [49–56]. Human lactate infusion and mouse MCT1 loss show tumour-cell lactate uptake [49], while acetate provides another brain-tumour fuel [52]. These findings reject a simple inactive-mitochondria account without proving that every oxidative route is necessary.

FDG is a transport/phosphorylation proxy, not an assay of lactate flux. In untreated surgical NSCLC, SUV did not track glycolytic-intermediate labelling [54]. Regional perfusion and non-glucose nutrients matter [50]. Mouse absolute-flux work found lower primary-solid-tumour TCA flux and calculated ATP synthesis despite increased glycolysis, with different metastasis behaviour [33]. Label enrichment, absolute flux and calculated ATP must not be conflated.

HK2 supplies an important positive case. Conditional deletion reduced initiation and maintenance in KRAS lung and ErbB2 mammary models [46]. HCC genetics reduced tumour incidence, while WT HK2 but not kinase-dead or mitochondrial-binding-deficient protein restored cellular phenotypes [44]. These results narrow a real dependency, but HK2 supplies PPP, serine and other outputs, and both catalytic and mitochondrial-binding functions matter. Necessity of HK2 is not proof that lactate fermentation is the primary lesion.

PKM2 illustrates the danger of treating isoform identity as metabolic-state identity. PKM2-to-PKM1 replacement impaired xenografts [11], whereas PKM2 deletion accelerated Brca1 mammary tumour formation [47], left APC colon tumour outcomes unchanged [45] and delayed sarcoma onset without preventing outgrowth [48]. Sarcoma lineage reporters assigned remaining PKM2 staining to stroma and confirmed PKM1-positive tumour proliferation. These are stage and lineage boundaries, not proof that all glucose metabolism is unnecessary.

The LDHA SCC experiments directly constrain fermentation necessity. LDHA loss left timing, number and volume unchanged in KRAS/p53, two Cre-defined and chemical carcinogenesis paradigms despite reduced metabolic and FDG signals [67]. A later study showed GLS-only loss could also be bypassed through lactate use, while joint LDHA/GLS deletion prevented tumour formation [66]. The genes target particular routes; neither experiment removes every glucose reaction. Compensation that preserves growth is a genuine failure of single-route necessity, even when combined targeting is effective.

Redox experiments provide another discriminator. Increasing NADH-shuttle capacity lowered lactate production without reducing cell proliferation [64]. PDH activation could inhibit proliferation through NAD+/NADH imbalance, with uncoupling or ATP turnover rescuing growth despite increased glucose oxidation [63]. These are mechanistically strong cell tests, with an explicit limit on long-term endogenous tumour inference. LDHA-inhibited melanoma also retained growth through ATF4/glutamine adaptation [65].

Oncogene-to-metabolism direction has direct support. Inducible KRAS withdrawal changes native PDAC maintenance and anabolic glucose routing [57]. Mutant KRAS allele enrichment changes the metabolic program during lung progression [31]; KRAS4A directly regulates HK1 [39]; NIX and p38gamma connect KRAS to metabolic and growth outputs [40–41]. These data are compatible with feedback, but they do not place fermentation before the driver.

The strongest upstream metabolic steelman is GLUT3-driven glucose uptake in nonmalignant 3D breast cells, which activated signalling and disrupted polarity [37]. YAP/TAZ and HIF feedback reinforce this possibility [10,14]. Overexpression, culture architecture and multifunctional protein interactions prevent those findings from establishing a general endogenous fermentation-first initiating lesion. Oncogene-specific fibroblast models also achieve transformation through distinct metabolic programs [34].

Metastatic and relapse states are separate tests. BACH1-linked glycolysis promoted KRAS lung metastasis [13], while PGC-1alpha-dependent respiration affected metastasis without primary growth [23]. KRAS-ablation survivors, endocrine persisters and recurrent HPV-associated cancers can favour respiration [26,30,62]. A primary tumour phenotype cannot be presumed to remain the rate-limiting state in every surviving or disseminated clone.

Clinical translation is also distinct from causal biology. IACS-010759 preclinical activity [20] was followed by phase-I trials in which lactate elevation and neurotoxicity prevented sustained target exposure and no recommended phase-II dose was established [60]. This is an exposure-limited efficacy test, not clean proof that respiration has no tumour role. The small uncontrolled DCA cohort supplies biomarker and tolerability observations without a reliable treatment-effect estimate [61].

Consensus helped identify shuttle, LDHA bypass and stage-specific gaps and extracted selected indexed full-text details for HK2, PKM2 and Wang. It could not source-validate the Ying or Pathria Methods in the targeted check; those papers were independently retrieved and examined at abstract depth. Original accessible body checks verified APC/sarcoma deletion, HCC rescue, aspartate bypass, mitochondrial editing and the SCC null/flexibility studies. Its predominantly affirmative meter conflicted with its narrative conclusion, demonstrating why neither meter nor screening count is used as an HSEF grade.

Several linked papers carry correction notices, some without supplied PMID details, and additional notices accompany diet, OXPHOS/relapse and p38gamma reports. Exact correction effects were not comprehensively audited; no precise affected numerical claim is used as a decisive premise. Human tracing cohort independence is also unresolved rather than assumed. These limits lower confidence and are retained in individual cards.

HYPOTHESIS STATE EVALUATION FRAMEWORK (HSEF v5.3)

Subclaim matrix and aggregation

Preregistered aggregation: primary-subclaim. Primary-subclaim SC3 returns Γ. Cell-level feedback and strong context-specific metabolic contributions make the hypothesis more than association, but no examined evidence set establishes general upstream metabolic primacy across tumour classes. SC2 remains Γ for the registered broad state-necessity claim; named endogenous dependencies reach Ε as a competing model. Weakest-link sensitivity is Β due to descriptive SC1; primary and scope-matched sensitivities are Γ. The real Warburg observation is not labelled refuted. Sensitivities: weakest-link Β; primary-subclaim Γ; scope-matched Γ.

Decisive-perturbation inventory and causal tally

[3] Independent splice-switching ASOs altered PKM isoforms and inhibited xenograft and genetic HCC tumourigenesis. Splice replacement is not PKM2 deletion; genotype, increased pyruvate-kinase activity and nonmetabolic functions remain relevant.

[4] mTOR-NEAT1-paraspeckle regulation controlled glycolytic splicing and tumour growth in an oncogene-driven model. The initiating oncogene and broad RNA-processing changes precede the metabolic phenotype; selective flux mediation is not fully isolated.

[13] BACH1-linked glycolysis promoted antioxidant-associated metastasis and BACH1 targeting prevented that increase. Metastatic amplification after KRAS-driven cancer is distinct from primary transformation and BACH1 has broader transcriptional functions.

[15] PKM1 supported tumour growth in defined KRAS/carcinogen models and proliferation in pulmonary neuroendocrine cancers. This counters PKM2-specific universality; active pyruvate kinase can support malignancy in particular lineages.

[25] Serine/glycine restriction improved survival in Apc and Myc models, with weaker responses in KRAS-driven cancers. Dietary amino-acid dependence is genotype-specific and distinct from lactate fermentation; linked erratum is retained.

[30] Cells surviving KRAS ablation relied more on mitochondrial respiration and supported relapse. A selected persistence state differs from untreated primary growth. A linked 2026 erratum is retained and its affected details were not audited.

[31] Endogenous mutant allele enrichment accompanied progression and a glycolytic/TCA/glutathione metabolic program. Genotype precedes rewiring, and early versus advanced tumours differ; enrichment is not proof that metabolism independently initiated cancer.

[32] High-dose vitamin C caused redox-linked GAPDH inhibition and impaired growth in Apc/KRAS mice. Pleiotropic oxidative toxicity and genotype selection prevent treating this as a clean test of fermentation-only necessity or clinical efficacy.

[35] COX2 deletion attenuated HFD-driven KRAS hyperactivation, glycolysis and dysplasia. Diet, inflammation and oncogene activation are upstream; COX2 perturbation is not selective glycolytic subtraction.

[38] TFAM loss reduced KRAS-driven tumorigenesis; mitochondrial ROS and glutamine metabolism supported transformation. Mitochondrial output necessity in this model contradicts a simple defective-respiration account without making it universal.

[40] Nix loss delayed PDAC progression and improved survival, while later mitochondrial-clearance compensation allowed progression. The KRAS-NIX program is upstream of altered flux; delay is not permanent abrogation or generic metabolic sufficiency.

[41] Pancreatic p38gamma loss reduced glycolytic intermediates/signalling and PDAC, with PFKFB3 pathway epistasis. The kinase has broader signalling roles and sits downstream of KRAS. Linked erratum PMID39957350 limits unaudited details.

[44] Liver HK2 deletion reduced tumour incidence; WT but not kinase-dead or mitochondrial-binding-deficient HK2 restored cell phenotypes. Both catalysis and mitochondrial binding were required in the rescue. Reduced glycolysis coexisted with maintained TCA flux; linked correction retained.

