In plain terms

The claim being assessed: The immune system actively sculpts tumors, driving all developing tumors through an obligate sequence of elimination, equilibrium and escape while selecting immune-resistant variants.

Where the evidence stands: Β — Association only. The pattern is real and repeatable, but nobody has shown it causes anything.

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

Strongest evidence for it: SC6 gates the registered universal scope and remains Β. Human phase histories are chiefly inferred from sampled lesions/genomes without an all-etiology causal inception dataset. This does not cap scoped immune elimination or presentation escape, which reach Ζ through matching genetic and functional contrasts. Low-antigen KPC, innate-only editing and early evasion delimit the universal inference.

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

Immune cells can remove tumors and select resistant variants. This review asks whether every developing tumor must pass through elimination, equilibrium and escape.

Implications for health and treatment: SC6 gates the registered universal scope and remains Β. Human phase histories are chiefly inferred from sampled lesions/genomes without an all-etiology causal inception dataset. This does not cap scoped immune elimination or presentation escape, which reach Ζ through matching genetic and functional contrasts. Low-antigen KPC, innate-only editing and early evasion delimit the universal inference.

Limits for individual care: 63 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 complete abstracts/status records and selected original partial-body passages. No 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

Figures

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

Full assessment

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

Sources: 63 primary reports; claims: 70; verified: 70; unverified: 0. Verification is bounded attribution. Access: 8 partial-body and 55 complete abstract/status examinations. Literature cutoff: 2026-10-08; source examination UTC date: 2026-10-09. 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. 7 corrections applied (see ADVERSARIAL REVIEW AND APPROVAL). Status: Approved as revised.

AT A GLANCE

Immune selection is causal in defined cancers; an obligatory elimination–equilibrium–escape sequence across every developing human tumor is unsupported.

Deciding experiment: Prospectively barcode autochthonous clones across predefined high/low-antigen etiologies; serially trace immune contact, viable occult persistence and outgrowth with delayed antigen-specific interruption and physiological restoration; pair with untreated human surveillance.

Belief versus evidence: SC6 gates the registered universal scope and remains Β. Human phase histories are chiefly inferred from sampled lesions/genomes without an all-etiology causal inception dataset. This does not cap scoped immune elimination or presentation escape, which reach Ζ through matching genetic and functional contrasts. Low-antigen KPC, innate-only editing and early evasion delimit the universal inference.

Best-supported rival: Etiology-stratified immune editing. KPC sustained prediagnostic T-cell depletion with repeated verification and an engineered OVA positive control separates universal necessity from baseline-antigen-dependent editing (PMID27642636).

COMPETING MECHANISTIC MODELS

Rank 1: Etiology-stratified immune editing (Ζ). Baseline antigenicity, lineage and immunity determine editing routes. MCA/antigen causal contrasts, KPC and human cSCC/NSCLC (PMID22318521,27642636,42276071). Earliest human transition frequencies remain unresolved; scope cannot be chosen after outcomes.

Rank 2: Antigen-presentation attrition (Ζ). Mutations can persist while HLA/B2M presentation or recognition fails. Autologous recognition, model MHC selection and organoid assays (PMID22318517,25294904,40614739). Different defects/shared cohorts do not verify every HLA event.

Rank 3: Immunological ignorance (Ε). Expressed antigens need not recruit effective specific recognition. Retained melanoma epitopes without reactive clones, vaccine-broadened responses and KPC positive control (PMID31685621,27642636). Sampling may miss reactive clones; no detected response is not no historical pressure.

Rank 4: Innate/myeloid pre-emption (Ε). Lineage and early suppressive niches permit outgrowth before effective adaptive engagement. Innate-only editing and IL1/SOX2 precancer intervention (PMID22927549,42446924). No prior brief equilibrium is not traced across all systems.

Rank 5: Universal obligatory three phases (Β). All developing human tumors require sequential adaptive elimination, equilibrium and escape. Classical mouse phases are a serious steelman (PMID11323675,18026089,22318517). No causal human all-phase dataset; low-antigen and early-escape contrasts constrain extrapolation.

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

The immune system actively sculpts tumors, driving all developing tumors through an obligate sequence of elimination, equilibrium and escape while selecting immune-resistant variants.

Use the index universal-sequence form; SC6 gates that scope. Elimination, equilibrium and different escape substrates are scored separately, including immunogenic versus low-burden etiologies. HLA loss is not evidence of point-mutational neoantigen depletion.

