In plain terms

The claim being assessed: Laetrile/amygdalin treatment provides clinically meaningful anticancer benefit through selective tumour toxicity at regimens promoted for human treatment.

Where the evidence stands: Α — Speculative. An idea with a story behind it, but no evidence yet that it actually happens.

What this mechanism does: The evidence reviewed does not show this mechanism causing the outcome. That does not show it has no effect.

Strongest evidence for it: Overall Α follows the preregistered primary SC2 human-regimen clinical-benefit gate. Partial selectivity and experimental activity remain Γ in SC1; they cannot be transferred into the clinical-benefit score. Toxicity avoidance is separately contradicted at Α, and generic care-substitution survival associations remain Β. Route-dependent cyanide harm has its own Ζ model assessment. No finding of toxicity alone determines the benefit grade.

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

Laetrile has no demonstrated clinically meaningful benefit in the examined human cancer regimens; oral use can cause serious cyanide poisoning, while newer preclinical delivery studies remain experimental.

Implications for health and treatment: Overall Α follows the preregistered primary SC2 human-regimen clinical-benefit gate. Partial selectivity and experimental activity remain Γ in SC1; they cannot be transferred into the clinical-benefit score. Toxicity avoidance is separately contradicted at Α, and generic care-substitution survival associations remain Β. Route-dependent cyanide harm has its own Ζ model assessment. No finding of toxicity alone determines the benefit grade.

Limits for individual care: 67 primary reports with 8 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: 67 primary reports; claims: 72; verified: 72; unverified: 0. Verification is bounded attribution. Access: 8 partial-body and 59 bounded abstract/status examinations. Literature cutoff: 2026-10-12; source examination UTC date: 2026-10-12. 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. 3 corrections applied (see ADVERSARIAL REVIEW AND APPROVAL). Status: Approved as revised.

AT A GLANCE

Laetrile has no demonstrated clinically meaningful benefit in the examined human cancer regimens; oral use can cause serious cyanide poisoning, while newer preclinical delivery studies remain experimental.

Deciding experiment: First characterize formulation identity, parent/metabolite kinetics and cyanide generation with clinically anchored exposure in matched malignant and normal systems. Establish selective functional activity, normal-organ margin and pharmacodynamic tumour exposure in controlled relevant models using carrier/enzyme-only and route-matched controls. A future new-platform add-on trial would require those preclinical criteria and independent review; it must preserve effective cancer care and is not a recommendation to revive the historical unbeneficial regimen.

Belief versus evidence: Overall Α follows the preregistered primary SC2 human-regimen clinical-benefit gate. Partial selectivity and experimental activity remain Γ in SC1; they cannot be transferred into the clinical-benefit score. Toxicity avoidance is separately contradicted at Α, and generic care-substitution survival associations remain Β. Route-dependent cyanide harm has its own Ζ model assessment. No finding of toxicity alone determines the benefit grade.

Best-supported rival: Route-dependent nonselective cyanide toxicity. Human oral/IV parent and cyanide pharmacology (PMID 7005480; 7273266), intestinal-versus-tumour glucosidase assays (PMID 6796962), and controlled mouse activation measurements (PMID 7448350) distinguish systemic activation from a consistently selective tumour mechanism. Engineered magnetic activation (PMID 32636623) changes that delivery problem experimentally, but its formulation is different and its body retains cardiac/liver biomarker effects.

COMPETING MECHANISTIC MODELS

Rank 1: Route-dependent nonselective cyanide toxicity (Ζ). Microbial or food-enzyme activation and coingestions generate systemic cyanide rather than a reliably tumour-selective clinical exposure. Human route pharmacology, prescribed-dose poisoning, food crossover, gut-flora manipulations and animal thiosulfate rescue constrain distinct errors and converge on harm. Not every route or dose causes poisoning; small IV settings and some chronic series show little clinical toxicity. Cases cannot determine population incidence.

Rank 2: Contextual cell activity without a clinical exposure bridge (Γ). Culture effects and engineered delivery can be real without being achievable as a safe effective promoted human regimen. Matched normal-cell positives, drug-combination effects and targeted enzyme/nanocarrier models support restricted activity; human plasma and culture exposure contexts differ. Some native animal tumour models report activity, so it would be wrong to assert the molecule is biologically inactive in every setting.

Rank 3: Selection and concurrent treatment explain apparent clinical benefit (Β). Solicited favorable cases and provider impressions can reflect selection, prior treatment and subjective endpoints rather than drug efficacy. The 1978 case-record design solicits only perceived successes; modern provider survey uses impressions and limited monitoring, and substitution cohorts show treatment-refusal differences. This does not prove every individual response is attributable to prior treatment or quantify confounding in every testimonial.

Rank 4: Selective tumour toxicity produces clinical benefit (Α). Promoted laetrile exposure safely targets tumour cells and improves human disease control. Selected cell-pair sparing, engineered antibody-enzyme activation and newer animal delivery studies provide its strongest good-faith support. Human regimen benefit lacks a credible positive comparator, matched marrow shows no selectivity, and systemic oral poisoning directly opposes a broad safe selective mechanism.

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

Laetrile or amygdalin provides clinically meaningful anticancer benefit through selective tumour toxicity at regimens promoted for human treatment.

Evaluated wording: Laetrile/amygdalin treatment provides clinically meaningful anticancer benefit through selective tumour toxicity at regimens promoted for human treatment. The original registered wording, including "laetrile or amygdalin", is retained above and in preregistration. The slash denotes the registered treatment family, with chemical formulation, route and dose still assessed separately. This is a grammatical restatement, without changing the claim or broadening evidence across formulations.

Primary SC2 is the registered clinical policy/benefit claim. Etiologic mechanisms and separately stated harms are reported at their own scope; proof of harm does not itself assign the nominal benefit level.

Selection: position 21 in this three-case user-selected harm-related batch, registered on 2026-10-12; this is not a measured popularity ranking. Registered claim assessed against original sources. Any carried-forward edition and prior visibility are recorded in the change record; no independent blinded calibration is claimed.

Evidence and causal interpretation

Rubric fit: Rubric 02 directly governs tumour control and clinical survival. Its literal endogenous-driver/lineage definitions are designed mainly for cancer mechanisms, so an explicit exposure-treatment adaptation is used: a field-valid drug-benefit test requires measured cancer response/control or survival under a specified exposure, with a credible comparator, not merely marker shifts. Toxicology SC3 separately concerns clinical symptoms and measured exposure under routes/coingestions with specificity, dechallenge/antidote and controlled model contrasts. SC4 concerns actual treatment receipt and clinical survival, while observational coding and selection prevent a laetrile-specific causal estimate. Original rubric quotes are preserved; this adaptation does not amend v5.3.

The original 1982 human abstract reports 178 treated with a practitioner-like metabolic programme, no substantive benefit and cyanide hazards. This is a nonrandomized trial, not a controlled survival effect. The 1978 original selected-case analysis does include judged responses, but only presumed successes were solicited and records were incomplete. Modern provider impressions cannot correct that attribution problem. The exact 175 evaluable/one temporary response detail is confirmed in an auxiliary NCI summary, not directly verified in the unavailable original Methods/Results. PMID 7033783; 683212; 38999817.

Route changes both parent exposure and cyanide. Six-patient original pharmacology and related measurements show much higher IV parent exposure with urinary elimination and low observed toxicity, while oral parent exposure is far lower and cyanide rises, with almond interaction. Intestinal enzymes and microbial studies challenge the idea that activation is naturally restricted to tumours. These small studies do not prove every IV regimen safe. PMID 7005480; 7273266; 6796962; 7222669; 7448350.

Clinical poisoning is not speculative. Prescribed oral-tablet ingestion precedes shock/acidosis with antidote response in the checked case body; other cases include measured cyanide, neurological injury and interacting foods/vitamin C. Healthy food crossover defines matrix-specific exposure, while controlled animal route/antibiotic and thiosulfate tests add causal specificity. Cases do not estimate incidence, and massive overdoses are distinguished from intended regimens. PMID 28912981; 16014371; 25605411; 6301673; 25708890; 7063858; 203726.

Positive preclinical findings are retained. Selected fibroblast and WI38 comparators show differential toxicity, whereas matched normal marrow and leukemia clonogens are inhibited similarly. Modern renal combination body distinguishes little monodrug activity from synergy without apoptosis. Concentrations of several mg/mL in culture differ substantially from low oral parent exposure, and brief IV plasma peaks do not establish tumour concentration. PMID 32778045; 37726740; 36204233; 6929726; 39519581; 7273266.

Engineered delivery is a different intervention. Magnetic enzyme particles achieve mouse tumour-growth effects with inactive targeting controls but retain AST/CK/LDH increases despite limited histological injury. The new Consensus-derived niosomal gel body uses small DMBA skin-lesion groups and topical loaded gel versus blank gel/oral free compound, with local histology rather than systemic cyanide safety. Alginate-chitosan abstract reports delivery/cell activity, but its broad normal-tissue protection conclusion has not been independently verified by a body audit. PMID 32636623; 36355478; 32097740; 41882653.

Replacing effective care is a separate risk proposition. Alternative-only and refusal cohorts associate with worse survival; complementary-use mortality attenuates after refusal/delay adjustment. These observational results do not identify laetrile and cannot turn a generic hazard ratio into a causal drug estimate. Original Alberta Methods confirms limited matching and exclusion of older patients. PMID 28922780; 30027204; 22734852; 16978951.

HYPOTHESIS STATE EVALUATION FRAMEWORK (HSEF v5.3)

Subclaim matrix and aggregation

Preregistered aggregation: primary-subclaim. Overall Α follows the preregistered primary SC2 human-regimen clinical-benefit gate. Partial selectivity and experimental activity remain Γ in SC1; they cannot be transferred into the clinical-benefit score. Toxicity avoidance is separately contradicted at Α, and generic care-substitution survival associations remain Β. Route-dependent cyanide harm has its own Ζ model assessment. No finding of toxicity alone determines the benefit grade. Sensitivities: weakest-link Α; primary-subclaim Α; scope-matched Α.

Decisive-perturbation inventory and causal tally

[1] The original nonrandomized study treated 178 patients and reports no substantive cancer benefit with cyanide hazards. Uncontrolled heterogeneous trial, not an RCT; exact response numbers were unavailable in the original abstract/body.

[2] Six cancer patients show route differences: IV parent excretion with little toxicity, oral cyanide and almond-related symptoms. Small safety study does not establish clinical benefit or safety of every IV regimen.