[46] HK2 deletion reduced initiation and maintenance despite HK1 expression, and systemic deletion was therapeutic in mice. A first glucose-phosphorylation step supports multiple branches and enzyme functions; this does not prove primary fermentation initiation. Linked correction retained.

[48] PKM2 loss delayed onset but tumours eventually grew; lineage reporting assigned residual PKM2 to stroma and showed PKM1-positive proliferation. Facilitated initiation is distinct from maintenance necessity; bulk expression mixes tumour and stromal compartments.

[53] Glo1 ablation increased reactive-metabolite sensitivity and reduced mouse tumour growth. Detoxification of a glycolytic byproduct is a consequence vulnerability, not proof of glycolysis-first oncogenesis.

[57] KRAS withdrawal established maintenance dependence and controlled glucose routing into anabolic pathways. This places a driver oncogene upstream; metabolic branches and KRAS pleiotropy are not isolated by withdrawal alone.

[66] GLS loss alone preserved tumourigenesis with lactate-use compensation; combined LDHA/GLS loss prevented papilloma and carcinoma. A joint-pathway requirement maps flexibility, not necessity of either individual pathway. Preprint and journal report are one study family.

Tally: 18 discriminating contrasts. Null and negative contrasts are discriminators, not positive threshold crossings. Reports and replications are not independent causal lines. Only positive endogenous tumour or stage-specific contributions carry the causal flag. Biochemical/cell rescues, tracing, unrescued graft tests and null dispensability contrasts remain informative without being counted as positive broad primacy crossings.

Cap audit, precedence and Β/Γ/Δ boundary

“A perturbation crosses a threshold within the system it was actually run in. A universal cap governs the inference from that system to the hypothesis’s claimed scope.” Cancer rubric02 governs endogenous initiation, maintenance and relapse. Metabolism rubric06 informs flux and redox but does not impose lifespan endpoints on cancer. Positive HK2 and related autochthonous tests receive their scoped causal credit; the broad state-necessity and primacy claims remain Γ because the targeted causal inference is not established, not because the whole convergent literature is re-capped at its weakest member.

Gamma hard-rule audit: The universal in vitro and overexpression inference limits apply to GLUT3-induced culture architecture, not to valid autochthonous tumour results. Transplantation and enzyme perturbation are interpreted in their actual scopes. A field-valid dependency test does not alone identify fermentation-specific primacy.

SC1 cap: “Association-only cancer claims cap at Β.” Human and mouse tracing confirms increased glycolytic activity in many tumours, alongside respiration and fuel heterogeneity. This descriptive existence claim is supported without a causal upgrade [33,49–56].

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. Human and mouse tracing confirms increased glycolytic activity in many tumours, alongside respiration and fuel heterogeneity. This descriptive existence claim is supported without a causal upgrade [33,49–56].

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. Specific HK2 and other metabolic dependencies have positive endogenous causal tests, but the general aerobic fermentation state is not required in all assessed tumours. PKM2 isoform loss is not equivalent to removal of glycolysis; the LDHA SCC null more directly constrains fermentation necessity [44–48,65–67].

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. Metabolic feedback can activate oncogenic signals in culture, but endogenous KRAS withdrawal, copy-number and effector tests generally place an oncogene before reprogramming. A general glycolysis-first transformation/maintenance model has not crossed its causal threshold [31,35,37–41,57,68].

Upward threshold ladder

SC1 Δ: “Requires at least one meaningful causal perturbation showing the proposed mechanism contributes to initiation, maintenance, or relapse.” Satisfied: no. The association/existence statement is not upgraded to a causal proposition by additional tracing alone.

SC2 Δ: “Requires at least one meaningful causal perturbation showing the proposed mechanism contributes to initiation, maintenance, or relapse.” Satisfied: no. Selectively lower aerobic lactate fermentation while preserving glucose-derived biosynthesis and enzyme scaffolding, with tumour outcomes and orthogonal redox rescue across the prespecified endogenous lineages.

SC3 Δ: “Requires at least one meaningful causal perturbation showing the proposed mechanism contributes to initiation, maintenance, or relapse.” Satisfied: no. Endogenous physiological flux manipulation must induce or sustain malignant transformation with oncogenic pathways intact, establish temporal precedence, and survive matched epistasis and alternative-function rescue.

Upper ladder: Endogenous genetic natural experiments and therapeutic incidence/survival interventions can reach the upper ladder without an unethical human cancer-induction trial. Named oncogene and metabolic dependencies have strong causal evidence, but the assessed convergence does not establish general fermentation-first primacy. Human tracing gives substrate use rather than intervention-based necessity; phase-I dose limitation and uncontrolled DCA observations do not carry the registered claim further.

Independence: The lines test different failure modes but are not counted as sufficient human primacy convergence. Early and later Dallas tracing cohorts may overlap, and a larger follow-up is not automatically independent replication; no unverified cohort-independence claim is made. The HFSC SCC studies share a lineage platform, while their separate genetic interventions extend a boundary. Clinical IACS preclinical and phase-I reports are distinct stages, not replication of efficacy.

Endogenous tumour genetics [46, 45, 67, 57]: Reduces transplantation and ordinary clinical confounding; separates positive dependencies from genuine nulls.

Human in vivo carbon fate [50, 49, 54, 55]: Checks culture-derived assumptions and imaging proxies, with mixed-compartment and exchange limitations.

Function-specific metabolic rescue [28, 24, 64]: Separates enzyme presence and fermentation rate from nutrient/redox outputs, while retaining engineering and culture limitations.

Falsification and pitfalls

The nominated metabolic node can be dispensable. LDHA loss preserved SCC outcomes and PKM2 loss had lineage-dependent null, acceleration or delay effects [45,47–48,67]. Additional downgrade: False.

Metabolic rate and growth dissociate. Lower fermentation after shuttle modulation did not lower proliferation [64]. Additional downgrade: False.

Primary and recurrence states differ. KRAS-surviving cells and endocrine persisters rely more on respiration [26,30]. Additional downgrade: False.

Presence mistaken for primacy. FDG avidity and glycolytic expression are not flux-state necessity [54].

Enzyme function equated with the nominated metabolic state. HK2 mitochondrial binding, ENO1 immune regulation and PKM2 noncanonical roles [8,14,44].

Xenograft capacity equated with endogenous initiation. PKM isoform replacement and LDHA complementation [11,59].

Clinical pharmacological failure misread as biological falsification. IACS exposure was limited by toxicity, preventing a clean efficacy test [60].

Scope fitted to outcome. Genotype, tissue, oxygen and baseline treatment state are specified antecedently; responders are not used to define the claim.

Adversarial steelman: above, then below

The case for Δ is that glucose uptake can activate oncogenic pathways, metabolic rescue changes tumour phenotypes, and endogenous HK2 deletion strongly alters initiation and maintenance [37,44,46]. Those results establish more than passive metabolism and justify Ε for named dependencies.

The case for Β is that human primacy evidence is mainly observational and fermentation can be dispensable [54,64–67]. It overlooks experimentally demonstrated metabolic-to-signalling feedback and causal contributions in cancer models [10,37,44,46].

Reject Δ for general upstream primacy because selective endogenous fermentation-first causation across the registered scope has not been shown. Reject Β as the primary score because mechanistic feedback and scoped causal contributions exist. Retain Γ with clearly displayed real observation and bounded Ε rival models.

Final score, causal direction and M/S/D

Primary-subclaim SC3 returns Γ. Cell-level feedback and strong context-specific metabolic contributions make the hypothesis more than association, but no examined evidence set establishes general upstream metabolic primacy across tumour classes. SC2 remains Γ for the registered broad state-necessity claim; named endogenous dependencies reach Ε as a competing model. Weakest-link sensitivity is Β due to descriptive SC1; primary and scope-matched sensitivities are Γ. The real Warburg observation is not labelled refuted.

The demonstrated nominal contribution is largely execution and amplification downstream of oncogenic signalling, with feedback through redox, YAP/HIF, lactate and immune interactions. HK2 can be required within named models, but necessity of a glucose-phosphorylation enzyme does not establish fermentation as the initiating lesion. Rate limitation and general primacy do not follow from proliferation inhibition.

Momentum: Mechanistic only. The 2024 combined LDHA/GLS deletion study tests the limits of metabolic bypass [66], while the 2024 mitochondrial editing work tests redox and immune consequences [9]. These are discriminating perturbations even where the general primacy claim is not supported.

Stability: Stable. No verified imminent clinical or endogenous flux-isolation result is used to claim a pending score change. The proposed stage-resolved experiment could move SC3 but is not a known upcoming readout.

Divergence: Contested; no single implied-level arithmetic. These dimensions do not change level or confidence.

PROPOSED DECISIVE EXPERIMENTS

In prespecified autochthonous KRAS, APC and mammary tumour models, independently manipulate cytosolic NADH recycling and lactate fermentation at physiological exposure while preserving glucose uptake, PPP/hexosamine carbon routing and enzyme scaffolding. Apply perturbations before initiation and separately after tumour establishment, with orthogonal flux rescue, matched oncogene signalling and lineage-resolved metastasis/relapse follow-up.