Selection: public-interest rank 37 from the editorial ranking compiled 2026-09-23; categorical prior verdicts were visible. Private fresh review; earlier documents are preserved.

Evidence and causal interpretation

The review separates experimentally identified immune functions from the registered claim that every developing human tumor must traverse three phases. The 63-report corpus includes primary negative immune interventions, computational null tests and recent positive human selection studies. Evidence count does not mean 63 independent causal confirmations.

Lymphocyte/IFNgamma loss alters tumor emergence (PMID11323675), and DC-restricted type-I-IFN sensitivity is required for cross-priming/rejection (PMID21930769). Perforin, NKT and TRAIL work extends the effector picture. Protection differs among lymphomas: v-Abl/bcl2 respond to perforin loss while aggressive Myc does not (PMID19196996). Mammary onset/count changes can reflect NK rather than CD8 effects without survival benefit (PMID17545627). Defined immune surveillance is real, but the effectors and phenotypes are not interchangeable.

Koebel reports prolonged restraint of occult transformed cells and later spontaneous edited escape in chemical carcinogenesis (PMID18026089). Its abstract explicitly permits independent or sequential phases. The body was unavailable, so exact intervention-release frequencies are not claimed as checked. Adoptive CTL-induced IFNgamma cytostasis (PMID25127875) is a different therapeutic state rather than independent universal spontaneous equilibrium.

Matsushita identifies endogenous mutant spectrin-beta2 in hypermutated Rag2-derived sarcomas, restores rejection with mutant but not WT antigen, and tracks selection of pre-existing antigen-negative clones (PMID22318521). Transplantation and the carcinogen context delimit that causal result. DuPage uses autochthonous KRAS/p53 tumors but introduces model epitopes; tolerance, depletion and antigen/MHC loss identify selection within that system (PMID22318517). Neither establishes phase order across all human etiologies.

KPC supplies a direct boundary test (PMID27642636). Juvenile sustained T-cell depletion is repeatedly verified, yet spontaneous tumor timing and survival remain statistically unchanged; depleted-donor tumors also grow in competent recipients. OVA introduction restores CD8-dependent rejection and mixed-population antigen-negative selection, including in the pancreatic microenvironment. This positive control argues against intrinsic immune incapacity, while the result remains a low-predicted-antigen mouse context rather than all human pancreatic disease.

Untreated human mutation-depletion evidence is disputed. Selection-free mutations matched for trinucleotide context reproduce apparent depletion, and finer correction diminishes signals (PMID31768072). The null challenges an uncorrected metric, not all selection routes. Wu reports distribution-based phase scores (PMID35486454), whereas the 2025 original comparison finds large classification differences dependent on references and neoantigen criteria (PMID40132544). Quantitative model/data fits (PMID32929288) constrain hypotheses but cannot alone identify causal phases.

The 2026 treatment-naive cSCC study is a substantive positive human test (PMID42276071). High-infiltrate intact-host tumors restrict strong-binding and immunogenic-like antigens toward lower clonal frequencies with purity/growth controls. Predictions use retained HLA alleles, and HLA loss does not differ among strata. Host immunosuppression is observational and age differs between groups, so the inference is a defined cohort result rather than every tumor or verified elimination of each predicted peptide. It does not establish equilibrium or erase the pan-cancer null.

Presenter attrition is distinct from antigen mutation loss. HLA LOH occurs in 36/90 untreated NSCLCs, often subclonally, with more predicted neoantigens binding lost alleles (PMID29107330); the coding mutation may remain. Rosenthal separately identifies copy-number loss, expression depletion and promoter methylation (PMID30894752). Methylation supports a selection hypothesis without functional demethylation/recognition rescue. HLA transcriptomic and organoid–TIL work refines heterogeneity (PMID39358601,40614739), while shared TRACERx samples do not provide independent population replications.

Epigenetic immune mechanisms must be named precisely. DuPage Aza restores reporter antigen expression (PMID22318517). SETDB1 loss induces TE peptides and therapy sensitivity requiring MHC-I/CD8, but little untreated growth effect in the examined models (PMID33953401). Developmental PRC2 can pre-silence antigen processing (PMID31564637), while NSD1 loss shows retroelement expression can coexist with immune exclusion (PMID36206767). These findings do not prove selective rescue of every silenced natural mutant antigen.