[8] Intestinal mucosa has activating glucosidase activity not detected in the human tumour tissues tested. Specific tissue panel challenges selective activation but cannot exclude every alternate cellular mechanism.

[13] Prescribed oral tablets precede hypotension, acidosis and altered consciousness with rapid antidote response in a cancer patient. Selected body does not report a measured cyanide value; temporal clinical attribution is not a poisoning-rate estimate.

[26] Drug alone and enzyme combination fail tumour tests near tolerated-dose limits; enzyme potentiates lethality. Three transplantable models are restricted contexts; negative contrasts earn no positive efficacy crossing.

[28] Early visual metastasis benefits fail independent, cooperative and blinded assessments; leukemia/prevention tests are negative. Disputed positive observations are retained with replication limits, not erased or generalized to new delivery platforms.

[38] Matched marrow and myeloid leukemia clonogenic tests find no selective killing, including enzyme activation. Specific assay cannot exclude every cell-pair effect, while normal-tissue cost is directly retained.

[52] Magnetic enzyme-loaded nanoparticles permit targeted activation and mouse tumour-growth inhibition. Different experimental platform; body reports AST, CK and LDH increases despite no obvious histological organ injury.

Tally: 0 positive qualifying study units across scoped subclaims; 54 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. The zero positive crossing tally concerns the stated benefit/safety propositions; it does not count or negate causal support for the separately scored opposite cyanide-harm model. Pivotal nulls and poisoning events do not become positive clinical benefit or harmlessness crossings. Cell assays, engineered platforms and repeated reports are not independent human efficacy studies.

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

SC1 retains partial-system Γ support from matched-cell and experimental delivery findings. The preregistered SC2 gate asks about benefit of promoted human regimens; different experimental platforms cannot supply that clinical result. The uncontrolled negative treatment programme is informative without being a randomized causal effect. SC3 poisoning contradicts toxicity avoidance; it supports the opposite harm model rather than an efficacy crossing. SC4 remains a non-drug-specific survival association.

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 death, apoptosis markers and docking do not establish a clinically attainable tumour-selective exposure. The old human study is nonrandomized and its abstract says no substantive benefit; clinical toxicity is not proof of anticancer efficacy. New enzyme/nanocarrier/gel platforms are distinct interventions, not confirmations of historical formulations. Matched normal-cell positives are retained, but normal marrow negative and systemic poisoning limit generalization. Substitution cohorts lack laetrile-specific identification and causal randomization.

SC1 cap: “In vitro evidence alone cannot exceed Γ.” Some matched malignant/normal cell comparisons and engineered activation systems show differential effects. Other matched marrow studies show no selective killing, and high culture concentrations, tissue-mismatched controls and enzyme/carrier additions prevent a general human therapeutic window. Partial system selectivity is retained at Γ, not promoted to clinical efficacy.

SC4 cap: “Pure association evidence — however large, however replicated — cannot exceed Β.” Treatment substitution/refusal cohorts associate with worse survival and treatment-refusal adjustment attenuates the complementary-use association. These are clinically important warnings, but the retrieved cohorts are nonrandomized and not laetrile-specific. This separate broad causal proposition is held at Β under the association-only cap; the observed hazard ratios are not randomized causal estimates.

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. Some matched malignant/normal cell comparisons and engineered activation systems show differential effects. Other matched marrow studies show no selective killing, and high culture concentrations, tissue-mismatched controls and enzyme/carrier additions prevent a general human therapeutic window. Partial system selectivity is retained at Γ, not promoted to clinical efficacy.

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. The registered human-regimen benefit claim has no credible, reproducible positive clinical support in the examined set. The 178-patient nonrandomized treatment programme reported no substantive benefit. Six judged responses in solicited favorable historical records lack a credible counterfactual and reliable attribution. Laboratory activity and new-platform animal effects remain SC1 support and do not establish SC2. Α is evidence-derived unsupported clinical benefit, rather than a penalty for the absence of an RCT or a claim of zero biological activity.

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. The blanket claim of avoiding clinically important systemic cyanide toxicity is directly challenged by prescribed oral-regimen poisoning, measured cyanide, food interactions and animal rescue/exposure tests. Some small IV pharmacology settings lack observed toxicity, but that cannot establish safety across promoted regimens. The positive avoidance claim is unsupported at Α; the opposite route-dependent poisoning model scores separately and higher.

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. Treatment substitution/refusal cohorts associate with worse survival and treatment-refusal adjustment attenuates the complementary-use association. These are clinically important warnings, but the retrieved cohorts are nonrandomized and not laetrile-specific. This separate broad causal proposition is held at Β under the association-only cap; the observed hazard ratios are not randomized causal estimates.

SC1 applied rubric adaptation (distinct from the unchanged literal criterion): For a drug-selectivity mechanism, matched malignant and normal systems plus clinically anchored tumour/normal exposure and in vivo confirmation are the applied criterion. Cell death in vitro or a tumour-only marker shift cannot establish a human therapeutic window.

SC2 applied rubric adaptation (distinct from the unchanged literal criterion): For a treatment-benefit claim, the applied field-valid criterion is objective tumour response/control or survival under the specified regimen with a credible counterfactual, clinically relevant exposure and interpretable concurrent therapy. The negative uncontrolled study is a discriminator, not a positive causal crossing or randomized effect size.

SC3 applied rubric adaptation (distinct from the unchanged literal criterion): For the avoidance-of-toxicity claim, the applied criterion is route/dose-specific clinical safety with measured exposure, interacting-food controls and a meaningful observation horizon. Clinical poisoning with timing, cyanide or antidote/dechallenge supports the opposite harm proposition; it does not cross the positive avoidance threshold.

SC4 applied rubric adaptation (distinct from the unchanged literal criterion): For treatment-substitution harm, compare actual treatment receipt/refusal with survival in predefined cancers while controlling selection and treatment indication. The retrieved association-only cohorts do not establish a laetrile-specific or randomized effect.

SC1 literal Γ hard rule: “Sphere formation alone” Not triggered literally for this assembled scoped case. Any adapted clinical, proxy or system-inference restriction is stated separately in the subclaim rationale.

SC1 literal Γ hard rule: “Marker enrichment alone” Not triggered literally for this assembled scoped case. Any adapted clinical, proxy or system-inference restriction is stated separately in the subclaim rationale.

SC1 literal Γ hard rule: “Xenograft tumor-initiation without lineage-resolved endogenous evidence” Not triggered literally for this assembled scoped case. Any adapted clinical, proxy or system-inference restriction is stated separately in the subclaim rationale.

SC1 literal Γ hard rule: “In vitro-only mechanistic studies, however elegant” Applies to the uncontrolled or cell-only positive support class, with ceiling Γ; it does not upgrade that class or describe the whole source set.

SC2 literal Γ hard rule: “Sphere formation alone” Not triggered literally for this assembled scoped case. Any adapted clinical, proxy or system-inference restriction is stated separately in the subclaim rationale.

SC2 literal Γ hard rule: “Marker enrichment alone” Not triggered literally for this assembled scoped case. Any adapted clinical, proxy or system-inference restriction is stated separately in the subclaim rationale.

SC2 literal Γ hard rule: “Xenograft tumor-initiation without lineage-resolved endogenous evidence” Not triggered literally for this assembled scoped case. Any adapted clinical, proxy or system-inference restriction is stated separately in the subclaim rationale.

SC2 literal Γ hard rule: “In vitro-only mechanistic studies, however elegant” Not triggered literally for this assembled scoped case. Any adapted clinical, proxy or system-inference restriction is stated separately in the subclaim rationale.

SC3 literal Γ hard rule: “Sphere formation alone” Not triggered literally for this assembled scoped case. Any adapted clinical, proxy or system-inference restriction is stated separately in the subclaim rationale.

SC3 literal Γ hard rule: “Marker enrichment alone” Not triggered literally for this assembled scoped case. Any adapted clinical, proxy or system-inference restriction is stated separately in the subclaim rationale.

SC3 literal Γ hard rule: “Xenograft tumor-initiation without lineage-resolved endogenous evidence” Not triggered literally for this assembled scoped case. Any adapted clinical, proxy or system-inference restriction is stated separately in the subclaim rationale.

SC3 literal Γ hard rule: “In vitro-only mechanistic studies, however elegant” Not triggered literally for this assembled scoped case. Any adapted clinical, proxy or system-inference restriction is stated separately in the subclaim rationale.

SC4 literal Γ hard rule: “Sphere formation alone” Not triggered literally for this assembled scoped case. Any adapted clinical, proxy or system-inference restriction is stated separately in the subclaim rationale.

SC4 literal Γ hard rule: “Marker enrichment alone” Not triggered literally for this assembled scoped case. Any adapted clinical, proxy or system-inference restriction is stated separately in the subclaim rationale.

SC4 literal Γ hard rule: “Xenograft tumor-initiation without lineage-resolved endogenous evidence” Not triggered literally for this assembled scoped case. Any adapted clinical, proxy or system-inference restriction is stated separately in the subclaim rationale.

SC4 literal Γ hard rule: “In vitro-only mechanistic studies, however elegant” Not triggered literally for this assembled scoped case. Any adapted clinical, proxy or system-inference restriction is stated separately in the subclaim rationale.

Upward threshold ladder

SC1 Δ: “Requires at least one meaningful causal perturbation showing the proposed mechanism contributes to initiation, maintenance, or relapse.” Satisfied: no. Demonstrate tumour-selective functional activity at measured, clinically attainable parent/metabolite exposure in matched normal and malignant systems with orthogonal death/readout assays and in vivo pharmacokinetics; establish a benefit-to-toxic-exposure margin.

SC2 Δ: “Requires at least one meaningful causal perturbation showing the proposed mechanism contributes to initiation, maintenance, or relapse.” Satisfied: no. First validate formulation identity, achievable selective exposure and systemic toxicology in relevant controlled models. Only a new platform meeting those conditions could justify an independently reviewed add-on clinical study with objective cancer endpoints; the historical regimen should not be assumed worth retesting or substituted for effective care.

SC3 Δ: “Requires at least one meaningful causal perturbation showing the proposed mechanism contributes to initiation, maintenance, or relapse.” Satisfied: no. Define route, dose, formulation, coingestions and monitoring prospectively; quantify clinical toxic events and parent/cyanide/thiocyanate exposure over time. A few symptom-free IV administrations do not establish broad-regimen safety.