Readout: Incident tumours, clonal growth, maintenance regression and metastatic/relapse burden alongside absolute fermentation, TCA flux, redox, ATP demand and biosynthetic routing. Controls: Catalytic versus scaffold rescue, wild-type and nonmalignant controls, matched oxygen/perfusion and nutrient supply, orthogonal redox instruments, blinded pathology and prespecified tissues/genotypes. Supporting: Physiological fermentation elevation initiates malignant state before oncogenic changes, and selective reversal suppresses the same state despite preserved nonfermentation outputs across the registered lineages. Falsifying: Validated fermentation reduction preserves initiation or maintenance, or effects disappear after biosynthetic/redox rescue while the nominated fermentation state remains altered.

RESEARCH TIMELINE

KNOWLEDGE GAPS

Stage-resolved endogenous flux manipulation that isolates fermentation from glucose-derived biosynthesis and enzyme scaffolding.

Paired lineage-resolved human primary/metastatic tracing, with measured perfusion, substrate exchange and immune/stromal contribution.

Prospectively define metabolic dependence using functional flux rather than FDG or transcript abundance alone.

Test adaptive transporter and nutrient bypass with orthogonal rescue before declaring an enzyme universally necessary.

Read linked corrections and inaccessible full Methods/supplements before promoting precise numerical or translational claims; improve therapeutic windows without treating exposure-limited failures as clean biological nulls.

METHODS AND CONSENSUS CONTRIBUTION

Consensus was used through the signed-in browser for discovery, counterevidence, research-gap and timeline suggestions, and checks of specific indexed full-text passages. Extracted Methods/Results details were reconciled with original-paper passages where available. Provider screening counts, citation counts and meters were not used as primary counts or HSEF scores. This source log records access depth; it does not claim complete-paper or supplement review. The monthly API allowance was conserved: zero Consensus API calls in this batch.

REVIEWER NOTES

68 primary reports were individually appraised. Seven original full texts were examined in selected passages, with exact receipts; other sources have complete abstract/status examination. Consensus supplied additional indexed full-text checks but those are not counted as independently examined original bodies. No complete-paper, supplement, exhaustive correction or blinded audit is claimed. The review is a bounded critical synthesis with reproducible discovery logs.

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.

ADVERSARIAL REVIEW AND APPROVAL

Authoring model: Sol 6.1. Adversarial reviewer: Claude Opus 5.5 (Anthropic). Date: 2026-10-06.

Scope of checks. All 68 reference PMIDs re-verified against PubMed records (title, first author, journal, year, DOI, status flags): 68/68 resolve to the stated work; no retractions or expressions of concern, and the correction notices the review already flags remain attached; the only year differences are print-issue versus e-pub year. 14 load-bearing claims spot-checked against PubMed abstracts and the local examination records, including the HK2 conditional-deletion result, the liver HK2 rescue experiments, the four PKM2 experiments and the original PKM2-to-PKM1 replacement study, the LDHA-dispensable squamous-carcinoma experiment and the later combined LDHA/GLS deletion, the NSCLC FDG-versus-tracing study, the GLUT3 three-dimensional culture study, the inducible KRAS maintenance experiment, and the complex I inhibitor phase I trials. Scoring logic re-derived from framework/hsef-hypothesis-review-v5/SKILL.md and references/hsef-framework.md: cap order and cap precedence, the index-prescribed primary-subclaim aggregation with SC3 as the gate, the Β/Γ/Δ boundary including the requirement that the mechanism-specific perturbation return a positive result, and the Δ-versus-Ε distinction.

Rejected findings (checked, no change). (a) SC2 at Γ rather than Δ or Ε — argued at length against the first answer. Hexokinase 2 deletion is a positive endogenous perturbation in autochthonous models, which looks Δ-qualifying; but the nominated mechanism is lactate fermentation despite oxygen, the enzyme-specific test of that state (LDHA ablation in the squamous carcinoma cell of origin) is null while the positive perturbations also remove biosynthetic glucose routing, and Step 9 requires the qualifying perturbation's result to be positive for the mechanism actually claimed. Γ is the defensible reading and the framework text does not force a change. (b) Overall Γ by primary-subclaim aggregation — the index designates SC3 as the gate, SC3 is Γ, and the weakest-link and scope-matched sensitivities are reported as required. (c) SC1 at Β — universal cap 2 and the cancer rubric's association cap bind a descriptive existence claim. (d) The headline wording, which declines to call the Warburg observation refuted — this is exactly what "Grade a claim, not a topic" requires. (e) Citation years differing by one from the PubMed e-pub year (PMIDs 25394852, 24316969, 25482511, 31846640, 33382985) — print/issue-year convention. (f) The study descriptions checked against sources (HK2 initiation and maintenance with continued HK1 expression; PKM2 deletion accelerating Brca1-driven mammary tumours; Ldha-null tumours unchanged in number, timing and volume with reduced glycolytic metabolites and FDG uptake; SUV not correlating with glycolytic intermediate labelling; no recommended phase 2 dose for the complex I inhibitor) all match.

Residual limitations. Access remains bounded: 7 partial-body and 61 abstract/status examinations with no complete-paper or supplement audit; several linked correction notices are retained but their numerical effects were not audited, and the independence of the human tracing cohorts is unresolved rather than established. This pass checked citation integrity, attribution and framework conformance; it is not independent blinded peer review.

Approval: Following adversarial review and the corrections listed above, the revised document is approved. — Claude Opus 5.5 (Anthropic), 2026-10-06.

REFERENCES

1. Chen H, Li Y, Li H et al. NBS1 lactylation is required for efficient DNA repair and chemotherapy resistance. Nature. 2024. PMID 38961290. DOI: 10.1038/s41586-024-07620-9. Access: abstract_metadata. Study: PMID38961290.

2. Li L, Liang Y, Kang L et al. Transcriptional Regulation of the Warburg Effect in Cancer by SIX1. Cancer cell. 2018. PMID 29455928. DOI: 10.1016/j.ccell.2018.01.010. Access: abstract_metadata. Study: PMID29455928.

3. Ma WK, Voss DM, Scharner J et al. ASO-Based PKM Splice-Switching Therapy Inhibits Hepatocellular Carcinoma Growth. Cancer research. 2022. PMID 34921016. DOI: 10.1158/0008-5472.CAN-20-0948. Access: abstract_metadata. Study: PMID34921016.

4. Zhang H, Su X, Burley SK et al. mTOR regulates aerobic glycolysis through NEAT1 and nuclear paraspeckle-mediated mechanism in hepatocellular carcinoma. Theranostics. 2022. PMID 35547764. DOI: 10.7150/thno.72581. Access: abstract_metadata. Study: PMID35547764.

5. Hu C, Liu T, Han C et al. HPV E6/E7 promotes aerobic glycolysis in cervical cancer by regulating IGF2BP2 to stabilize m6A-MYC expression. International journal of biological sciences. 2022. PMID 35002506. DOI: 10.7150/ijbs.67770. Access: abstract_metadata. Study: PMID35002506.

6. Wang Z, Zhang L, Zhang D et al. Glycolysis inhibitor 2-deoxy-D-glucose suppresses carcinogen-induced rat hepatocarcinogenesis by restricting cancer cell metabolism. Molecular medicine reports. 2015. PMID 25394852. DOI: 10.3892/mmr.2014.2945. Access: abstract_metadata. Study: PMID25394852.

7. Uen W, Tseng T, Wu CP et al. Detachment stress mediated bioenergetic switch of malignant melanoma cells into anti-Warburg phenotype. Aging. 2022. PMID 35802540. DOI: 10.18632/aging.204164. Access: abstract_metadata. Study: PMID35802540.

8. Zhu Q, Li J, Sun H et al. O-GlcNAcylation of enolase 1 serves as a dual regulator of aerobic glycolysis and immune evasion in colorectal cancer. Proceedings of the National Academy of Sciences of the United States of America. 2024. PMID 39446384. DOI: 10.1073/pnas.2408354121. Access: abstract_metadata. Study: PMID39446384.

9. Mahmood M, Liu EM, Shergold AL et al. Mitochondrial DNA mutations drive aerobic glycolysis to enhance checkpoint blockade response in melanoma. Nature cancer. 2024. PMID 38286828. DOI: 10.1038/s43018-023-00721-w. Access: full_text_partial. Study: PMID38286828.

10. Enzo E, Santinon G, Pocaterra A et al. Aerobic glycolysis tunes YAP/TAZ transcriptional activity. The EMBO journal. 2015. PMID 25796446. DOI: 10.15252/embj.201490379. Access: abstract_metadata. Study: PMID25796446.

11. Christofk HR, Vander Heiden MG, Harris MH et al. The M2 splice isoform of pyruvate kinase is important for cancer metabolism and tumour growth. Nature. 2008. PMID 18337823. DOI: 10.1038/nature06734. Access: abstract_metadata. Study: PMID18337823.