Alternatives include expressed melanoma epitopes without detected reactive TCR clones (PMID31685621), innate editing in adaptive-deficient hosts (PMID22927549) and low NK ligands in SCLC (PMID33495232). Early CRC multiomics infers escape near transformation (PMID41193656); 2026 oral precancer work causally tests IL1-driven myeloid suppression (PMID42446924). Checkpoint-acquired JAK/B2M resistance informs therapy escape (PMID27433843,29070816), not untreated initiation. The tournament favors etiology-defined selection with multiple escape substrates; the universal obligate sequence stays Β.

HYPOTHESIS STATE EVALUATION FRAMEWORK (HSEF v5.3)

Subclaim matrix and aggregation

Preregistered aggregation: scope-matched. SC6 gates the registered universal scope and remains Β. Human phase histories are chiefly inferred from sampled lesions/genomes without an all-etiology causal inception dataset. This does not cap scoped immune elimination or presentation escape, which reach Ζ through matching genetic and functional contrasts. Low-antigen KPC, innate-only editing and early evasion delimit the universal inference. Sensitivities: weakest-link Β; primary-subclaim Β; scope-matched Β.

Decisive-perturbation inventory and causal tally

[2] Both antigen-specific CD4 and CD8 responses contribute to rejection with local class-II antigen requirements. Does not show: scoped antigens and models; not all human tumors.

[6] Lymphocytes and IFNgamma suppress tumors and select less immunogenic variants. Does not show: broad immune deficiency and specific etiologies; every tumor sequence not established.

[30] Abstract reports prolonged immune-controlled occult transformed cells and edited spontaneous escape. Does not show: full body unavailable in this examination; no animal-level depletion/spontaneous-frequency audit; phases can act independently.

[31] Mutant spectrin-beta2 recognition, antigen restoration and mixed-clone tracking establish T-cell-dependent selection. Does not show: transplanted hypermutated mouse clones; not every untreated human tumor.

[32] DC type-I IFN response is required for cross-priming and tumor rejection; NK/myeloid contrasts exclude broad effector attribution. Does not show: highly immunogenic challenge models; no universal initiation phases.

[33] Immune/tolerance contrasts and antigen or presentation loss establish selection; Aza restores reporter antigen expression. Does not show: engineered antigens and RNA/reporter restoration; no universal endogenous neoantigen rescue across human tumors.

[57] Physiological NKT transfer protection requires CD1d, downstream NK/CD8, IFNgamma and perforin. Does not show: specific carcinogen challenge and pathway; NKT is not proof of general conventional-T sequence.

Tally: 7 discriminating contrasts. Null and negative contrasts are discriminators, not positive threshold crossings. Reports and replications are not independent causal lines. Shared cohorts/platforms and multiple outcomes do not constitute independent causal lines.

Cap audit, precedence and Β/Γ/Δ boundary

Ζ for scoped elimination/presentation reflects matching convergent endogenous/functional classes; transplantation caps do not erase them. Those same sources cannot upgrade all-human phase order or the distinct mutant-locus reversible-rescue endpoint. Negative contrasts never enter positive-causality tallies.

Gamma hard-rule audit: Engineered antigen/TCR challenge demonstrates scoped capacity, not naturally obligatory phases. Transplantation and checkpoint-acquired resistance differ from untreated autochthonous initiation. HLA LOH removes a presenter rather than the antigen-bearing coding mutation. Methylation association and reporter re-expression require a separate immune-recognition rescue. Broad immune defects require antigen-specific and cell-restricted controls.

SC5 cap: “Mechanistically plausible cell culture or transplantation-linked models without endogenous in vivo dependency cap at Γ.” Aza restores DuPage reporter antigen expression; Rosenthal methylation associates with non-expression (PMID22318517,30894752). A naturally silenced mutant human locus with presentation/recognition rescue is not established by the examined passages. SETDB1 TE-antigen/MHC/CD8 epistasis under therapy is a distinct broader mechanism (PMID33953401). Exact SC5 stays Γ.

SC6 cap: “Association-only cancer claims cap at Β.” No all-etiology human inception study identifies the obligatory sequence. Sustained validated juvenile T-cell depletion leaves KPC onset/survival intact, with OVA positive control (PMID27642636); early CRC escape and innate editing delimit phase generalization (PMID41193656,22927549). Human universal sequence inference remains Β.