SC4 Δ: “Requires at least one meaningful causal perturbation showing the proposed mechanism contributes to initiation, maintenance, or relapse.” Satisfied: no. Use stronger counterfactual treatment-receipt designs and validated alternative-modality identification, with indication, stage, prognosis and refusal reasons retained. Do not infer a laetrile-specific hazard ratio from coarse alternative-care categories.

Upper ladder: Therapeutic cancer endpoints, exposure pharmacology, controlled tumour-model panels, objective case-record review and matched-cell selectivity were considered. No positive controlled clinical benefit at the promoted regimen scope was verified; the absence of an RCT is not itself the blocker. Existing human tumour-course evidence and failure to bridge safe exposure are substantive limits. The opposite route-dependent harm case has convergent independent classes and can score above the nominal without reapplying association caps to all its clinical cases. Complementary-use mortality cohorts alone remain association evidence.

Independence: The causal convergence at Ζ belongs to the opposite route-dependent poisoning model, not to clinical anticancer benefit or harmlessness. Human pharmacology papers from the NCI programme are not automatically separate lines. Historical tumour panels share laboratories/regimens; many renal/prostate cell reports share lines and assays. Multiple death/apoptosis markers from one exposure are not independent tests. Cases across settings plus controlled metabolism/route/rescue and direct exposure measurements constrain different failure modes, without producing a population poisoning rate. NCDB substitution/complementary reports share a database and coarse coding.

Human route and exposure pharmacology [2, 3, 18]: Measures systemic parent/cyanide and food-matrix differences, reducing uncertainty about exposure; some studies share a programme and healthy challenges do not measure cancer efficacy.

Clinical poisoning and reversal [10, 13, 11, 17]: Independent temporal exposure, symptoms and antidote/dechallenge reduce a culture-only explanation, while selection and interacting exposures prevent incidence estimates.

Controlled activation and toxic rescue [20, 22, 24]: Microbial/route interventions and thiosulfate rescue constrain cyanide mediation against nonspecific illness; species limits remain.

Restricted experimental tumour activity [50, 52, 66]: Matched cells and altered delivery test biological activity, but share preclinical exposure/specificity translation limits and do not converge on human regimen benefit.

Falsification and pitfalls

Broad toxicity-avoidance prediction fails under promoted oral exposure. Prescribed tablets, food interactions, measured cyanide cases and controlled animal comparisons contradict SC3. The sparse apparent-benign associational safety case is downgraded from Β to Α; the opposite harm model is graded independently. Additional downgrade: True.

Historical human trial labeled randomized. PMID 7033783 is nonrandomized; exact response denominator must not be invented from its abstract.

Positive culture activity equated with clinical exposure. mg/mL and engineered activation do not establish a safe patient tumour concentration.

Every route assigned the same toxicity. Oral/IV and interacting-food differences are original findings.

Generic alternative-care mortality assigned to laetrile. NCDB coding and refusal cohorts do not identify the compound.

Unmonitored provider impressions treated as safety. Most survey providers do not monitor cyanide.

Adversarial steelman: above, then below

One rung above, Β: six judged responses in the selected 1978 case records and clinical testimonials offer a possible clinical association. Favorable-case solicitation, incomplete records, concurrent treatments and no credible comparator prevent a reproducible benefit association attributable to the eligible regimen. Γ would require importing laboratory or new-platform activity into a different clinical scope; that evidence is retained in SC1 instead.

Α is the lowest HSEF rung, so no lower letter exists. A stronger dismissal asserting that amygdalin is biologically inactive in all systems is rejected: matched-cell and animal delivery positives are real within their examined contexts. The nominal clinical score does not erase Γ mechanistic support or the separate cyanide-harm model.

Retain Α for the registered human-regimen benefit gate and Γ for partial-system selectivity. The selected clinical responses cannot establish replicated attribution; the 178-patient negative human course is not an RCT or a universal zero-effect proof. New delivery platforms retain their own evidence and would require a new clinical scope and justified exposure bridge.

Final score, causal direction and M/S/D

Overall Α follows the preregistered primary SC2 human-regimen clinical-benefit gate. Partial selectivity and experimental activity remain Γ in SC1; they cannot be transferred into the clinical-benefit score. Toxicity avoidance is separately contradicted at Α, and generic care-substitution survival associations remain Β. Route-dependent cyanide harm has its own Ζ model assessment. No finding of toxicity alone determines the benefit grade.

For the registered clinical anticancer benefit, no demonstrated disease-control dependency in either direction is established, so Step 2 applies to purported human benefit. This classification is not a description of cyanide poisoning: administered exposure, metabolism and antidote/dechallenge demonstrate genuine harm causation, which is graded in the opposite model. Nor does it deny laboratory cytotoxicity; that is a different scoped phenotype.

Momentum: Mechanistic only. The 2020 magnetic enzyme platform (PMID 32636623), 2022 niosomal gel (PMID 36355478), 2024 renal combination (PMID 39519581) and 2026 nanocomposite (PMID 41882653) test preclinical delivery/activity. The 2024 provider survey (PMID 38999817) supplies use patterns, not a verified clinical efficacy trial underway.

Stability: Stable. No verified named pending result was identified that demonstrably determines the full-scope letter; this is a bounded search finding, not proof that none exists.

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

PROPOSED DECISIVE EXPERIMENTS

First characterize formulation identity, parent/metabolite kinetics and cyanide generation with clinically anchored exposure in matched malignant and normal systems. Establish selective functional activity, normal-organ margin and pharmacodynamic tumour exposure in controlled relevant models using carrier/enzyme-only and route-matched controls. A future new-platform add-on trial would require those preclinical criteria and independent review; it must preserve effective cancer care and is not a recommendation to revive the historical unbeneficial regimen.

Readout: Measured parent, cyanide and thiocyanate in tumour and normal compartments; objective tumour function/control and systemic injury; conditional later clinical progression/survival. Controls: Formulation/purity, free drug versus carrier/enzyme, oral/IV/topical route, microbiota/foods, normal tissue proliferation, delivered tumour exposure, systemic toxicity, concurrent standard therapy and prespecified endpoints. Supporting: A new platform produces replicated selective functional tumour control at attainable exposure with verified normal-organ safety and then, only if justified, an add-on clinical endpoint benefit. Falsifying: Cell activity requires unachievable or toxic exposure, disappears against normal tissue/carrier controls, or delivery improves markers without tumour control and systemic safety.

KNOWLEDGE GAPS

Bridge cell concentrations to measured tumour and normal-organ exposure rather than equating plasma peak with sustained tumour concentration.

Separate free compound, historical products and new carrier/enzyme platforms.

Verify actual normal-tissue comparators and systemic cyanide margins for modern selectivity claims.

Resolve route, foods and microbial activation without assuming every IV dose is safe or every oral exposure is lethal.

Retain the difference between 178 treated and 175 evaluable, and label the temporary response detail as secondary-source verification until original body is obtained.

Do not use generic alternative-care mortality coding as a laetrile-specific causal estimate.

Audit current retractions before reusing older positive cell papers.

REVIEWER NOTES

Sixty-seven relevant original primary reports were examined, with eight selected original bodies. Most full papers and supplements were not audited. New Consensus-derived originals PMID 36355478 and PMID 32097740 were actually examined; PMID 683212 preserves selected favorable case observations. Narrative PMID 26272547 and retracted PMID 23137229/34536497 were excluded from primary credit. The NCI professional summary is auxiliary secondary context, not another primary study or original trial-body audit. Printed 2019 IV dose units remain unverified against body. Scite quota was unavailable and no retry/purchase occurred. Final coordinating adjudication, source-access reconciliation and sealing remain pending; structural checks are not scientific confirmation. Coordinating source/claim checks and registered-scope adjudication are recorded in COORDINATOR_ADJUDICATION.json. This is not an independent blinded regrade. HSEF letters measure support for the exact positive claim; negative clinical findings and documented harms are explained separately. Divergence is not a probability or a claim of evenly balanced medical evidence.

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

Our original abstract check agrees: practitioner-like metabolic programme, no substantive benefit and cyanide hazards. The 175 evaluable/one temporary partial response detail is supplied by a later review in Consensus and by auxiliary NCI context in our check; it is not directly verified in an original trial body. Do not call this an RCT or zero responses. Original PMID 7033783.

The original 1978 abstract retains six responses judged by a blinded oncology panel, while only presumed-success cases were solicited and incomplete records excluded. These observations justify interest without proving a treatment effect or becoming an independent controlled trial. Original PMID 683212.

Original six-patient abstract confirms low observed toxicity with IV administration, increased oral cyanide and transient symptoms in one almond-challenged patient. Small route-specific observations cannot prove safety of every IV dose or efficacy. Original PMID 7005480.

Original pharmacology distinguishes high, brief IV parent plasma levels from very low oral parent levels with cyanide increase. These are parent plasma measurements, not sustained tumour concentration. Culture mg/mL activity cannot be equated to the same human exposure. Original PMID 7273266.

Original Results verifies HepG2 versus WI38 differential IC50 and sorafenib synergy at mg/mL concentrations. Unmatched tissue types, cell proliferation and missing whole-organ exposure still limit a human therapeutic window; the positive comparator is retained. Original PMID 37726740.

Original body distinguishes minimal growth effect from 1 mg/mL alone from strong sulforaphane-combination effects, and reports no induced apoptosis in those assays. A low dose in that protocol is not verified patient exposure. Original PMID 39519581.

Original magnetic-enzyme Results agree with a restricted delivery benefit and retain inactive targeting controls. AST, CK and LDH increase despite no obvious organ histology injury. Neither the carrier nor enzyme addition is historical free-drug therapy, and limited histology cannot be labeled complete safety. Original PMID 32636623.

Original record and selected Methods/Results were examined: thirty rats in small DMBA groups, topical loaded gel versus blank gel/oral free compound, reduced local carcinoma volume and skin pathology. These results remain positive in their context, but do not measure a human regimen, systemic cyanide margin or complete normal-organ safety. Provider graphics are not republished. Original PMID 36355478.

The original abstract reports release, mucin/flow and H1299 cell activity, then makes a broad protection conclusion. Our original-body access is unavailable, so an independent normal-tissue endpoint was not verified. Its abstract supports a delivery lead, not clinical proof of protection. Original PMID 32097740.

Original abstract agrees that refusal is important: mortality association becomes nonsignificant after refusal/delay adjustment. This observational attenuation is not randomized mediation proof, not a compound-specific hazard ratio and not evidence that every complementary use causes death. Original PMID 30027204.