12. Masui K, Tanaka K, Akhavan D et al. mTOR complex 2 controls glycolytic metabolism in glioblastoma through FoxO acetylation and upregulation of c-Myc. Cell metabolism. 2013. PMID 24140020. DOI: 10.1016/j.cmet.2013.09.013. Access: abstract_metadata. Study: PMID24140020.

13. Wiel C, Le Gal K, Ibrahim MX et al. BACH1 Stabilization by Antioxidants Stimulates Lung Cancer Metastasis. Cell. 2019. PMID 31257027. DOI: 10.1016/j.cell.2019.06.005. Access: abstract_metadata. Study: PMID31257027.

14. Luo W, Hu H, Chang R et al. Pyruvate kinase M2 is a PHD3-stimulated coactivator for hypoxia-inducible factor 1. Cell. 2011. PMID 21620138. DOI: 10.1016/j.cell.2011.03.054. Access: abstract_metadata. Study: PMID21620138.

15. Morita M, Sato T, Nomura M et al. PKM1 Confers Metabolic Advantages and Promotes Cell-Autonomous Tumor Cell Growth. Cancer cell. 2018. PMID 29533781. DOI: 10.1016/j.ccell.2018.02.004. Access: abstract_metadata. Study: PMID29533781.

16. Kim D, Fiske BP, Birsoy K et al. SHMT2 drives glioma cell survival in ischaemia but imposes a dependence on glycine clearance. Nature. 2015. PMID 25855294. DOI: 10.1038/nature14363. Access: abstract_metadata. Study: PMID25855294.

17. Sakamaki T, Casimiro MC, Ju X et al. Cyclin D1 determines mitochondrial function in vivo. Molecular and cellular biology. 2006. PMID 16809779. DOI: 10.1128/MCB.02074-05. Access: abstract_metadata. Study: PMID16809779.

18. Anastasiou D, Poulogiannis G, Asara JM et al. Inhibition of pyruvate kinase M2 by reactive oxygen species contributes to cellular antioxidant responses. Science (New York, N.Y.). 2011. PMID 22052977. DOI: 10.1126/science.1211485. Access: abstract_metadata. Study: PMID22052977.

19. Zhang J, Wang S, Jiang B et al. c-Src phosphorylation and activation of hexokinase promotes tumorigenesis and metastasis. Nature communications. 2017. PMID 28054552. DOI: 10.1038/ncomms13732. Access: abstract_metadata. Study: PMID28054552.

20. Molina JR, Sun Y, Protopopova M et al. An inhibitor of oxidative phosphorylation exploits cancer vulnerability. Nature medicine. 2018. PMID 29892070. DOI: 10.1038/s41591-018-0052-4. Access: abstract_metadata. Study: PMID29892070.

21. Krall AS, Mullen PJ, Surjono F et al. Asparagine couples mitochondrial respiration to ATF4 activity and tumor growth. Cell metabolism. 2021. PMID 33609439. DOI: 10.1016/j.cmet.2021.02.001. Access: abstract_metadata. Study: PMID33609439.

22. Sullivan LB, Gui DY, Hosios AM et al. Supporting Aspartate Biosynthesis Is an Essential Function of Respiration in Proliferating Cells. Cell. 2015. PMID 26232225. DOI: 10.1016/j.cell.2015.07.017. Access: abstract_metadata. Study: PMID26232225.

23. LeBleu VS, O'Connell JT, Gonzalez Herrera KN et al. PGC-1α mediates mitochondrial biogenesis and oxidative phosphorylation in cancer cells to promote metastasis. Nature cell biology. 2014. PMID 25241037. DOI: 10.1038/ncb3039. Access: abstract_metadata. Study: PMID25241037.

24. Birsoy K, Wang T, Chen WW et al. An Essential Role of the Mitochondrial Electron Transport Chain in Cell Proliferation Is to Enable Aspartate Synthesis. Cell. 2015. PMID 26232224. DOI: 10.1016/j.cell.2015.07.016. Access: abstract_metadata. Study: PMID26232224.

25. Maddocks ODK, Athineos D, Cheung EC et al. Modulating the therapeutic response of tumours to dietary serine and glycine starvation. Nature. 2017. PMID 28425994. DOI: 10.1038/nature22056. Access: abstract_metadata. Study: PMID28425994.

26. Tau S, Chamberlin MD, Yang H et al. Oxidative Phosphorylation Is a Metabolic Vulnerability of Endocrine Therapy-Tolerant Persister Cells in ER+ Breast Cancer. Cancer research. 2025. PMID 39777474. DOI: 10.1158/0008-5472.CAN-24-1204. Access: abstract_metadata. Study: PMID39777474.

27. Commander R, Wei C, Sharma A et al. Subpopulation targeting of pyruvate dehydrogenase and GLUT1 decouples metabolic heterogeneity during collective cancer cell invasion. Nature communications. 2020. PMID 32210228. DOI: 10.1038/s41467-020-15219-7. Access: abstract_metadata. Study: PMID32210228.

28. Sullivan LB, Luengo A, Danai LV et al. Aspartate is an endogenous metabolic limitation for tumour growth. Nature cell biology. 2018. PMID 29941931. DOI: 10.1038/s41556-018-0125-0. Access: full_text_partial. Study: PMID29941931.

29. Masoud R, Reyes-Castellanos G, Lac S et al. Targeting Mitochondrial Complex I Overcomes Chemoresistance in High OXPHOS Pancreatic Cancer. Cell reports. Medicine. 2020. PMID 33294863. DOI: 10.1016/j.xcrm.2020.100143. Access: abstract_metadata. Study: PMID33294863.

30. Viale A, Pettazzoni P, Lyssiotis CA et al. Oncogene ablation-resistant pancreatic cancer cells depend on mitochondrial function. Nature. 2014. PMID 25119024. DOI: 10.1038/nature13611. Access: abstract_metadata. Study: PMID25119024.

31. Kerr EM, Gaude E, Turrell FK et al. Mutant Kras copy number defines metabolic reprogramming and therapeutic susceptibilities. Nature. 2016. PMID 26909577. DOI: 10.1038/nature16967. Access: abstract_metadata. Study: PMID26909577.

32. Yun J, Mullarky E, Lu C et al. Vitamin C selectively kills KRAS and BRAF mutant colorectal cancer cells by targeting GAPDH. Science (New York, N.Y.). 2015. PMID 26541605. DOI: 10.1126/science.aaa5004. Access: abstract_metadata. Study: PMID26541605.

33. Bartman CR, Weilandt DR, Shen Y et al. Slow TCA flux and ATP production in primary solid tumours but not metastases. Nature. 2023. PMID 36725930. DOI: 10.1038/s41586-022-05661-6. Access: abstract_metadata. Study: PMID36725930.

34. Peruzzo P, Comelli M, Di Giorgio E et al. Transformation by different oncogenes relies on specific metabolic adaptations. Cell cycle (Georgetown, Tex.). 2016. PMID 27485932. DOI: 10.1080/15384101.2016.1215387. Access: abstract_metadata. Study: PMID27485932.

35. Wang D, Bi Y, Hu L et al. Obesogenic high-fat diet heightens aerobic glycolysis through hyperactivation of oncogenic KRAS. Cell communication and signaling : CCS. 2019. PMID 30819189. DOI: 10.1186/s12964-019-0333-7. Access: abstract_metadata. Study: PMID30819189.

36. Gough DJ, Corlett A, Schlessinger K et al. Mitochondrial STAT3 supports Ras-dependent oncogenic transformation. Science (New York, N.Y.). 2009. PMID 19556508. DOI: 10.1126/science.1171721. Access: abstract_metadata. Study: PMID19556508.

37. Onodera Y, Nam JM, Bissell MJ et al. Increased sugar uptake promotes oncogenesis via EPAC/RAP1 and O-GlcNAc pathways. The Journal of clinical investigation. 2014. PMID 24316969. DOI: 10.1172/JCI63146. Access: abstract_metadata. Study: PMID24316969.

38. Weinberg F, Hamanaka R, Wheaton WW et al. Mitochondrial metabolism and ROS generation are essential for Kras-mediated tumorigenicity. Proceedings of the National Academy of Sciences of the United States of America. 2010. PMID 20421486. DOI: 10.1073/pnas.1003428107. Access: abstract_metadata. Study: PMID20421486.

39. Amendola CR, Mahaffey JP, Parker SJ et al. KRAS4A directly regulates hexokinase 1. Nature. 2019. PMID 31827279. DOI: 10.1038/s41586-019-1832-9. Access: abstract_metadata. Study: PMID31827279.

40. Humpton TJ, Alagesan B, DeNicola GM et al. Oncogenic KRAS Induces NIX-Mediated Mitophagy to Promote Pancreatic Cancer. Cancer discovery. 2019. PMID 31263025. DOI: 10.1158/2159-8290.CD-18-1409. Access: abstract_metadata. Study: PMID31263025.