SC1 field-valid definition: “ - Endogenous genetic manipulation in autochthonous models - Patient-derived system perturbation with matched rescue AND in vivo confirmation - Clonal tracking demonstrating persistence of dependency through selection pressure” Positive full-scope result: yes. Lymphocyte/IFNgamma loss, DC-specific IFNAR1 deletion, antigen tolerance and effector transfer change tumor emergence/rejection (PMID11323675,21930769,22318517,12093876). Orthogonal contrasts support Ζ in specified etiologies. Perforin-insensitive Myc lymphoma and KPC outgrowth prevent universal Η.

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: yes. Koebel reports transformed occult cells under prolonged immune restraint and later edited spontaneous escape in chemical carcinogenesis (PMID18026089), supporting Δ within that model. Only its complete primary abstract was accessible; exact body/supplement intervention frequencies remain unverified. Therapeutic CTL-induced arrest (PMID25127875) is not independent spontaneous equilibrium.

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: yes. Matsushita tracks selection of pre-existing mutant-spectrin-negative clones and restores antigen-dependent rejection (PMID22318521). Treatment-naive 2026 cSCC adds clonal restriction with purity/growth and HLA controls (PMID42276071). Context-matched nulls undermine uncorrected pan-cancer depletion (PMID31768072). This supports Ε in mapped antigen/clonal contexts, not all untreated tumors.

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: yes. Autochthonous model-antigen/MHC contrasts, serial human autologous recognition/B2M loss, clonal organoid–TIL assays and allele-specific HLA evolution converge (PMID22318517,25294904,40614739,29107330). Defined presentation escape reaches Ζ. Not every HLA event is functionally restored or isolated from passenger CIN.

SC5 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. Aza restores DuPage reporter antigen expression; Rosenthal methylation associates with non-expression (PMID22318517,30894752). A naturally silenced mutant human locus with presentation/recognition rescue is not established by the examined passages. SETDB1 TE-antigen/MHC/CD8 epistasis under therapy is a distinct broader mechanism (PMID33953401). Exact SC5 stays Γ.

SC6 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. No all-etiology human inception study identifies the obligatory sequence. Sustained validated juvenile T-cell depletion leaves KPC onset/survival intact, with OVA positive control (PMID27642636); early CRC escape and innate editing delimit phase generalization (PMID41193656,22927549). Human universal sequence inference remains Β.

Upward threshold ladder

SC1 Δ: “Requires at least one meaningful causal perturbation showing the proposed mechanism contributes to initiation, maintenance, or relapse.” Satisfied: yes.

SC1 Ε: “Requires context-specific causal validity with explicit limits. Cancer stem cells often belong here when a stem-like hierarchy exists in some tumor contexts but fails as a universal model.” Satisfied: yes.

SC1 Ζ: “Requires: - Independent causal confirmation across models - Reduced viability of major alternatives - Relevance beyond one assay class” Satisfied: yes.

SC1 Η: “Requires: - Strong endogenous evidence - Therapeutic or mechanistic consequence - Reproducibility across disease settings within the claimed scope” Satisfied: no. Prospectively replicate immune-specific loss/restoration across fixed etiologies without changing carcinogenesis or viability.

SC2 Δ: “Requires at least one meaningful causal perturbation showing the proposed mechanism contributes to initiation, maintenance, or relapse.” Satisfied: yes.

SC2 Ε: “Requires context-specific causal validity with explicit limits. Cancer stem cells often belong here when a stem-like hierarchy exists in some tumor contexts but fails as a universal model.” Satisfied: no. Independent lineage-confirmed occult persistence with delayed immune interruption/restoration; finish original methods/supplements.

SC3 Δ: “Requires at least one meaningful causal perturbation showing the proposed mechanism contributes to initiation, maintenance, or relapse.” Satisfied: yes.

SC3 Ε: “Requires context-specific causal validity with explicit limits. Cancer stem cells often belong here when a stem-like hierarchy exists in some tumor contexts but fails as a universal model.” Satisfied: yes.

SC3 Ζ: “Requires: - Independent causal confirmation across models - Reduced viability of major alternatives - Relevance beyond one assay class” Satisfied: no. Validate presented mutant epitopes and TCR recognition, use powered sequence-context nulls, and trace antigen-clone loss separately from presenter attrition.