Original Methods confirms retrospective primary-treatment refusal, limited age/year/stage matching and exclusion of older patients. The association is clinically relevant but cannot identify laetrile exposure or remove every selection factor. Original PMID 22734852.

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: Laetrile / Amygdalin as Cancer Treatment

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

1965 — chemical comparison documents nonidentical formulations and tests the proposed biological rationale. Original PMID 14281087.

1975 — controlled tumour models test amygdalin with enzyme activation near tolerated doses and find no activity with increased toxicity. Original PMID 1060511.

1978 — selected retrospective case submissions include judged responses but no definite attribution. Original PMID 683212.

1981 — human pharmacology maps oral/IV and interacting-food differences. Original PMID 7005480.

1982 — nonrandomized human metabolic-regimen trial finds no substantive benefit and toxicity hazards. Original PMID 7033783.

2017 — prescribed oral tablets precede a clinically documented poisoning syndrome with antidote response. Original PMID 28912981.

2020 — engineered magnetic-enzyme activation tests a different tumour-delivery platform in mice. Original PMID 32636623.

2022 — niosomal-gel study measures restricted rat skin-tumour volume and local histology. Original PMID 36355478.

2026 — nanocomposite cell assays explore uptake and delivery; human-regimen validation remains absent. Original PMID 41882653.

Research gaps

Bridge cell concentrations to measured tumour and normal-organ exposure rather than equating plasma peak with sustained tumour concentration.

Separate free compound, historical products and new carrier/enzyme platforms.

Verify actual normal-tissue comparators and systemic cyanide margins for modern selectivity claims.

Resolve route, foods and microbial activation without assuming every IV dose is safe or every oral exposure is lethal.

Retain the difference between 178 treated and 175 evaluable, and label the temporary response detail as secondary-source verification until original body is obtained.

Do not use generic alternative-care mortality coding as a laetrile-specific causal estimate.

Audit current retractions before reusing older positive cell papers.

Boundary experiment

First characterize formulation identity, parent/metabolite kinetics and cyanide generation with clinically anchored exposure in matched malignant and normal systems. Establish selective functional activity, normal-organ margin and pharmacodynamic tumour exposure in controlled relevant models using carrier/enzyme-only and route-matched controls. A future new-platform add-on trial would require those preclinical criteria and independent review; it must preserve effective cancer care and is not a recommendation to revive the historical unbeneficial regimen.

ADVERSARIAL REVIEW AND APPROVAL

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

Scope of checks. All 67 REFERENCES PMIDs verified against PubMed metadata (67/67 resolve to the stated first author, year, journal and title; 0 failures, 0 substitutions). The three identifiers cited in REVIEWER NOTES only were also verified: PMID 26272547 resolves as described, and PMID 23137229 and PMID 34536497 both carry the PubMed "Retracted Publication" type, confirming the review's exclusion of them from primary credit. 14 load-bearing claims were checked against the original records, including Moertel 1982 (178 patients treated with a representative Laetrile-practice regimen; "No substantive benefit was observed"; cyanide levels approaching the lethal range), Ellison 1978 (only cases thought to show objective benefit were solicited; 93 cases submitted, 26 eliminated for insufficient documentation, 160 courses read by a 12-oncologist panel with no knowledge of treatment, six Laetrile courses judged responses — two complete, four partial), Moertel 1981 (six patients; intravenous amygdalin largely excreted unchanged with no toxicity, oral dosing producing blood cyanide to 2.1 microgram/mL, one patient symptomatic after a large raw-almond intake), Newmark 1981 (high small-intestinal mucosal beta-glucosidase activity; no glucose release detected in any human neoplastic tissue examined), Newton 1981 (oral cyanide release dependent on gut flora and suppressed by antibiotic pretreatment) and Willhite 1982 (oral but not intravenous laetrile teratogenic in hamsters; thiosulfate protected the embryos). All matched. Scoring logic was re-derived against references/hsef-framework.md and SKILL.md Steps 2–3: the universal caps, the pre-registered primary-subclaim aggregation with its three reported sensitivities, the Ζ independence basis recorded for the rank-1 harm model, and the Momentum/Stability/Divergence artifacts.

Rejected findings (checked, no change). 1. Headline level Α. Argued for Β on the strength of the six panel-judged responses in the 1978 NCI review. Rejected: those records were solicited only where benefit was believed to have occurred, 26 of 93 were too poorly documented to use, and there is no counterfactual — the result is not a reproducible positive association, and cap 2 would cap it at Β even if it were. Α is correct. 2. Softness of the refutation. Considered whether the review understates the result by writing "no demonstrated clinically meaningful benefit" where Moertel's own abstract concludes that amygdalin "is a toxic drug that is not effective as a cancer treatment." Rejected as a change, because the review does report the trial's negative finding and its cyanide hazard, grades the claim at the lowest rung, records "Additional downgrade: True" against the safety subclaim, and separately grades the route-dependent poisoning model at Ζ. The refutation is carried by the structure rather than by a single adjective, and the review correctly declines the one overclaim available here — that amygdalin is biologically inactive in every system, which the matched-cell and engineered-delivery positives do not support. 3. Ζ on the rank-1 harm model. Checked against the v5.2 shared-failure test and the requirement that a Ζ carry a recorded independence basis. Four lines are recorded with the distinct error each excludes, and they do fail differently: human route pharmacology, clinical poisoning with antidote dechallenge, controlled microbial/route manipulation with thiosulfate rescue, and restricted experimental tumour activity. Oral-versus-intravenous teratogenicity with thiosulfate rescue (Willhite 1982) is a genuinely different failure structure from case reports. No change. 4. "Best-supported rival." Verified against the scorecard: models_evaluated[0] has rank_in_tournament: 1 and is_nominal: false; the nominal is rank 4. The known renderer defect that prints the nominal as the rival does not fire here. 5. Momentum. "Mechanistic only" is correct and carries named artifacts (the 2020 magnetic enzyme/prodrug platform, the 2022 niosomal gel, the 2024 renal combination and the 2026 nanocomposite). It is not the "Decisive work underway" boilerplate at issue elsewhere in this batch. 6. Divergence "Contested." Considered whether this manufactures a split where mainstream oncology is not split. Retained: the review states explicitly that divergence "is not a probability or a claim of evenly balanced medical evidence" and that no field consensus is being manufactured, and the promoted-use literature does imply a different level from the assessed one. A reviewer judgement, defensibly recorded. 7. The "175 evaluable / one temporary response" figure. Checked and left as written. It does not appear in the Moertel 1982 abstract, and the review already labels it as secondary-source verification pending access to the original body, and lists that as a knowledge gap. Flagging it as verified would have been the error. 8. Collective authors. No dropped-consortium defect; no reference begins with a stray comma. 9. Rubric routing. Rubric 02's cancer-biology hard rules are quoted verbatim and marked "not triggered literally", with the exposure-treatment adaptation declared separately, as the cap-precedence rule requires.

Residual limitations. 59 of 67 records were accessed at abstract/status depth; the numeric checks above were made against abstracts and the eight examined bodies, not against full reports. The 2019 intravenous dose units and the alginate-chitosan normal-tissue protection conclusion remain unverified against their bodies, as the review states. provenance_header.generated_by reads "Codex; HSEF v5.3" rather than naming Sol 6.1; this is a package-wide field and was left for the lead.

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

REFERENCES

1. Moertel CG, Fleming TR, Rubin J et al. A clinical trial of amygdalin (Laetrile) in the treatment of human cancer. The New England journal of medicine. 1982. PMID 7033783. DOI: 10.1056/NEJM198201283060403. Access: abstract_metadata. Study: PMID7033783.

2. Moertel CG, Ames MM, Kovach JS et al. A pharmacologic and toxicological study of amygdalin. JAMA. 1981. PMID 7005480. No DOI in PubMed record. Access: abstract_metadata. Study: NCI_CLINICAL_PHARMACOLOGY_PROGRAMME.

3. Ames MM, Moyer TP, Kovach JS et al. Pharmacology of amygdalin (laetrile) in cancer patients. Cancer chemotherapy and pharmacology. 1981. PMID 7273266. DOI: 10.1007/BF00253010. Access: abstract_metadata. Study: NCI_CLINICAL_PHARMACOLOGY_PROGRAMME.

4. Rauws AG, Olling M, Timmerman A. The pharmacokinetics of amygdalin. Archives of toxicology. 1982. PMID 7092570. DOI: 10.1007/BF00347879. Access: abstract_metadata. Study: PMID7092570.

5. Mani J, Rutz J, Maxeiner S et al. Cyanide and lactate levels in patients during chronic oral amygdalin intake followed by intravenous amygdalin administration. Complementary therapies in medicine. 2019. PMID 30935547. DOI: 10.1016/j.ctim.2019.03.002. Access: abstract_metadata. Study: PMID30935547.

6. Markowitsch SD, Binali S, Rutz J et al. Survey of Physicians and Healers Using Amygdalin to Treat Cancer Patients. Nutrients. 2024. PMID 38999817. DOI: 10.3390/nu16132068. Access: full_text_partial. Study: PMID38999817.

7. Davignon JP, Trissel LA, Kleinman LM. Pharmaceutical assessment of amygdalin (Laetrile) products. Cancer treatment reports. 1978. PMID 627001. No DOI in PubMed record. Access: abstract_metadata. Study: PMID627001.

8. Newmark J, Brady RO, Grimley PM et al. Amygdalin (Laetrile) and prunasin beta-glucosidases: distribution in germ-free rat and in human tumor tissue. Proceedings of the National Academy of Sciences of the United States of America. 1981. PMID 6796962. DOI: 10.1073/pnas.78.10.6513. Access: abstract_metadata. Study: PMID6796962.

9. LEVI L, FRENCH WN, BICKIS IJ et al. LAETRILE: A STUDY OF ITS PHYSICOCHEMICAL AND BIOCHEMICAL PROPERTIES. Canadian Medical Association journal. 1965. PMID 14281087. No DOI in PubMed record. Access: abstract_metadata. Study: PMID14281087.

10. Bromley J, Hughes BG, Leong DC et al. Life-threatening interaction between complementary medicines: cyanide toxicity following ingestion of amygdalin and vitamin C. The Annals of pharmacotherapy. 2005. PMID 16014371. DOI: 10.1345/aph.1E634. Access: abstract_metadata. Study: PMID16014371.