41. Wang F, Qi XM, Wertz R et al. p38γ MAPK Is Essential for Aerobic Glycolysis and Pancreatic Tumorigenesis. Cancer research. 2020. PMID 32580961. DOI: 10.1158/0008-5472.CAN-19-3281. Access: abstract_metadata. Study: PMID32580961.

42. Anastasiou D, Yu Y, Israelsen WJ et al. Pyruvate kinase M2 activators promote tetramer formation and suppress tumorigenesis. Nature chemical biology. 2012. PMID 22922757. DOI: 10.1038/nchembio.1060. Access: abstract_metadata. Study: PMID22922757.

43. Lunt SY, Muralidhar V, Hosios AM et al. Pyruvate kinase isoform expression alters nucleotide synthesis to impact cell proliferation. Molecular cell. 2015. PMID 25482511. DOI: 10.1016/j.molcel.2014.10.027. Access: abstract_metadata. Study: PMID25482511.

44. DeWaal D, Nogueira V, Terry AR et al. Hexokinase-2 depletion inhibits glycolysis and induces oxidative phosphorylation in hepatocellular carcinoma and sensitizes to metformin. Nature communications. 2018. PMID 29386513. DOI: 10.1038/s41467-017-02733-4. Access: full_text_partial. Study: PMID29386513.

45. Lau AN, Israelsen WJ, Roper J et al. PKM2 is not required for colon cancer initiated by APC loss. Cancer & metabolism. 2017. PMID 29214019. DOI: 10.1186/s40170-017-0172-1. Access: full_text_partial. Study: PMID29214019.

46. Patra KC, Wang Q, Bhaskar PT et al. Hexokinase 2 is required for tumor initiation and maintenance and its systemic deletion is therapeutic in mouse models of cancer. Cancer cell. 2013. PMID 23911236. DOI: 10.1016/j.ccr.2013.06.014. Access: abstract_metadata. Study: PMID23911236.

47. Israelsen WJ, Dayton TL, Davidson SM et al. PKM2 isoform-specific deletion reveals a differential requirement for pyruvate kinase in tumor cells. Cell. 2013. PMID 24120138. DOI: 10.1016/j.cell.2013.09.025. Access: abstract_metadata. Study: PMID24120138.

48. Dayton TL, Gocheva V, Miller KM et al. Isoform-specific deletion of PKM2 constrains tumor initiation in a mouse model of soft tissue sarcoma. Cancer & metabolism. 2018. PMID 29854399. DOI: 10.1186/s40170-018-0179-2. Access: full_text_partial. Study: PMID29854399.

49. Faubert B, Li KY, Cai L et al. Lactate Metabolism in Human Lung Tumors. Cell. 2017. PMID 28985563. DOI: 10.1016/j.cell.2017.09.019. Access: abstract_metadata. Study: DALLAS-NSCLC-LACTATE-2017-cohort-overlap-unresolved.

50. Hensley CT, Faubert B, Yuan Q et al. Metabolic Heterogeneity in Human Lung Tumors. Cell. 2016. PMID 26853473. DOI: 10.1016/j.cell.2015.12.034. Access: abstract_metadata. Study: DALLAS-NSCLC-TRACING-2016-cohort-overlap-unresolved.

51. Hui S, Ghergurovich JM, Morscher RJ et al. Glucose feeds the TCA cycle via circulating lactate. Nature. 2017. PMID 29045397. DOI: 10.1038/nature24057. Access: abstract_metadata. Study: PMID29045397.

52. Mashimo T, Pichumani K, Vemireddy V et al. Acetate is a bioenergetic substrate for human glioblastoma and brain metastases. Cell. 2014. PMID 25525878. DOI: 10.1016/j.cell.2014.11.025. Access: abstract_metadata. Study: PMID25525878.

53. Luengo A, Abbott KL, Davidson SM et al. Reactive metabolite production is a targetable liability of glycolytic metabolism in lung cancer. Nature communications. 2019. PMID 31811141. DOI: 10.1038/s41467-019-13419-4. Access: abstract_metadata. Study: PMID31811141.

54. Kernstine KH, Faubert B, Do QN et al. Does Tumor FDG-PET Avidity Represent Enhanced Glycolytic Metabolism in Non-Small Cell Lung Cancer? The Annals of thoracic surgery. 2020. PMID 31846640. DOI: 10.1016/j.athoracsur.2019.10.061. Access: abstract_metadata. Study: DALLAS-NSCLC-FDG-2020-cohort-overlap-unresolved.

55. Maher EA, Marin-Valencia I, Bachoo RM et al. Metabolism of [U-13 C]glucose in human brain tumors in vivo. NMR in biomedicine. 2012. PMID 22419606. DOI: 10.1002/nbm.2794. Access: abstract_metadata. Study: PMID22419606.

56. Marin-Valencia I, Yang C, Mashimo T et al. Analysis of tumor metabolism reveals mitochondrial glucose oxidation in genetically diverse human glioblastomas in the mouse brain in vivo. Cell metabolism. 2012. PMID 22682223. DOI: 10.1016/j.cmet.2012.05.001. Access: abstract_metadata. Study: PMID22682223.

57. Ying H, Kimmelman AC, Lyssiotis CA et al. Oncogenic Kras maintains pancreatic tumors through regulation of anabolic glucose metabolism. Cell. 2012. PMID 22541435. DOI: 10.1016/j.cell.2012.01.058. Access: abstract_metadata. Study: PMID22541435.

58. Sonveaux P, Végran F, Schroeder T et al. Targeting lactate-fueled respiration selectively kills hypoxic tumor cells in mice. The Journal of clinical investigation. 2008. PMID 19033663. DOI: 10.1172/JCI36843. Access: abstract_metadata. Study: PMID19033663.

59. Fantin VR, St-Pierre J, Leder P et al. Attenuation of LDH-A expression uncovers a link between glycolysis, mitochondrial physiology, and tumor maintenance. Cancer cell. 2006. PMID 16766262. DOI: 10.1016/j.ccr.2006.04.023. Access: abstract_metadata. Study: PMID16766262.

60. Yap TA, Daver N, Mahendra M et al. Complex I inhibitor of oxidative phosphorylation in advanced solid tumors and acute myeloid leukemia: phase I trials. Nature medicine. 2023. PMID 36658425. DOI: 10.1038/s41591-022-02103-8. Access: abstract_metadata. Study: PMID36658425.

61. Michelakis ED, Sutendra G, Dromparis P et al. Metabolic modulation of glioblastoma with dichloroacetate. Science translational medicine. 2010. PMID 20463368. DOI: 10.1126/scitranslmed.3000677. Access: abstract_metadata. Study: PMID20463368.

62. Vyas A, Harbison RA, Faden DL et al. Recurrent Human Papillomavirus-Related Head and Neck Cancer Undergoes Metabolic Reprogramming and Is Driven by Oxidative Phosphorylation. Clinical cancer research : an official journal of the American Association for Cancer Research. 2021. PMID 34407971. DOI: 10.1158/1078-0432.CCR-20-4789. Access: abstract_metadata. Study: PMID34407971.

63. Luengo A, Li Z, Gui DY et al. Increased demand for NAD+ relative to ATP drives aerobic glycolysis. Molecular cell. 2021. PMID 33382985. DOI: 10.1016/j.molcel.2020.12.012. Access: abstract_metadata. Study: PMID33382985.

64. Wang Y, Stancliffe E, Fowle-Grider R et al. Saturation of the mitochondrial NADH shuttles drives aerobic glycolysis in proliferating cells. Molecular cell. 2022. PMID 35973426. DOI: 10.1016/j.molcel.2022.07.007. Access: abstract_metadata. Study: PMID35973426.

65. Pathria G, Scott DA, Feng Y et al. Targeting the Warburg effect via LDHA inhibition engages ATF4 signaling for cancer cell survival. The EMBO journal. 2018. PMID 30209241. DOI: 10.15252/embj.201899735. Access: abstract_metadata. Study: PMID30209241.

66. Galvan C, Flores AA, Cerrilos V et al. Defining metabolic flexibility in hair follicle stem cell induced squamous cell carcinoma. Science advances. 2024. PMID 39303037. DOI: 10.1126/sciadv.adn2806. Access: full_text_partial. Study: HFSC-SCC-FLEXIBILITY-2024-journal.

67. Flores A, Sandoval-Gonzalez S, Takahashi R et al. Increased lactate dehydrogenase activity is dispensable in squamous carcinoma cells of origin. Nature communications. 2019. PMID 30626875. DOI: 10.1038/s41467-018-07857-9. Access: full_text_partial. Study: HFSC-SCC-LDHA-2019.

68. Elstrom RL, Bauer DE, Buzzai M et al. Akt stimulates aerobic glycolysis in cancer cells. Cancer research. 2004. PMID 15172999. DOI: 10.1158/0008-5472.CAN-03-2904. Access: abstract_metadata. Study: PMID15172999.