SC4 Δ: “Requires at least one meaningful causal perturbation showing the proposed mechanism contributes to initiation, maintenance, or relapse.” Satisfied: yes.

SC4 Ε: “Requires context-specific causal validity with explicit limits. Cancer stem cells often belong here when a stem-like hierarchy exists in some tumor contexts but fails as a universal model.” Satisfied: yes.

SC4 Ζ: “Requires: - Independent causal confirmation across models - Reduced viability of major alternatives - Relevance beyond one assay class” Satisfied: yes.

SC4 Η: “Requires: - Strong endogenous evidence - Therapeutic or mechanistic consequence - Reproducibility across disease settings within the claimed scope” Satisfied: no. Physiologically restore lost HLA alleles in matched clones; test autologous killing and tumor control with CIN passenger controls.

SC5 Δ: “Requires at least one meaningful causal perturbation showing the proposed mechanism contributes to initiation, maintenance, or relapse.” Satisfied: no. Selectively reverse endogenous mutant-antigen silencing, recover peptide presentation and recognition, then restore silence to reverse control.

SC6 Δ: “Requires at least one meaningful causal perturbation showing the proposed mechanism contributes to initiation, maintenance, or relapse.” Satisfied: no. Prospective untreated inception cohorts across predefined etiologies must trace all causal phase transitions and count bypasses.

Upper ladder: Immune genetics, antigen tolerance/rescue, cell-restricted pathways, functional patient assays and host immunosuppression are live upper-ladder routes without RCT-only requirements. They strengthen narrow claims while universal phase order and endogenous mutant-locus rescue retain separate missing endpoints.

Independence: Convergence spans immune genetics, DC-specific cross-priming, autochthonous antigen/tolerance and human functional recognition, addressing different errors. TRACERx and TCGA are repeatedly shared; independent cSCC is observational. No report count creates universal phase order.

Primary immune-genetic tumor contrasts [6, 57]: Incidence and transfer contrasts exclude mere infiltration association; shared carcinogen pathways remain one family.

Cell-restricted cross-priming [32]: DC IFNAR1/chimeras narrow systemic immune-deficiency alternatives.

Autochthonous antigen/tolerance [33]: Induction, depletion and tolerance isolate recognition; engineered epitopes bound scope.

Patient-derived functional recognition [50, 17]: Autologous and organoid assays go beyond genomic association; culture selection and small samples remain.

Mutant antigen restoration/clonal tracking [31]: WT/mutant rescue and mixtures isolate selection in one carcinogen clone family.

Falsification and pitfalls

Perturbing the proposed feature leaves the claimed output intact. KPC verified sustained T-cell depletion leaves onset/survival intact with OVA positive control (PMID27642636). Scoped contradiction of universal adaptive necessity. Additional downgrade: False.

Phase labels reified. Scores/morphological stages do not observe the same clone traversing all phases.

Presenter loss equals antigen loss. HLA/B2M and coding mutation depletion have different substrates.

Treatment generalized to untreated initiation. Checkpoint-acquired JAK/B2M escape stays treatment-scoped.

Shared cohorts treated as independence. TRACERx/TCGA repeatedly reuse data.

Provider substitution/date accepted. Consensus Borden date error and Wu preprint were corrected with journal records.

Adversarial steelman: above, then below

One rung higher, Γ: classical mouse phases, engineered autochthonous antigens and human lesion/genomic trajectories make the expansive sequence mechanistically conceivable.

One rung lower, Α: the whole account is a narrative without meaningful evidence. That misses experimentally demonstrated selected-phase causality.

Reject Α because real antigen-dependent selection and occult restraint exist. Reject Γ for the exact every-human obligatory gate because model-specific plausibility does not identify that scope and faces endogenous counterexamples. Keep Ζ for matching narrow functions.

Final score, causal direction and M/S/D

SC6 gates the registered universal scope and remains Β. Human phase histories are chiefly inferred from sampled lesions/genomes without an all-etiology causal inception dataset. This does not cap scoped immune elimination or presentation escape, which reach Ζ through matching genetic and functional contrasts. Low-antigen KPC, innate-only editing and early evasion delimit the universal inference.

Immune pressure can restrain and sculpt already transformed variants; upstream oncogenic and baseline antigen/lineage context remains. All-human initiation necessity plus sufficiency is not shown, so steps 3/4 do not license universal primary-driver status. This role applies to supported scoped selection.