11. Sauer H, Wollny C, Oster I et al. Severe cyanide poisoning from an alternative medicine treatment with amygdalin and apricot kernels in a 4-year-old child. Wiener medizinische Wochenschrift (1946). 2015. PMID 25605411. DOI: 10.1007/s10354-014-0340-7. Access: abstract_metadata. Study: PMID25605411.

12. Dalkıran T, Kandur Y, Ozaslan M et al. Role of Hemodialysis in the Management of Cyanide Intoxication From Apricot Kernels in a 3-Year-Old Child. Pediatric emergency care. 2020. PMID 30399062. DOI: 10.1097/PEC.0000000000001644. Access: abstract_metadata. Study: PMID30399062.

13. Dang T, Nguyen C, Tran PN. Physician Beware: Severe Cyanide Toxicity from Amygdalin Tablets Ingestion. Case reports in emergency medicine. 2017. PMID 28912981. DOI: 10.1155/2017/4289527. Access: full_text_partial. Study: PMID28912981.

14. Shively RM, Harding SA, Hoffman RS et al. Rebound metabolic acidosis following intentional amygdalin supplement overdose. Clinical toxicology (Philadelphia, Pa.). 2020. PMID 31322009. DOI: 10.1080/15563650.2019.1640369. Access: abstract_metadata. Study: PMID31322009.

15. Konstantatos A, Shiv Kumar M, Burrell A et al. An unusual presentation of chronic cyanide toxicity from self-prescribed apricot kernel extract. BMJ case reports. 2017. PMID 28893740. DOI: 10.1136/bcr-2017-220814. Access: abstract_metadata. Study: PMID28893740.

16. Akyildiz BN, Kurtoğlu S, Kondolot M et al. Cyanide poisoning caused by ingestion of apricot seeds. Annals of tropical paediatrics. 2010. PMID 20196932. DOI: 10.1179/146532810X12637745451951. Access: abstract_metadata. Study: PMID20196932.

17. Kalyanaraman UP, Kalyanaraman K, Cullinan SA et al. Neuromyopathy of cyanide intoxication due to "laetrile" (amygdalin). A clinicopathologic study. Cancer. 1983. PMID 6301673. DOI: 10.1002/1097-0142(19830601)51:11<2126::aid-cncr2820511129>3.0.co;2-l. Access: abstract_metadata. Study: PMID6301673.

18. Abraham K, Buhrke T, Lampen A. Bioavailability of cyanide after consumption of a single meal of foods containing high levels of cyanogenic glycosides: a crossover study in humans. Archives of toxicology. 2016. PMID 25708890. DOI: 10.1007/s00204-015-1479-8. Access: full_text_partial. Study: PMID25708890.

19. Holzbecher MD, Moss MA, Ellenberger HA. The cyanide content of laetrile preparations, apricot, peach and apple seeds. Journal of toxicology. Clinical toxicology. 1984. PMID 6098693. DOI: 10.3109/15563658408992565. Access: abstract_metadata. Study: PMID6098693.

20. Newton GW, Schmidt ES, Lewis JP et al. Amygdalin toxicity studies in rats predict chronic cyanide poisoning in humans. The Western journal of medicine. 1981. PMID 7222669. No DOI in PubMed record. Access: abstract_metadata. Study: PMID7222669.

21. Khandekar JD, Edelman H. Studies of amygdalin (laetrile) toxicity in rodents. JAMA. 1979. PMID 448892. No DOI in PubMed record. Access: abstract_metadata. Study: PMID448892.

22. Hill HZ, Backer R, Hill GJ. Blood cyanide levels in mice after administration of amygdalin. Biopharmaceutics & drug disposition. 1980. PMID 7448350. DOI: 10.1002/bdd.2510010409. Access: abstract_metadata. Study: PMID7448350.

23. Basu TK. High-dose ascorbic acid decreases detoxification of cyanide derived from amygdalin (laetrile): studies in guinea pigs. Canadian journal of physiology and pharmacology. 1983. PMID 6661693. DOI: 10.1139/y83-203. Access: abstract_metadata. Study: PMID6661693.

24. Willhite CC. Congenital malformations induced by laetrile. Science (New York, N.Y.). 1982. PMID 7063858. DOI: 10.1126/science.7063858. Access: abstract_metadata. Study: PMID7063858.

25. Schmidt ES, Newton GW, Sanders SM et al. Laetrile toxicity studies in dogs. JAMA. 1978. PMID 203726. No DOI in PubMed record. Access: abstract_metadata. Study: PMID203726.

26. Laster WR, Schabel FM. Experimental studies of the antitumor activity of amygdalin MF (NSC-15780) alone and in combination with beta-glucosidase (NSC-128056). Cancer chemotherapy reports. 1975. PMID 1060511. No DOI in PubMed record. Access: abstract_metadata. Study: NCI_RODENT_TEST_PROGRAMME.

27. Wodinsky I, Swiniarski JK. Antitumor activity of amygdalin MF (NSC-15780) as a single agent and with beta-glucosidase (NSC-128056) on a spectrum of transplantable rodent tumors. Cancer chemotherapy reports. 1975. PMID 1203898. No DOI in PubMed record. Access: abstract_metadata. Study: NCI_RODENT_TEST_PROGRAMME.

28. Stock CC, Martin DS, Sugiura K et al. Antitumor tests of amygdalin in spontaneous animal tumor systems. Journal of surgical oncology. 1978. PMID 347176. DOI: 10.1002/jso.2930100203. Access: abstract_metadata. Study: PMID347176.

29. Stock CC, Tarnowski GS, Schmid FA et al. Antitumor tests of amygdalin in transplantable animal tumor systems. Journal of surgical oncology. 1978. PMID 642516. DOI: 10.1002/jso.2930100202. Access: abstract_metadata. Study: NCI_RODENT_TEST_PROGRAMME.

30. Hill GJ, Shine TE, Hill HZ et al. Failure of amygdalin to arrest B16 melanoma and BW5147 AKR leukemia. Cancer research. 1976. PMID 1268862. No DOI in PubMed record. Access: abstract_metadata. Study: PMID1268862.

31. Chitnis MP, Adwankar MK, Amonkar AJ. Studies on high-dose chemotherapy of amygdalin in murine P388 lymphocytic leukaemia and P815 mast cell leukaemia. Journal of cancer research and clinical oncology. 1985. PMID 4008516. DOI: 10.1007/BF00390359. Access: abstract_metadata. Study: PMID4008516.

32. El-Masry T, Al-Shaalan N, Tousson E et al. Potential therapy of vitamin B17 against Ehrlich solid tumor induced changes in Interferon gamma, Nuclear factor kappa B, DNA fragmentation, p53, Bcl2, survivin, VEGF and TNF-α Expressions in mice. Pakistan journal of pharmaceutical sciences. 2020. PMID 32122873. No DOI in PubMed record. Access: abstract_metadata. Study: PMID32122873.

33. Attia AA, Salama AF, Eldiasty JG et al. Amygdalin potentiates the anti-cancer effect of Sorafenib on Ehrlich ascites carcinoma and ameliorates the associated liver damage. Scientific reports. 2022. PMID 35444229. DOI: 10.1038/s41598-022-10517-0. Access: abstract_metadata. Study: PMID35444229.

34. Hosny S, Sahyon H, Youssef M et al. Prunus Armeniaca L. Seed Extract and Its Amygdalin Containing Fraction Induced Mitochondrial-Mediated Apoptosis and Autophagy in Liver Carcinogenesis. Anti-cancer agents in medicinal chemistry. 2021. PMID 32510292. DOI: 10.2174/1871520620666200608124003. Access: abstract_metadata. Study: PMID32510292.

35. Mageid ADA, Abdel-Wadoud IM, Salim EI et al. The protective and chemotherapeutical role of amygdalin in induced mammary cancer in experimental mice and upregulation of related genes. Scientific reports. 2025. PMID 40097629. DOI: 10.1038/s41598-025-93620-2. Access: abstract_metadata. Study: PMID40097629.

36. Lin S, Wen J, Xu X et al. Amygdalin Induced Mitochondria-Mediated Apoptosis of Lung Cancer Cells via Regulating NF[Formula: see text]B-1/NF[Formula: see text]B Signaling Cascade in Vitro and in Vivo. The American journal of Chinese medicine. 2022. PMID 35681261. DOI: 10.1142/S0192415X22500586. Access: abstract_metadata. Study: PMID35681261.

37. Syrigos KN, Rowlinson-Busza G, Epenetos AA. In vitro cytotoxicity following specific activation of amygdalin by beta-glucosidase conjugated to a bladder cancer-associated monoclonal antibody. International journal of cancer. 1998. PMID 9833764. DOI: 10.1002/(sici)1097-0215(19981209)78:6<712::aid-ijc8>3.0.co;2-d. Access: abstract_metadata. Study: PMID9833764.

38. Koeffler HP, Lowe L, Golde DW. Amygdalin (Laetrile): effect on clonogenic cells from human myeloid leukemia cell lines and normal human marrow. Cancer treatment reports. 1980. PMID 6929726. No DOI in PubMed record. Access: abstract_metadata. Study: PMID6929726.

39. Qian L, Xie B, Wang Y et al. Amygdalin-mediated inhibition of non-small cell lung cancer cell invasion in vitro. International journal of clinical and experimental pathology. 2015. PMID 26191238. No DOI in PubMed record. Access: abstract_metadata. Study: PMID26191238.

40. Albogami S, Alnefaie A. Role of Amygdalin in Blocking DNA Replication in Breast Cancer In Vitro. Current pharmaceutical biotechnology. 2021. PMID 33535947. DOI: 10.2174/1389201022666210203123803. Access: abstract_metadata. Study: PMID33535947.

41. Sireesha D, Reddy BS, Reginald BA et al. Effect of amygdalin on oral cancer cell line: An in vitro study. Journal of oral and maxillofacial pathology : JOMFP. 2019. PMID 31110425. DOI: 10.4103/jomfp.JOMFP_281_18. Access: abstract_metadata. Study: PMID31110425.

42. Park HJ, Yoon SH, Han LS et al. Amygdalin inhibits genes related to cell cycle in SNU-C4 human colon cancer cells. World journal of gastroenterology. 2005. PMID 16127745. DOI: 10.3748/wjg.v11.i33.5156. Access: abstract_metadata. Study: PMID16127745.

43. Lee HM, Moon A. Amygdalin Regulates Apoptosis and Adhesion in Hs578T Triple-Negative Breast Cancer Cells. Biomolecules & therapeutics. 2016. PMID 26759703. DOI: 10.4062/biomolther.2015.172. Access: abstract_metadata. Study: PMID26759703.