Source-specific evidence cards

[1] NBS1 lactylation is required for efficient DNA repair and chemotherapy resistance.

Shows: LDHA reduction and site-specific NBS1 lactylation work link lactate to DNA repair and chemotherapy resistance.

Does not show: Resistance is distinct from initiation; the abstract's anaerobic wording does not redefine aerobic glycolysis as oxygen absence.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[2] Transcriptional Regulation of the Warburg Effect in Cancer by SIX1.

Shows: SIX1 transcriptional regulation increased glycolytic genes and tumour growth.

Does not show: A transcriptional driver upstream of metabolism does not establish glycolysis as the primary initiating event.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

[3] ASO-Based PKM Splice-Switching Therapy Inhibits Hepatocellular Carcinoma Growth.

Shows: Independent splice-switching ASOs altered PKM isoforms and inhibited xenograft and genetic HCC tumourigenesis.

Does not show: Splice replacement is not PKM2 deletion; genotype, increased pyruvate-kinase activity and nonmetabolic functions remain relevant.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[4] mTOR regulates aerobic glycolysis through NEAT1 and nuclear paraspeckle-mediated mechanism in hepatocellular carcinoma.

Shows: mTOR-NEAT1-paraspeckle regulation controlled glycolytic splicing and tumour growth in an oncogene-driven model.

Does not show: The initiating oncogene and broad RNA-processing changes precede the metabolic phenotype; selective flux mediation is not fully isolated.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

[5] HPV E6/E7 promotes aerobic glycolysis in cervical cancer by regulating IGF2BP2 to stabilize m6A-MYC expression.

Shows: E6/E7 knockdown reduced glycolysis and growth, with IGF2BP2 rescue and MYC-linked regulation.

Does not show: The viral oncogenes sit upstream and rescue affects RNA regulation broadly; this is not proof of glycolysis-first initiation.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

[6] Glycolysis inhibitor 2-deoxy-D-glucose suppresses carcinogen-induced rat hepatocarcinogenesis by restricting cancer cell metabolism.

Shows: 2-DG delayed carcinogenesis and prolonged survival after DEN exposure.

Does not show: 2-DG affects multiple metabolic and glycosylation processes; DEN remains the initiating exposure and no selective metabolic rescue was presented in the abstract.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[7] Detachment stress mediated bioenergetic switch of malignant melanoma cells into anti-Warburg phenotype.

Shows: Detachment reduced lactate secretion and increased respiratory-inhibitor sensitivity.

Does not show: Culture stress is not an endogenous metastatic lesion; adaptive oxidation challenges a fixed state without establishing universal respiration dependence.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2.

[8] O-GlcNAcylation of enolase 1 serves as a dual regulator of aerobic glycolysis and immune evasion in colorectal cancer.

Shows: Distinct ENO1 glycosylation sites regulated catalysis and PD-L1 stability, with tumour and immune effects.

Does not show: One enzyme has separable metabolic and immune functions; site effects cannot all be attributed to glycolytic flux.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[9] Mitochondrial DNA mutations drive aerobic glycolysis to enhance checkpoint blockade response in melanoma.

Shows: Mt-Nd5 truncation caused redox-linked glycolytic changes and enhanced checkpoint response in models.

Does not show: A glycolytic shift can improve immune treatment response; isogenic cell growth was not substantially changed and human heteroplasmy response is observational.

Access/status: Complete primary abstract and linked-status metadata examined. Selected body sections: 1:Results:paragraph0, 1:Results:paragraph1, 1:Results:paragraph3. No complete-paper or supplement audit claimed. UpdateOf: PMID36993533. Subclaims: SC1, SC2.

[10] Aerobic glycolysis tunes YAP/TAZ transcriptional activity.

Shows: Glycolytic perturbation modulated YAP/TAZ through PFK1-TEAD interaction and altered overgrowth in flies.

Does not show: Feedback can reinforce oncogenic signalling but does not establish endogenous mammalian tumour initiation by flux alone.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

[11] The M2 splice isoform of pyruvate kinase is important for cancer metabolism and tumour growth.

Shows: Replacing PKM2 with PKM1 reduced lactate, increased oxygen use and impaired xenograft formation.

Does not show: Transplanted cell replacement does not establish endogenous PKM2 necessity or generalize to every tumour lineage.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[12] mTOR complex 2 controls glycolytic metabolism in glioblastoma through FoxO acetylation and upregulation of c-Myc.

Shows: mTORC2 regulated FoxO acetylation and MYC-linked glycolytic reprogramming.

Does not show: Oncogenic signalling is upstream and patient survival correlations are not metabolic initiation evidence.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

[13] BACH1 Stabilization by Antioxidants Stimulates Lung Cancer Metastasis.

Shows: BACH1-linked glycolysis promoted antioxidant-associated metastasis and BACH1 targeting prevented that increase.

Does not show: Metastatic amplification after KRAS-driven cancer is distinct from primary transformation and BACH1 has broader transcriptional functions.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[14] Pyruvate kinase M2 is a PHD3-stimulated coactivator for hypoxia-inducible factor 1.

Shows: PKM2-PHD3 interactions increased HIF coactivation and glycolytic flux.

Does not show: PKM2 coactivation is a noncanonical enzyme function; loss of a regulator cannot isolate fermentation as the causal driver.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

[15] PKM1 Confers Metabolic Advantages and Promotes Cell-Autonomous Tumor Cell Growth.

Shows: PKM1 supported tumour growth in defined KRAS/carcinogen models and proliferation in pulmonary neuroendocrine cancers.

Does not show: This counters PKM2-specific universality; active pyruvate kinase can support malignancy in particular lineages.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[16] SHMT2 drives glioma cell survival in ischaemia but imposes a dependence on glycine clearance.

Shows: SHMT2 lowered pyruvate-kinase activity and oxygen consumption, supporting survival in ischemic glioma regions.

Does not show: Hypoxic survival adaptation differs from aerobic initiation; glycine toxicity and microenvironment shape the dependency.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[17] Cyclin D1 determines mitochondrial function in vivo.

Shows: Cyclin D1 perturbation reciprocally altered mitochondrial function and metabolic gene expression in mammary models.

Does not show: Cell-cycle control and mitochondrial changes are coupled; the abstract does not isolate metabolic mediation of tumorigenesis.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC3.

[18] Inhibition of pyruvate kinase M2 by reactive oxygen species contributes to cellular antioxidant responses.

Shows: Oxidation-resistant PKM2 reduced redox adaptation and impaired xenograft formation under oxidative stress.

Does not show: PPP redox diversion is not equivalent to lactate dependence; transplanted growth retains the cancer-rubric scope cap.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[19] c-Src phosphorylation and activation of hexokinase promotes tumorigenesis and metastasis.

Shows: Src phosphorylation activated HK1/HK2; phosphorylation-site mutation impaired growth and metastatic phenotypes.

Does not show: Src is upstream, and low-glucose HK1 dependence limits claims of universal HK2 dependency.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

[20] An inhibitor of oxidative phosphorylation exploits cancer vulnerability.

Shows: IACS-010759 inhibited OXPHOS-dependent tumour growth with energy and aspartate effects.

Does not show: Preclinical therapeutic tolerability did not predict the subsequent human therapeutic window; respiration inhibition is not proof of glycolysis primacy.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[21] Asparagine couples mitochondrial respiration to ATF4 activity and tumor growth.

Shows: Asparagine restored proliferation and signalling during respiratory impairment; combined restrictions impaired tumour growth.

Does not show: Nutrient availability bypasses a respiratory output in defined contexts; no universal metabolic state is necessary in every tumour.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[22] Supporting Aspartate Biosynthesis Is an Essential Function of Respiration in Proliferating Cells.

Shows: Alternative electron acceptors or aspartate restored proliferation after respiratory-chain impairment.

Does not show: Cell proliferation mechanistic rescue is strong within culture but not an endogenous tumour-initiation test.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[23] PGC-1α mediates mitochondrial biogenesis and oxidative phosphorylation in cancer cells to promote metastasis.

Shows: PGC-1alpha silencing reduced metastasis without reducing primary tumour growth, separating oxidative motility from proliferation.

Does not show: Stage-specific effects are not universal respiration necessity. Linked erratum details were not fully audited.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. ErratumIn: PMIDnot supplied. Subclaims: SC2.

[24] An Essential Role of the Mitochondrial Electron Transport Chain in Cell Proliferation Is to Enable Aspartate Synthesis.

Shows: GOT1-dependent aspartate production and supplementation explained proliferation rescue after ETC inhibition.

Does not show: Cell metabolic outputs do not establish the general tumour-driver claim; supraphysiological supplementation and transporter expression matter.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[25] Modulating the therapeutic response of tumours to dietary serine and glycine starvation.

Shows: Serine/glycine restriction improved survival in Apc and Myc models, with weaker responses in KRAS-driven cancers.

Does not show: Dietary amino-acid dependence is genotype-specific and distinct from lactate fermentation; linked erratum is retained.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. ErratumIn: PMID28770843. Subclaims: SC2.