Momentum: Mechanistic only. The named artifacts are completed mechanistic and boundary studies, not experiments now in progress: 2025 timed-antigen and CRC epigenomic work plus the 2026 treatment-naive cSCC and IL1 precancer studies. No experiment addressing the SC6 boundary check — a prospective untreated all-etiology inception cohort tracing phase transitions — was identified as underway, and no unnamed pending trial is claimed.

Stability: Stable. No named pending result is asserted to move the grade imminently.

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

PROPOSED DECISIVE EXPERIMENTS

Prospectively barcode autochthonous clones across predefined high/low-antigen etiologies; serially trace immune contact, viable occult persistence and outgrowth with delayed antigen-specific interruption and physiological restoration; pair with untreated human surveillance.

Readout: Clone survival, occult proliferation balance, antigen/presenter substrate, autologous recognition and tumor outcomes. Controls: Antigen-negative lineages, tolerant/depleted/intact hosts, restoration, CIN-matched HLA passengers, mutation-signature nulls and blinded phase labels. Supporting: Every predefined developing clone traverses causally verified sequential phases with matching interruption/restoration effects. Falsifying: Tracked clones bypass the required adaptive phase or inherit early escape without preceding causal equilibrium.

RESEARCH TIMELINE

KNOWLEDGE GAPS

Trace phases within the same untreated clone.

Prespecify antigen/etiology scope and count bypasses.

Replicate and fully audit occult-equilibrium experiments.

Validate presented epitopes and TCR recognition beyond predictions.

Control sequence context, purity, growth and detection power.

Restore HLA with matched CIN passenger controls.

Reverse endogenous mutant-antigen methylation with functional recognition rescue.

Track early myeloid pre-emption versus brief equilibrium.

Separate innate ligand editing and retained helper antigens.

Audit errata, inaccessible bodies and supplements before complete-paper publication claims.

METHODS AND CONSENSUS CONTRIBUTION

Consensus was used through the signed-in browser for discovery, counterevidence, research-gap and timeline suggestions, and attempted checks of specific indexed full-text passages. Provider statements about Methods/Results and any returned passages were compared with original-paper passages where available; failed retrievals, source substitutions and unverified aspects are recorded in the case reconciliation. Successful original-passage comparisons are distinguished from provider summaries and assertions of access. Provider screening counts, citation counts and meters were not used as primary counts or HSEF scores. This source log records access depth; it does not claim complete-paper or supplement review. The monthly API allowance was conserved: zero Consensus API calls in this batch.

REVIEWER NOTES

63 primary reports; eight selected original partial-body examinations and 55 complete abstract/status examinations. Original computational comparisons are primary analyses, not interventions. Koebel body unavailable; IFN erratum lacks resolved PMID; organoid correction details remain unaudited. No independent blinded or complete-paper review.

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

ADVERSARIAL REVIEW AND APPROVAL

Authoring model: Sol 6.1. Adversarial reviewer: Claude Opus 5.5 (Anthropic). Date: 2026-10-14. Edition reviewed: the 9 October 2026 revised edition, as it now stands.

Scope of checks. All 63 REFERENCES PMIDs plus the two that appear only in the body (36993286, 42155448) were resolved against PubMed and matched on first author, title, journal, year and DOI: 65/65 resolved, no wrong identifiers, no retracted publication and no expression of concern. Nine load-bearing claims were spot-checked against the source abstracts and, for PMID 29107330, against the retrieved full text held in the originating package: PMIDs 27642636, 18026089, 29107330, 42276071, 31768072, 19196996, 17545627, 22927549 and 41193656. Scoring logic was re-derived from references/hsef-framework.md, SKILL.md Steps 2-3 and 8, references/evidence-rubrics/02-cancer-biology.md and the Cancer immunoediting (three-phase model) entry in references/hypothesis-index.md. The revised edition was also diffed against the 2026-10-08 edition to separate inherited defects from revision regressions, and the renderer, provenance and confidence artifacts reported across this batch were each checked here.