44. Mosayyebi B, Mohammadi L, Kalantary-Charvadeh A et al. Amygdalin Decreases Adhesion and Migration of MDA-MB-231 and MCF-7 Breast Cancer Cell Lines. Current molecular pharmacology. 2021. PMID 32778045. DOI: 10.2174/1874467213666200810141251. Access: abstract_metadata. Study: PMID32778045.

45. Juengel E, Thomas A, Rutz J et al. Amygdalin inhibits the growth of renal cell carcinoma cells in vitro. International journal of molecular medicine. 2016. PMID 26709398. DOI: 10.3892/ijmm.2015.2439. Access: abstract_metadata. Study: PMID26709398.

46. Makarević J, Tsaur I, Juengel E et al. Amygdalin delays cell cycle progression and blocks growth of prostate cancer cells in vitro. Life sciences. 2016. PMID 26827990. DOI: 10.1016/j.lfs.2016.01.039. Access: abstract_metadata. Study: PMID26827990.

47. Tsaur I, Thomas A, Monecke M et al. Amygdalin Exerts Antitumor Activity in Taxane-Resistant Prostate Cancer Cells. Cancers. 2022. PMID 35804883. DOI: 10.3390/cancers14133111. Access: abstract_metadata. Study: PMID35804883.

48. Abboud MM, Al Awaida W, Alkhateeb HH et al. Antitumor Action of Amygdalin on Human Breast Cancer Cells by Selective Sensitization to Oxidative Stress. Nutrition and cancer. 2019. PMID 30407870. DOI: 10.1080/01635581.2018.1508731. Access: abstract_metadata. Study: PMID30407870.

49. Moradipoodeh B, Jamalan M, Zeinali M et al. In vitro and in silico anticancer activity of amygdalin on the SK-BR-3 human breast cancer cell line. Molecular biology reports. 2019. PMID 31583572. DOI: 10.1007/s11033-019-05080-3. Access: abstract_metadata. Study: PMID31583572.

50. El-Sewedy T, Salama AF, Mohamed AE et al. Hepatocellular Carcinoma cells: activity of Amygdalin and Sorafenib in Targeting AMPK /mTOR and BCL-2 for anti-angiogenesis and apoptosis cell death. BMC complementary medicine and therapies. 2023. PMID 37726740. DOI: 10.1186/s12906-023-04142-1. Access: full_text_partial. Study: PMID37726740.

51. Markowitsch SD, Pham T, Rutz J et al. Growth of Renal Cancer Cell Lines Is Strongly Inhibited by Synergistic Activity of Low-Dosed Amygdalin and Sulforaphane. Nutrients. 2024. PMID 39519581. DOI: 10.3390/nu16213750. Access: full_text_partial. Study: PMID39519581.

52. Zhou J, Hou J, Rao J et al. Magnetically Directed Enzyme/Prodrug Prostate Cancer Therapy Based on β-Glucosidase/Amygdalin. International journal of nanomedicine. 2020. PMID 32636623. DOI: 10.2147/IJN.S242359. Access: full_text_partial. Study: PMID32636623.

53. Baghdady SI, Sabra SA, Darwish AB et al. Targeted nanocomposite delivery system of amygdalin using chitosan/reduced graphene oxide-zinc oxide/hyaluronic acid for treatment of head and neck squamous cell carcinoma. BMC oral health. 2026. PMID 41882653. DOI: 10.1186/s12903-026-07943-1. Access: abstract_metadata. Study: PMID41882653.

54. Meenatchi V, Narayanan KB, Sood A et al. Formation of amygdalin/β-cyclodextrin derivatives inclusion complexes for anticancer activity assessment in human cervical carcinoma HeLa cell line. International journal of pharmaceutics. 2024. PMID 38823468. DOI: 10.1016/j.ijpharm.2024.124293. Access: abstract_metadata. Study: PMID38823468.

55. Hakimi F, Sharifyrad M, Safari H et al. Amygdalin/chitosan-polyvinyl alcohol/cerium-tannic acid hydrogel as biodegradable long-time implant for cancer recurrence care applications: An in vitro study. Heliyon. 2023. PMID 38027646. DOI: 10.1016/j.heliyon.2023.e21835. Access: abstract_metadata. Study: PMID38027646.

56. Christodoulou P, Boutsikos P, Neophytou CM et al. Amygdalin as a chemoprotective agent in co-treatment with cisplatin. Frontiers in pharmacology. 2022. PMID 36204233. DOI: 10.3389/fphar.2022.1013692. Access: abstract_metadata. Study: PMID36204233.

57. Todorova A, Pesheva M, Iliev I et al. Antimutagenic, Antirecombinogenic, and Antitumor Effect of Amygdalin in a Yeast Cell-Based Test and Mammalian Cell Lines. Journal of medicinal food. 2017. PMID 28146364. DOI: 10.1089/jmf.2016.0108. Access: abstract_metadata. Study: PMID28146364.

58. Dimitrov M, Iliev I, Bardarov K et al. Phytochemical characterization and biological activity of apricot kernels' extract in yeast-cell based tests and hepatocellular and colorectal carcinoma cell lines. Journal of ethnopharmacology. 2021. PMID 34146630. DOI: 10.1016/j.jep.2021.114333. Access: abstract_metadata. Study: PMID34146630.

59. Lehmane H, Kohonou AN, Tchogou AP et al. Antioxidant, Anti-Inflammatory, and Anti-Cancer Properties of Amygdalin Extracted from Three Cassava Varieties Cultivated in Benin. Molecules (Basel, Switzerland). 2023. PMID 37299029. DOI: 10.3390/molecules28114548. Access: abstract_metadata. Study: PMID37299029.

60. Abdel-Gawad DRI, Ibrahim MA, El-Banna HA et al. Evaluating the therapeutic potential of amygdalin: Cytotoxic and antimicrobial properties. Tissue & cell. 2024. PMID 38908223. DOI: 10.1016/j.tice.2024.102443. Access: abstract_metadata. Study: PMID38908223.

61. Johnson SB, Park HS, Gross CP et al. Use of Alternative Medicine for Cancer and Its Impact on Survival. Journal of the National Cancer Institute. 2018. PMID 28922780. DOI: 10.1093/jnci/djx145. Access: abstract_metadata. Study: PMID28922780.

62. Johnson SB, Park HS, Gross CP et al. Complementary Medicine, Refusal of Conventional Cancer Therapy, and Survival Among Patients With Curable Cancers. JAMA oncology. 2018. PMID 30027204. DOI: 10.1001/jamaoncol.2018.2487. Access: abstract_metadata. Study: PMID30027204.

63. Joseph K, Vrouwe S, Kamruzzaman A et al. Outcome analysis of breast cancer patients who declined evidence-based treatment. World journal of surgical oncology. 2012. PMID 22734852. DOI: 10.1186/1477-7819-10-118. Access: full_text_partial. Study: PMID22734852.

64. Chang EY, Glissmeyer M, Tonnes S et al. Outcomes of breast cancer in patients who use alternative therapies as primary treatment. American journal of surgery. 2006. PMID 16978951. DOI: 10.1016/j.amjsurg.2006.05.013. Access: abstract_metadata. Study: PMID16978951.

65. Ellison NM, Byar DP, Newell GR. Special report on Laetrile: the NCI Laetrile Review. Results of the National Cancer Institute's retrospective Laetrile analysis. The New England journal of medicine. 1978. PMID 683212. DOI: 10.1056/NEJM197809072991013. Access: abstract_metadata. Study: PMID683212.

66. El-Ela FIA, Gamal A, Elbanna HA et al. In Vitro and In Vivo Evaluation of the Effectiveness and Safety of Amygdalin as a Cancer Therapy. Pharmaceuticals (Basel, Switzerland). 2022. PMID 36355478. DOI: 10.3390/ph15111306. Access: full_text_partial. Study: PMID36355478.

67. Sohail R, Abbas SR. Evaluation of amygdalin-loaded alginate-chitosan nanoparticles as biocompatible drug delivery carriers for anticancerous efficacy. International journal of biological macromolecules. 2020. PMID 32097740. DOI: 10.1016/j.ijbiomac.2020.02.191. Access: abstract_metadata. Study: PMID32097740.

Source-specific evidence cards

[1] A clinical trial of amygdalin (Laetrile) in the treatment of human cancer.

Shows: The original nonrandomized study treated 178 patients and reports no substantive cancer benefit with cyanide hazards.

Does not show: Uncontrolled heterogeneous trial, not an RCT; exact response numbers were unavailable in the original abstract/body.

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.

[2] A pharmacologic and toxicological study of amygdalin.

Shows: Six cancer patients show route differences: IV parent excretion with little toxicity, oral cyanide and almond-related symptoms.

Does not show: Small safety study does not establish clinical benefit or safety of every IV regimen.

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.

[3] Pharmacology of amygdalin (laetrile) in cancer patients.

Shows: IV parent exposure is high and rapidly cleared; oral parent exposure is low while cyanide rises and almonds amplify it.

Does not show: Related pharmacology programme; plasma parent drug is not tumour exposure or an efficacy endpoint.

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.

[4] The pharmacokinetics of amygdalin.

Shows: Dog and rat pharmacokinetics demonstrate renal elimination and limited oral parent bioavailability with prunasin formation.

Does not show: Species limits; no efficacy trial or equivalence of oral and IV toxicity.

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.

[5] Cyanide and lactate levels in patients during chronic oral amygdalin intake followed by intravenous amygdalin administration.

Shows: Retrospective oral-plus-IV exposure in 55 patients shows increased cyanide but decreased lactate around infusion.

Does not show: Biomarkers without comparator or objective tumour outcome; printed dose units require body verification.

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.

[6] Survey of Physicians and Healers Using Amygdalin to Treat Cancer Patients.

Shows: Providers report perceived progression delay and few adverse effects, with variable dosing and little cyanide monitoring.

Does not show: Selected 38 of 76 providers; impressions are not audited patient outcomes or evidence of safety.

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

[7] Pharmaceutical assessment of amygdalin (Laetrile) products.

Shows: Selected historical commercial products were subpotent, mislabeled, contaminated or pyrogenic.

Does not show: Manufacturer and batch findings cannot describe every modern preparation or determine cancer efficacy.

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.

[8] Amygdalin (Laetrile) and prunasin beta-glucosidases: distribution in germ-free rat and in human tumor tissue.