[26] Oxidative Phosphorylation Is a Metabolic Vulnerability of Endocrine Therapy-Tolerant Persister Cells in ER+ Breast Cancer.

Shows: Endocrine-tolerant persisters showed OXPHOS reliance; complex-I targeting suppressed tumour regrowth potential.

Does not show: Human neoadjuvant specimens establish markers, not randomized OXPHOS inhibitor efficacy; persistence differs from initiation.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. UpdateOf: PMID39386444. Subclaims: SC2.

[27] Subpopulation targeting of pyruvate dehydrogenase and GLUT1 decouples metabolic heterogeneity during collective cancer cell invasion.

Shows: Respiratory leaders and glucose-using followers required different metabolic interventions during collective invasion.

Does not show: Subpopulation culture cooperation does not prove the same dependencies in intact human metastases.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2.

[28] Aspartate is an endogenous metabolic limitation for tumour growth.

Shows: gpASNase1 increased growth of selected transplanted tumours and bypassed metformin suppression; AsPC-1 growth did not increase.

Does not show: Engineered enzyme sufficiency maps an endogenous nutrient limitation, not universal respiratory necessity or primary cancer initiation.

Access/status: Complete primary abstract and linked-status metadata examined. Selected body sections: 0:Methods:paragraph1, 0:Methods:paragraph2, 0:Methods:paragraph11, 0:Methods:paragraph12, 2:Main text (unsectioned body paragraphs):paragraph6, 2:Main text (unsectioned body paragraphs):paragraph7, 2:Main text (unsectioned body paragraphs):paragraph8, 2:Main text (unsectioned body paragraphs):paragraph9, 2:Main text (unsectioned body paragraphs):paragraph10, 2:Main text (unsectioned body paragraphs):paragraph11, 2:Main text (unsectioned body paragraphs):paragraph12. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[29] Targeting Mitochondrial Complex I Overcomes Chemoresistance in High OXPHOS Pancreatic Cancer.

Shows: High-OXPHOS PDAC subsets responded to phenformin plus gemcitabine in orthotopic models.

Does not show: Clinical response was proposed rather than demonstrated; baseline tumour heterogeneity matters.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[30] Oncogene ablation-resistant pancreatic cancer cells depend on mitochondrial function.

Shows: Cells surviving KRAS ablation relied more on mitochondrial respiration and supported relapse.

Does not show: A selected persistence state differs from untreated primary growth. A linked 2026 erratum is retained and its affected details were not audited.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. ErratumIn: PMID41951747. Subclaims: SC2, SC3.

[31] Mutant Kras copy number defines metabolic reprogramming and therapeutic susceptibilities.

Shows: Endogenous mutant allele enrichment accompanied progression and a glycolytic/TCA/glutathione metabolic program.

Does not show: Genotype precedes rewiring, and early versus advanced tumours differ; enrichment is not proof that metabolism independently initiated cancer.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC3.

[32] Vitamin C selectively kills KRAS and BRAF mutant colorectal cancer cells by targeting GAPDH.

Shows: High-dose vitamin C caused redox-linked GAPDH inhibition and impaired growth in Apc/KRAS mice.

Does not show: Pleiotropic oxidative toxicity and genotype selection prevent treating this as a clean test of fermentation-only necessity or clinical efficacy.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[33] Slow TCA flux and ATP production in primary solid tumours but not metastases.

Shows: Quantitative tracing found suppressed primary-solid-tumour TCA flux and calculated ATP synthesis, with higher TCA flux in metastases.

Does not show: Calculated ATP depends on flux modelling assumptions; increased glycolysis did not imply greater total ATP production or a causal initiating lesion.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2.

[34] Transformation by different oncogenes relies on specific metabolic adaptations.

Shows: RAS and mutant HDAC4 produced different metabolic adaptations, with transformation without increased lactate in the latter.

Does not show: In vitro engineered transformation does not establish every endogenous tumour route, but refutes a unique cell-model metabolic program.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC3.

[35] Obesogenic high-fat diet heightens aerobic glycolysis through hyperactivation of oncogenic KRAS.

Shows: COX2 deletion attenuated HFD-driven KRAS hyperactivation, glycolysis and dysplasia.

Does not show: Diet, inflammation and oncogene activation are upstream; COX2 perturbation is not selective glycolytic subtraction.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

[36] Mitochondrial STAT3 supports Ras-dependent oncogenic transformation.

Shows: Mitochondria-targeted STAT3 mutants supported RAS transformation without canonical nuclear signalling.

Does not show: Engineered protein localization narrows a mitochondrial contribution rather than universal upstream glycolytic causation.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[37] Increased sugar uptake promotes oncogenesis via EPAC/RAP1 and O-GlcNAc pathways.

Shows: GLUT3 overexpression increased glucose uptake and activated oncogenic signalling and disorganized architecture; reduced uptake restored polarity.

Does not show: Overexpression and tissue architecture are not endogenous incidence or proof of aerobic lactate flux as sufficient primary cause.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3.

[38] Mitochondrial metabolism and ROS generation are essential for Kras-mediated tumorigenicity.

Shows: TFAM loss reduced KRAS-driven tumorigenesis; mitochondrial ROS and glutamine metabolism supported transformation.

Does not show: Mitochondrial output necessity in this model contradicts a simple defective-respiration account without making it universal.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[39] KRAS4A directly regulates hexokinase 1.

Shows: GTP-dependent KRAS4A binding directly regulated HK1, providing an oncogene-to-metabolic-enzyme link.

Does not show: Direct binding is not itself an endogenous tumour-survival intervention or complete clinical mechanism.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

[40] Oncogenic KRAS Induces NIX-Mediated Mitophagy to Promote Pancreatic Cancer.

Shows: Nix loss delayed PDAC progression and improved survival, while later mitochondrial-clearance compensation allowed progression.

Does not show: The KRAS-NIX program is upstream of altered flux; delay is not permanent abrogation or generic metabolic sufficiency.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

[41] p38γ MAPK Is Essential for Aerobic Glycolysis and Pancreatic Tumorigenesis.

Shows: Pancreatic p38gamma loss reduced glycolytic intermediates/signalling and PDAC, with PFKFB3 pathway epistasis.

Does not show: The kinase has broader signalling roles and sits downstream of KRAS. Linked erratum PMID39957350 limits unaudited details.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. ErratumIn: PMID39957350. Subclaims: SC2, SC3.

[42] Pyruvate kinase M2 activators promote tetramer formation and suppress tumorigenesis.

Shows: PKM2 activators promoted an active tetramer and reduced xenograft growth.

Does not show: Activity modulation differs from deletion and from PKM1 expression in every lineage; linked correction details were not fully audited.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. ErratumIn: PMIDnot supplied. Subclaims: SC2.

[43] Pyruvate kinase isoform expression alters nucleotide synthesis to impact cell proliferation.

Shows: PKM1 expression after PKM2 deletion limited nucleotide synthesis; thymine rescued proliferation.

Does not show: This is normal-cell proliferation, a specificity control rather than a cancer-initiating mechanism.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[44] Hexokinase-2 depletion inhibits glycolysis and induces oxidative phosphorylation in hepatocellular carcinoma and sensitizes to metformin.

Shows: Liver HK2 deletion reduced tumour incidence; WT but not kinase-dead or mitochondrial-binding-deficient HK2 restored cell phenotypes.

Does not show: Both catalysis and mitochondrial binding were required in the rescue. Reduced glycolysis coexisted with maintained TCA flux; linked correction retained.

Access/status: Complete primary abstract and linked-status metadata examined. Selected body sections: 1:Results:paragraph0, 1:Results:paragraph1, 1:Results:paragraph4, 1:Results:paragraph5. No complete-paper or supplement audit claimed. ErratumIn: PMID29946147. Subclaims: SC2.

[45] PKM2 is not required for colon cancer initiated by APC loss.

Shows: PKM2 deletion left tumour incidence, multiplicity, survival and nuclear beta-catenin unchanged despite confirmed recombination.

Does not show: PKM2 loss is not removal of all glycolysis; PKM1 compensation and lineage boundaries remain explicit.

Access/status: Complete primary abstract and linked-status metadata examined. Selected body sections: 1:Results:paragraph0, 1:Results:paragraph1, 1:Results:paragraph2, 1:Results:paragraph3. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[46] Hexokinase 2 is required for tumor initiation and maintenance and its systemic deletion is therapeutic in mouse models of cancer.

Shows: HK2 deletion reduced initiation and maintenance despite HK1 expression, and systemic deletion was therapeutic in mice.

Does not show: A first glucose-phosphorylation step supports multiple branches and enzyme functions; this does not prove primary fermentation initiation. Linked correction retained.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. ErratumIn: PMIDnot supplied. Subclaims: SC2.

[47] PKM2 isoform-specific deletion reveals a differential requirement for pyruvate kinase in tumor cells.

Shows: PKM2 loss accelerated mammary tumour formation, with heterogeneous PKM1 expression and proliferating PKM2-null cells.