Findings

Rejected findings (checked, no change)

• The Oct-9 revision regression reported elsewhere in this batch is absent here, but a different one was found (F6). The revision spliced in one block, "Consensus provenance reconciliation — 9 October 2026", and changed nothing else; a line-by-line diff against the 2026-10-08 edition shows no other difference. The preserved ## METHODS AND CONSENSUS CONTRIBUTION section already states that Consensus was used through the signed-in browser with zero API calls in the batch, so the spliced block does not contradict it. There is no preserved sentence asserting that no Consensus work was performed.
• No Consensus-notes contradiction. The preserved ## METHODS AND CONSENSUS CONTRIBUTION and ## REVIEWER NOTES sections contain no sentence denying Consensus use, so the splice created no self-contradiction of that kind here.
• "Best-supported rival" renders the correct model. The score card marks the rank-1 model is_nominal: false and the nominal "Universal obligatory three phases" at rank 5; the AT A GLANCE prose names the rank-1 rival, etiology-stratified immune editing, so the renderer bug reported elsewhere did not fire here.
• Other provenance clauses. The Selection line (public-interest rank 37 from the 2026-09-23 editorial ranking, prior categorical verdicts visible, earlier documents preserved) and the zero-API-call statement were each checked against the package and apply to this case.
• Headline letter Β and scope-matched aggregation. Re-derived independently and correct. The index prescribes scope-matched; the registered claim is the universal obligate sequence across all developing human tumors; SC6 is the only subclaim written at that scope and sits at Β under the rubric's "Association-only cancer claims cap at Β". The Ζ on SC1 and SC4 is scope-restricted to defined host genotypes and to presentation escape, and correctly does not carry the headline — this is the "scope trap" the index names for this entry, and the review avoids it. The reported alternatives are right too: weakest-link takes the lowest of Ζ/Δ/Ε/Ζ/Γ/Β, and primary-subclaim with SC6 named returns the same letter.
• The subclaim decomposition matches the index SC1-SC6 verbatim in substance, and all five index competing models are enumerated. index_rule_deviation is correctly empty.
• Sixth competing model not enumerated. The index adjudication notes say a chromosomal-instability-passenger model "is worth adding whenever SC4 is the live question", and SC4 is live here at Ζ. The review does not rank it as a sixth model, but it does engage the confounder directly — SC4's decision records that not every HLA event is "isolated from passenger CIN", and the decisive-experiment controls list "CIN-matched HLA passengers". The index wording is advisory rather than mandatory and the substance is addressed, so no model was added; flagged for the lead rather than changed.
• Causal tally phrasing. The tally reads "7 discriminating contrasts" and does not separately count positive qualifying study units, which the sibling reviews in this corpus do. The score card's causal_tally_audit matches the document exactly (7 claimed, 7 after audit), and the Ζ derivations for SC1 and SC4 rest on the independence basis and the threshold blocks rather than on this tally, so nothing is double-counted or missing. Loose relative to house practice, not an error.
• PMID 42155448 and PMID 36993286. The first is the published erratum to reference 17 (PMID 40614739) and the second is the bioRxiv preprint of reference 20 (PMID 38815010). Neither is counted as a primary source: they appear only in the Access/status lines of the corresponding cards, as "ErratumIn: PMID42155448" and "UpdateOf: PMID36993286", which is where this package records linked notices.
• Seven references whose PubMed author arrays contain an empty entry mid-list (PMIDs 30894752, 29107330, 39358601, 40614739, 32461698, 38581685 and the erratum 42155448) carry the signature of a dropped consortium author — the TRACERx group in most cases — but none of them is in the first-author position, so the collective-author builder defect does not apply and every reference renders its correct first author.
• Citation years and dropped markup. Reference 7 (PMID 25127875) prints 2015 against a 2014 e-pub year; references 1 and 18 have PubMed titles that drop a superscript "+" and an italicised gene name. All three are rejected by standing convention.
• Score card review_metadata.reviewer reads "Codex". This is house provenance across the corpus and was not edited per case; the authoring model is stated as Sol 6.1 in the banner and in this section.

Residual limitations

The Koebel equilibrium paper (PMID 18026089), which is the sole support for SC2, was examined at abstract depth only, so the intervention-release and spontaneous-escape frequencies behind SC2's Δ are unverified; the review says so, and the abstract does support the two statements it is cited for, including that the three processes "function either independently or in sequence". Four of the eight partial-body sources carry claims this pass checked for abstract-level consistency rather than re-retrieving, and the organoid correction details remain unaudited, as REVIEWER NOTES records.