Shows: Intestinal mucosa has activating glucosidase activity not detected in the human tumour tissues tested.

Does not show: Specific tissue panel challenges selective activation but cannot exclude every alternate cellular mechanism.

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.

[9] LAETRILE: A STUDY OF ITS PHYSICOCHEMICAL AND BIOCHEMICAL PROPERTIES.

Shows: Canadian and US formulations differ; tested preparations do not support the manufacturers' biological rationale.

Does not show: Historic chemical and biological tests are not clinical endpoints or universal product identification.

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.

[10] Life-threatening interaction between complementary medicines: cyanide toxicity following ingestion of amygdalin and vitamin C.

Shows: Oral amygdalin with high-dose vitamin C precedes severe neurological and acidosis symptoms with rapid antidote response.

Does not show: Single case; exact interaction magnitude and population incidence are not determined.

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.

[11] Severe cyanide poisoning from an alternative medicine treatment with amygdalin and apricot kernels in a 4-year-old child.

Shows: A child receiving oral/IV amygdalin and kernels develops encephalopathy, acidosis and elevated cyanide, improving after antidote.

Does not show: Combined exposure prevents assignment to each component; case supports harm, not an incidence estimate.

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.

[12] Role of Hemodialysis in the Management of Cyanide Intoxication From Apricot Kernels in a 3-Year-Old Child.

Shows: Kernel ingestion precedes pediatric coma and measured cyanide, followed by recovery with antidote and dialysis.

Does not show: Small-child kernel exposure differs from an IV regimen; combined rescue does not isolate dialysis efficacy.

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.

[13] Physician Beware: Severe Cyanide Toxicity from Amygdalin Tablets Ingestion.

Shows: Prescribed oral tablets precede hypotension, acidosis and altered consciousness with rapid antidote response in a cancer patient.

Does not show: Selected body does not report a measured cyanide value; temporal clinical attribution is not a poisoning-rate estimate.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. Selected body sections: 1:2. Case Report:paragraph0, 1:2. Case Report:paragraph1. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3.

[14] Rebound metabolic acidosis following intentional amygdalin supplement overdose.

Shows: Intentional massive amygdalin overdose causes recurrent acidosis and hypotension requiring repeated antidotal treatment.

Does not show: A 20 g overdose is not a standard promoted regimen; severity must not be generalized to all dosing.

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.

[15] An unusual presentation of chronic cyanide toxicity from self-prescribed apricot kernel extract.

Shows: Chronic kernel extract is associated with measured cyanide and abnormal oximetry normalizing after cessation.

Does not show: Oximeter-interference explanation is inferred; single case and unavailable body limit broader attribution.

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.

[16] Cyanide poisoning caused by ingestion of apricot seeds.

Shows: Thirteen pediatric seed-ingestion cases include ventilation, coma and acidosis with recovery after treatment.

Does not show: Selected intensive-care series has no population denominator and does not test isolated IV products.

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.

[17] Neuromyopathy of cyanide intoxication due to "laetrile" (amygdalin). A clinicopathologic study.

Shows: Neuromyopathy after laetrile has elevated cyanide/thiocyanate and improves after discontinuation, with alternatives assessed.

Does not show: Nutritional deficiency contributes; single-patient evidence does not quantify risk or establish cancer efficacy.

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.

[18] Bioavailability of cyanide after consumption of a single meal of foods containing high levels of cyanogenic glycosides: a crossover study in humans.

Shows: Healthy-adult crossover demonstrates different cyanide peaks from foods with equal cyanide-equivalent doses.

Does not show: Small healthy sample and limited challenge measure exposure, not cancer efficacy or poisoning incidence.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. Selected body sections: 1:Materials and methods:paragraph2, 1:Materials and methods:paragraph6, 1:Materials and methods:paragraph8. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC3.

[19] The cyanide content of laetrile preparations, apricot, peach and apple seeds.

Shows: Preparations from an intoxicated patient and fruit seeds contain measurable cyanide.

Does not show: Product analysis complements clinical cases but is not a complete dose-risk or causation model.

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.

[20] Amygdalin toxicity studies in rats predict chronic cyanide poisoning in humans.

Shows: Rat antibiotic/gut-flora comparisons and human fecal bacteria show microbial cyanide release from amygdalin.

Does not show: Animal dosing and ex vivo activity do not specify human incidence or remove formulation/route 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: SC1, SC3.

[21] Studies of amygdalin (laetrile) toxicity in rodents.

Shows: Repeated high intraperitoneal doses produce rat mortality with cyanide signals.

Does not show: High-dose animal route is not a direct human-regimen risk estimate; nonselective toxicity is retained.

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.

[22] Blood cyanide levels in mice after administration of amygdalin.

Shows: Oral mouse dosing produces cyanide and distress; gut contents activate drug whereas tumour tissue releases little; parenteral exposure differs.

Does not show: Species, individual variability and near-lethal doses limit clinical extrapolation.

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.

[23] High-dose ascorbic acid decreases detoxification of cyanide derived from amygdalin (laetrile): studies in guinea pigs.

Shows: Vitamin C coadministration decreases cyanide detoxification through reduced cysteine availability in guinea pigs.

Does not show: Specific metabolic model does not quantify the interaction at every human dose.

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.

[24] Congenital malformations induced by laetrile.

Shows: Oral but not IV hamster exposure causes malformations with cyanide increase; thiosulfate protects embryos.

Does not show: Pregnancy and species scope retained; not a measured human teratogenic incidence.

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.

[25] Laetrile toxicity studies in dogs.

Shows: Oral laetrile with almonds causes death or neurological impairment in dogs at proposed human-equivalent doses.

Does not show: Food, dose and species limits; this does not describe every IV formulation.

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.

[26] Experimental studies of the antitumor activity of amygdalin MF (NSC-15780) alone and in combination with beta-glucosidase (NSC-128056).

Shows: Drug alone and enzyme combination fail tumour tests near tolerated-dose limits; enzyme potentiates lethality.

Does not show: Three transplantable models are restricted contexts; negative contrasts earn no positive efficacy crossing.

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.

[27] Antitumor activity of amygdalin MF (NSC-15780) as a single agent and with beta-glucosidase (NSC-128056) on a spectrum of transplantable rodent tumors.

Shows: Four rodent tumours show no activity under drug/enzyme schedules, with increased combination toxicity.

Does not show: Models share a testing programme and are not four independent studies or all human cancers.

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.

[28] Antitumor tests of amygdalin in spontaneous animal tumor systems.

Shows: Early visual metastasis benefits fail independent, cooperative and blinded assessments; leukemia/prevention tests are negative.

Does not show: Disputed positive observations are retained with replication limits, not erased or generalized to new delivery platforms.

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.

[29] Antitumor tests of amygdalin in transplantable animal tumor systems.

Shows: Broad rodent tumour panels show no effect from high-dose amygdalin, including chemotherapy interaction tests.

Does not show: Many models share a programme; this does not exclude distinct engineered delivery.

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.

[30] Failure of amygdalin to arrest B16 melanoma and BW5147 AKR leukemia.

Shows: Parenteral amygdalin does not arrest two specified murine tumours across a dose range.

Does not show: Model and route-specific negative, not a human randomized trial.

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.

[31] Studies on high-dose chemotherapy of amygdalin in murine P388 lymphocytic leukaemia and P815 mast cell leukaemia.

Shows: High-dose schedules do not prolong survival in two leukemia-bearing mouse models.

Does not show: No inference to an exact human hazard ratio or every formulation.

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.

[32] Potential therapy of vitamin B17 against Ehrlich solid tumor induced changes in Interferon gamma, Nuclear factor kappa B, DNA fragmentation, p53, Bcl2, survivin, VEGF and TNF-α Expressions in mice.

Shows: Amygdalin changes apoptosis, inflammatory and organ-function markers in Ehrlich tumour-bearing mice.

Does not show: Marker-heavy abstract does not show durable human tumour control; B17 terminology is not a validated vitamin status.

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.

[33] Amygdalin potentiates the anti-cancer effect of Sorafenib on Ehrlich ascites carcinoma and ameliorates the associated liver damage.

Shows: Amygdalin with sorafenib reduces Ehrlich tumour burden and liver injury measures in mice.

Does not show: Combination and route context, not clinical benefit or a proved selective cyanide mechanism.

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.

[34] Prunus Armeniaca L. Seed Extract and Its Amygdalin Containing Fraction Induced Mitochondrial-Mediated Apoptosis and Autophagy in Liver Carcinogenesis.

Shows: Apricot extract and amygdalin-containing fraction alter apoptosis/autophagy and proliferation in a chemically induced liver model.

Does not show: Mixed extract and molecular endpoints cannot establish pure-drug clinical efficacy.

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.

[35] The protective and chemotherapeutical role of amygdalin in induced mammary cancer in experimental mice and upregulation of related genes.

Shows: Amygdalin alone or with tamoxifen changes mammary histology and molecular/biochemical tumour measures.

Does not show: Induction, combination and marker limits; no human treatment outcome or selective exposure bridge.

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.

[36] Amygdalin Induced Mitochondria-Mediated Apoptosis of Lung Cancer Cells via Regulating NF[Formula: see text]B-1/NF[Formula: see text]B Signaling Cascade in Vitro and in Vivo.

Shows: Lung cell apoptosis and NFkB1 studies with xenograft findings support restricted model activity.

Does not show: No promoted human-dose selectivity; original abstract only and supplements unexamined.

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.

[37] In vitro cytotoxicity following specific activation of amygdalin by beta-glucosidase conjugated to a bladder cancer-associated monoclonal antibody.

Shows: Antibody-directed glucosidase activation produces antigen-specific killing with mismatched-antibody and coculture controls.

Does not show: Engineered activation differs from ordinary laetrile; two cancer lines do not establish systemic normal-tissue safety.

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.

[38] Amygdalin (Laetrile): effect on clonogenic cells from human myeloid leukemia cell lines and normal human marrow.

Shows: Matched marrow and myeloid leukemia clonogenic tests find no selective killing, including enzyme activation.

Does not show: Specific assay cannot exclude every cell-pair effect, while normal-tissue cost is directly retained.

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.

[39] Amygdalin-mediated inhibition of non-small cell lung cancer cell invasion in vitro.

Shows: Drug inhibits migration/invasion in selected lung cancer lines below concentrations needed for growth inhibition.

Does not show: No actual metastatic disease endpoint or matched normal-tissue/exposure test.

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.

[40] Role of Amygdalin in Blocking DNA Replication in Breast Cancer In Vitro.