Does not show: An isoform-specific null is a strong boundary, not proof that tumours have no glucose-metabolism requirement.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[48] Isoform-specific deletion of PKM2 constrains tumor initiation in a mouse model of soft tissue sarcoma.

Shows: PKM2 loss delayed onset but tumours eventually grew; lineage reporting assigned residual PKM2 to stroma and showed PKM1-positive proliferation.

Does not show: Facilitated initiation is distinct from maintenance necessity; bulk expression mixes tumour and stromal compartments.

Access/status: Complete primary abstract and linked-status metadata examined. Selected body sections: 2:Results:paragraph11, 2:Results:paragraph12, 2:Results:paragraph14, 1:Methods:paragraph9, 2:Results:paragraph2, 2:Results:paragraph3, 2:Results:paragraph6. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[49] Lactate Metabolism in Human Lung Tumors.

Shows: Human lactate infusion labelled tumour TCA intermediates; MCT1 loss in mouse tumours removed lactate-dependent labelling.

Does not show: Fuel tracing and uptake loss are not proof that lactate oxidation is universally required for growth.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2.

[50] Metabolic Heterogeneity in Human Lung Tumors.

Shows: Tumours showed glycolysis and glucose oxidation with region-dependent use of additional nutrients.

Does not show: Small surgical cohort and tissue heterogeneity limit extrapolation; flux coexistence is not an initiation experiment.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1.

[51] Glucose feeds the TCA cycle via circulating lactate.

Shows: Circulating lactate supplied TCA carbon in tissues and lung/pancreatic tumours, sometimes exceeding direct glucose contribution.

Does not show: Tracer contribution depends on feeding state and exchange; label incorporation does not alone establish growth necessity.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1.

[52] Acetate is a bioenergetic substrate for human glioblastoma and brain metastases.

Shows: Brain tumours oxidized acetate and glucose, with human acetate oxidation linked to ACSS2 expression.

Does not show: Substrate use is observed rather than selectively required; orthotopic models and human samples have distinct access limits.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1.

[53] Reactive metabolite production is a targetable liability of glycolytic metabolism in lung cancer.

Shows: Glo1 ablation increased reactive-metabolite sensitivity and reduced mouse tumour growth.

Does not show: Detoxification of a glycolytic byproduct is a consequence vulnerability, not proof of glycolysis-first oncogenesis.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[54] Does Tumor FDG-PET Avidity Represent Enhanced Glycolytic Metabolism in Non-Small Cell Lung Cancer?

Shows: FDG SUV did not correlate with glycolytic intermediate labelling and instead related to other respiratory substrate markers and proliferation.

Does not show: FDG uptake cannot stand in for measured lactate fermentation or metabolic causal dependence.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1.

[55] Metabolism of [U-13 C]glucose in human brain tumors in vivo.

Shows: Glucose carbon entered lactate, TCA metabolism, anaplerosis and biosynthetic pools in human brain tumours.

Does not show: Multiple carbon fates contradict a simple fermentation-only picture without measuring causal necessity.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1.

[56] Analysis of tumor metabolism reveals mitochondrial glucose oxidation in genetically diverse human glioblastomas in the mouse brain in vivo.

Shows: Three genetically distinct orthotopic GBM lines oxidized glucose and used it for biosynthesis.

Does not show: Patient-derived graft metabolism does not substitute for selective endogenous loss; linked correction retained.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. ErratumIn: PMIDnot supplied. Subclaims: SC1.

[57] Oncogenic Kras maintains pancreatic tumors through regulation of anabolic glucose metabolism.

Shows: KRAS withdrawal established maintenance dependence and controlled glucose routing into anabolic pathways.

Does not show: This places a driver oncogene upstream; metabolic branches and KRAS pleiotropy are not isolated by withdrawal alone.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

[58] Targeting lactate-fueled respiration selectively kills hypoxic tumor cells in mice.

Shows: MCT1 targeting shifted oxidative cells toward glucose use, retarded tumour growth and altered hypoxic-cell survival.

Does not show: Drug selectivity and microenvironmental nutrient competition limit a universal lactate/fermentation interpretation.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[59] Attenuation of LDH-A expression uncovers a link between glycolysis, mitochondrial physiology, and tumor maintenance.

Shows: LDHA reduction impaired tumourigenicity and human LDHA complementation reversed the phenotype.

Does not show: Transplantation-linked maintenance is not universal endogenous initiation; oxygen and rescue specificity matter and correction retained.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. ErratumIn: PMIDnot supplied. Subclaims: SC2.

[60] Complex I inhibitor of oxidative phosphorylation in advanced solid tumors and acute myeloid leukemia: phase I trials.

Shows: IACS-010759 trials could not establish a recommended phase-II dose because target exposure was limited by lactate elevation and neurotoxicity, with limited activity.

Does not show: Dose-limited pharmacology does not refute biological OXPHOS dependence; clinical safety and mechanistic validity are separate.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[61] Metabolic modulation of glioblastoma with dichloroacetate.

Shows: DCA altered mitochondrial signals in patient tissue and a small treated cohort, with reversible neuropathy limiting dose.

Does not show: No randomized control or reliable efficacy estimate; biomarker change and clinical indications cannot establish cancer primacy.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[62] Recurrent Human Papillomavirus-Related Head and Neck Cancer Undergoes Metabolic Reprogramming and Is Driven by Oxidative Phosphorylation.

Shows: Recurrent cancers showed OXPHOS enrichment and NRF2-related respiratory vulnerability in experimental models.

Does not show: Matched recurrence associations and xenograft inhibition do not show universal dependency or randomized clinical benefit.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2.

[63] Increased demand for NAD+ relative to ATP drives aerobic glycolysis.

Shows: PDH activation reduced NAD+/NADH and proliferation; uncoupling or ATP hydrolysis rescued both despite increased glucose oxidation.

Does not show: A redox/ATP-demand mechanism in cells is not an endogenous cancer-initiation study.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

[64] Saturation of the mitochondrial NADH shuttles drives aerobic glycolysis in proliferating cells.

Shows: Increasing NADH-shuttle flux reduced lactate fermentation without reducing proliferation; fermentation correlated with shuttle capacity rather than growth.

Does not show: Short-term cell behaviour and engineered shuttle activity cannot establish long-term tumour-wide dispensability.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

[65] Targeting the Warburg effect via LDHA inhibition engages ATF4 signaling for cancer cell survival.

Shows: LDHA inhibition failed to impair melanoma proliferation or tumour growth and activated GCN2-ATF4/glutamine survival programs.

Does not show: An effective adaptive bypass is actual dependency failure under that intervention; synthetic cotargeting does not restore LDHA-only necessity.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2.

[66] Defining metabolic flexibility in hair follicle stem cell induced squamous cell carcinoma.

Shows: GLS loss alone preserved tumourigenesis with lactate-use compensation; combined LDHA/GLS loss prevented papilloma and carcinoma.

Does not show: A joint-pathway requirement maps flexibility, not necessity of either individual pathway. Preprint and journal report are one study family.

Access/status: Complete primary abstract and linked-status metadata examined. Selected body sections: 1:RESULTS:paragraph0, 1:RESULTS:paragraph2, 1:RESULTS:paragraph6, 1:RESULTS:paragraph7, 2:DISCUSSION:paragraph0, 2:DISCUSSION:paragraph2, 2:DISCUSSION:paragraph3, 3:MATERIALS AND METHODS:paragraph1. No complete-paper or supplement audit claimed. UpdateOf: PMID37905122. Subclaims: SC2.

[67] Increased lactate dehydrogenase activity is dispensable in squamous carcinoma cells of origin.

Shows: LDHA abrogation or induction did not alter tested tumour outcomes despite reduced glycolytic metabolites and FDG uptake.

Does not show: Targeted fermentation is not all glucose metabolism. Multiple Cre/chemical paradigms strengthen a lineage-specific null; linked correction retained.

Access/status: Complete primary abstract and linked-status metadata examined. Selected body sections: 1:Results:paragraph4, 1:Results:paragraph5, 1:Results:paragraph6. No complete-paper or supplement audit claimed. ErratumIn: PMID30914644. Subclaims: SC2.

[68] Akt stimulates aerobic glycolysis in cancer cells.

Shows: Akt increased glucose consumption and aerobic glycolysis without increasing oxidative phosphorylation or culture proliferation.

Does not show: Oncogene activation precedes the switch, and glucose-withdrawal sensitivity does not isolate lactate fermentation from all glucose functions.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC3.

Documentary record

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

• SHA256.txt
• codex-correction-provenance.json
• consensus-addendum.md
• consensus-validation.json
• edition-approval-audit.json
• validation-report.json
• validation-round2-2026-10-14.json
• validation.txt
• warburg-cancer-driver.review.md
• warburg-cancer-driver.scorecard.json

Sources: 68 · load-bearing claims: 72 · verified: 72 · unverified: 0 · contradicted: 0 · schema 5.3