Shows: Treatment at the line's IC50 downregulates DNA-replication genes in MCF7 cells.

Does not show: Expression after cytotoxic exposure can be secondary; no matched normal or exposure validation.

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.

[41] Effect of amygdalin on oral cancer cell line: An in vitro study.

Shows: Apricot and almond ethanolic extracts reduce KB cell viability in an MTT assay.

Does not show: Extract identity and cell-line assay do not establish pure-drug oral-cancer clinical response.

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.

[42] Amygdalin inhibits genes related to cell cycle in SNU-C4 human colon cancer cells.

Shows: Amygdalin at 5 mg/mL changes cell-cycle gene expression in colon cancer cells.

Does not show: High concentration and transcript readout do not prove clinically achievable selectivity or tumour control.

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.

[43] Amygdalin Regulates Apoptosis and Adhesion in Hs578T Triple-Negative Breast Cancer Cells.

Shows: Breast cancer lines show apoptosis and adhesion changes with caspase, Bax/Bcl2 and p38 signals.

Does not show: Cancer-only culture findings lack normal-organ exposure and human prevention endpoints.

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.

[44] Amygdalin Decreases Adhesion and Migration of MDA-MB-231 and MCF-7 Breast Cancer Cell Lines.

Shows: Cancer growth and migration decrease while selected normal skin fibroblasts are spared at the tested concentrations.

Does not show: One normal comparator and culture context cannot define a systemic therapeutic window.

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.

[45] Amygdalin inhibits the growth of renal cell carcinoma cells in vitro.

Shows: Renal cancer cells exposed to 10 mg/mL show growth and cell-cycle effects with cdk1/cyclin B tests.

Does not show: Blockade phenocopy is not a drug-specific rescue; human exposure and normal-organ selectivity remain untested.

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.

[46] Amygdalin delays cell cycle progression and blocks growth of prostate cancer cells in vitro.

Shows: Dose-dependent prostate cell growth/clonogenic inhibition accompanies cell-cycle changes.

Does not show: High mg/mL exposure and cell-specific apoptosis effects limit clinical translation.

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.

[47] Amygdalin Exerts Antitumor Activity in Taxane-Resistant Prostate Cancer Cells.

Shows: Drug inhibits parental and taxane-resistant prostate lines at 10 mg/mL.

Does not show: Related cell programme, not independent clinical trials; exposure and safety bridge absent.

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.

[48] Antitumor Action of Amygdalin on Human Breast Cancer Cells by Selective Sensitization to Oxidative Stress.

Shows: Breast line growth inhibition correlates with oxidative stress at 39–45 mM IC50.

Does not show: Tumour-only correlations and high concentrations do not demonstrate normal-tissue sparing or mediator necessity.

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.

[49] In vitro and in silico anticancer activity of amygdalin on the SK-BR-3 human breast cancer cell line.

Shows: SKBR3 viability/apoptotic proteins change; docking proposes target interactions.

Does not show: Docking is not measured binding or necessity, and culture lacks a human-dose selective window.

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.

[50] Hepatocellular Carcinoma cells: activity of Amygdalin and Sorafenib in Targeting AMPK /mTOR and BCL-2 for anti-angiogenesis and apoptosis cell death.

Shows: Original body reports HepG2/WI38 IC50 differences and synergy with sorafenib.

Does not show: mg/mL exposure and unmatched tissue/proliferation contexts limit selectivity; the source's outdated HCC-treatment assertion is not adopted.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. Selected body sections: 2:Results:paragraph0, 2:Results:paragraph15, 2:Results:paragraph25. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1.

[51] Growth of Renal Cancer Cell Lines Is Strongly Inhibited by Synergistic Activity of Low-Dosed Amygdalin and Sulforaphane.

Shows: Body shows little monodrug growth effect at 1 mg/mL but stronger sulforaphane combinations, without induced apoptosis.

Does not show: Low dose is relative to culture, not verified patient exposure; no in vivo clinical feasibility result.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. Selected body sections: 2:3. Results:paragraph0, 2:3. Results:paragraph1, 2:3. Results:paragraph3. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1.

[52] Magnetically Directed Enzyme/Prodrug Prostate Cancer Therapy Based on β-Glucosidase/Amygdalin.

Shows: Magnetic enzyme-loaded nanoparticles permit targeted activation and mouse tumour-growth inhibition.

Does not show: Different experimental platform; body reports AST, CK and LDH increases despite no obvious histological organ injury.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. Selected body sections: 2:Results:paragraph7, 2:Results:paragraph21, 2:Results:paragraph28. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1.

[53] Targeted nanocomposite delivery system of amygdalin using chitosan/reduced graphene oxide-zinc oxide/hyaluronic acid for treatment of head and neck squamous cell carcinoma.

Shows: Loaded nanocomposite improves uptake and viability/migration assays in selected HNSCC cells.

Does not show: Carrier components and release may contribute; in vitro platform is not clinical free-drug benefit or complete systemic safety.

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.

[54] Formation of amygdalin/β-cyclodextrin derivatives inclusion complexes for anticancer activity assessment in human cervical carcinoma HeLa cell line.

Shows: Cyclodextrin complexes improve HeLa MTT cytotoxicity compared with free compound.

Does not show: Formulation culture assay does not establish a human therapeutic window or clinical 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.

[55] Amygdalin/chitosan-polyvinyl alcohol/cerium-tannic acid hydrogel as biodegradable long-time implant for cancer recurrence care applications: An in vitro study.

Shows: Amygdalin hydrogel causes colorectal-cell apoptosis alongside material/biocompatibility tests.

Does not show: Proposed implant has no demonstrated postoperative recurrence-prevention or clinical safety outcome.

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.

[56] Amygdalin as a chemoprotective agent in co-treatment with cisplatin.

Shows: Combination with cisplatin protects selected normal cells while enhancing cancer-cell inhibition.

Does not show: Specific cell types and schedules, not organismal safety or patient quality-of-life trial.

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.

[57] Antimutagenic, Antirecombinogenic, and Antitumor Effect of Amygdalin in a Yeast Cell-Based Test and Mammalian Cell Lines.

Shows: Yeast recombination/genotoxicity and mammalian cell comparisons suggest context-specific effects.

Does not show: Yeast surrogates are not human cancer prevention; native organism dependency and exposure bridge absent.

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.

[58] Phytochemical characterization and biological activity of apricot kernels' extract in yeast-cell based tests and hepatocellular and colorectal carcinoma cell lines.

Shows: Characterized kernel extract shows different toxicity in cancer and selected normal lines.

Does not show: Thousands of constituents and multiple cyanogenic glycosides prevent assigning all activity to pure amygdalin.

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.

[59] Antioxidant, Anti-Inflammatory, and Anti-Cancer Properties of Amygdalin Extracted from Three Cassava Varieties Cultivated in Benin.

Shows: Cassava extract tests report antioxidant, inflammatory and limited rat cancer-model changes.

Does not show: Mixed ingredients and chemical identity need scrutiny; no isolated human amygdalin therapeutic 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.

[60] Evaluating the therapeutic potential of amygdalin: Cytotoxic and antimicrobial properties.

Shows: Monocytic leukemia line shows concentration/time-dependent viability and apoptosis changes.

Does not show: No matched normal/exposure comparison; broad claims about incurable leukemia and clinical application are not adopted.

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.

[61] Use of Alternative Medicine for Cancer and Its Impact on Survival.

Shows: Matched NCDB patients using alternative treatment without conventional care have higher mortality association.

Does not show: Selection, residual confounding and coarse modality coding; not a laetrile-specific randomized hazard ratio.

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.

[62] Complementary Medicine, Refusal of Conventional Cancer Therapy, and Survival Among Patients With Curable Cancers.

Shows: Complementary-use mortality association attenuates and becomes nonsignificant after treatment refusal/delay adjustment.

Does not show: Shared NCDB and observational mediation; not proof every complementary use causes death.

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.

[63] Outcome analysis of breast cancer patients who declined evidence-based treatment.

Shows: Matched breast-cancer primary-treatment refusal is associated with worse survival.

Does not show: Body confirms age, year and stage matching without full TNM; not laetrile-specific and selection remains.

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. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

[64] Outcomes of breast cancer in patients who use alternative therapies as primary treatment.

Shows: Patients delaying or refusing standard breast-cancer care frequently progress or relapse.

Does not show: Small chart series; some mortality figures are projections rather than observed causal deaths and modalities differ.

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.

[65] Special report on Laetrile: the NCI Laetrile Review. Results of the National Cancer Institute's retrospective Laetrile analysis.

Shows: Blinded oncologists judge six responses in solicited favorable laetrile treatment courses.

Does not show: Only claimed-benefit cases solicited; nonrandomized courses, incomplete records and cointerventions prevent definite attribution.

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.

[66] In Vitro and In Vivo Evaluation of the Effectiveness and Safety of Amygdalin as a Cancer Therapy.

Shows: Original body compares topical loaded niosomal gel, blank gel, oral free compound and tamoxifen in small DMBA skin-tumour rat groups.

Does not show: Formulation/route differ, and selected local histology does not establish systemic cyanide safety, tumour pharmacokinetics or human benefit.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. Selected body sections: 1:2. Results and Discussion:paragraph8, 1:2. Results and Discussion:paragraph9, 1:2. Results and Discussion:paragraph10, 2:3. Materials and Methods:paragraph13, 2:3. Materials and Methods:paragraph14, 2:3. Materials and Methods:paragraph15, 2:3. Materials and Methods:paragraph16, 2:3. Materials and Methods:paragraph17, 2:3. Materials and Methods:paragraph18, 2:3. Materials and Methods:paragraph19, 2:3. Materials and Methods:paragraph20, 2:3. Materials and Methods:paragraph21. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1.

[67] Evaluation of amygdalin-loaded alginate-chitosan nanoparticles as biocompatible drug delivery carriers for anticancerous efficacy.

Shows: Alginate-chitosan particles sustain release and increase H1299 cell cytotoxicity compared with free drug.

Does not show: Abstract claims normal-tissue protection but supplied results do not detail an independent normal-tissue clinical endpoint; no whole-paper audit.

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.

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
• adversarial-findings.json
• codex-correction-provenance.json
• consensus-validation.json
• edition-approval-audit.json
• laetrile-cancer-treatment.review.md
• laetrile-cancer-treatment.scorecard.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: 67 · load-bearing claims: 72 · verified: 72 · unverified: 0 · contradicted: 0 · schema 5.3