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

REFERENCES

1. Rozalén C, Sangrador I, Avalle S et al. TIM3+ breast cancer cells license immune evasion during micrometastasis outbreak. Cancer cell. 2025. PMID 40645187. DOI: 10.1016/j.ccell.2025.06.015. Access: abstract_metadata. Study: PMID40645187.

2. Alspach E, Lussier DM, Miceli AP et al. MHC-II neoantigens shape tumour immunity and response to immunotherapy. Nature. 2019. PMID 31645760. DOI: 10.1038/s41586-019-1671-8. Access: abstract_metadata. Study: PMID31645760.

3. Łuksza M, Sethna ZM, Rojas LA et al. Neoantigen quality predicts immunoediting in survivors of pancreatic cancer. Nature. 2022. PMID 35589842. DOI: 10.1038/s41586-022-04735-9. Access: abstract_metadata. Study: PMID35589842.

4. Raju S, Kretzmer LZ, Koues OI et al. NKG2D-NKG2D Ligand Interaction Inhibits the Outgrowth of Naturally Arising Low-Grade B Cell Lymphoma In Vivo. Journal of immunology (Baltimore, Md. : 1950). 2016. PMID 27183590. DOI: 10.4049/jimmunol.1501982. Access: abstract_metadata. Study: PMID27183590.

5. De Mattos-Arruda L, Sammut SJ, Ross EM et al. The Genomic and Immune Landscapes of Lethal Metastatic Breast Cancer. Cell reports. 2019. PMID 31141692. DOI: 10.1016/j.celrep.2019.04.098. Access: abstract_metadata. Study: PMID31141692.

6. Shankaran V, Ikeda H, Bruce AT et al. IFNgamma and lymphocytes prevent primary tumour development and shape tumour immunogenicity. Nature. 2001. PMID 11323675. DOI: 10.1038/35074122. Access: abstract_metadata. Study: PMID11323675.

7. Matsushita H, Hosoi A, Ueha S et al. Cytotoxic T lymphocytes block tumor growth both by lytic activity and IFNγ-dependent cell-cycle arrest. Cancer immunology research. 2015. PMID 25127875. DOI: 10.1158/2326-6066.CIR-14-0098. Access: abstract_metadata. Study: PMID25127875.

8. Chang CH, Qiu J, O'Sullivan D et al. Metabolic Competition in the Tumor Microenvironment Is a Driver of Cancer Progression. Cell. 2015. PMID 26321679. DOI: 10.1016/j.cell.2015.08.016. Access: abstract_metadata. Study: PMID26321679.

9. Gubin MM, Zhang X, Schuster H et al. Checkpoint blockade cancer immunotherapy targets tumour-specific mutant antigens. Nature. 2014. PMID 25428507. DOI: 10.1038/nature13988. Access: abstract_metadata. Study: PMID25428507.

10. Dunn GP, Bruce AT, Sheehan KC et al. A critical function for type I interferons in cancer immunoediting. Nature immunology. 2005. PMID 15951814. DOI: 10.1038/ni1213. Access: abstract_metadata. Study: PMID15951814.

11. Kammertoens T, Friese C, Arina A et al. Tumour ischaemia by interferon-γ resembles physiological blood vessel regression. Nature. 2017. PMID 28445461. DOI: 10.1038/nature22311. Access: abstract_metadata. Study: PMID28445461.

12. Rosenthal R, Cadieux EL, Salgado R et al. Neoantigen-directed immune escape in lung cancer evolution. Nature. 2019. PMID 30894752. DOI: 10.1038/s41586-019-1032-7. Access: full_text_partial. Study: PMID30894752.

13. Zapata L, Pich O, Serrano L et al. Negative selection in tumor genome evolution acts on essential cellular functions and the immunopeptidome. Genome biology. 2018. PMID 29855388. DOI: 10.1186/s13059-018-1434-0. Access: abstract_metadata. Study: PMID29855388.

14. Beauchemin L, Slifker M, Rossell D et al. Characterizing MHC-I Genotype Predictive Power for Oncogenic Mutation Probability in Cancer Patients. Methods in molecular biology (Clifton, N.J.). 2020. PMID 32162254. DOI: 10.1007/978-1-0716-0389-5_8. Access: abstract_metadata. Study: PMID32162254.

15. McGranahan N, Rosenthal R, Hiley CT et al. Allele-Specific HLA Loss and Immune Escape in Lung Cancer Evolution. Cell. 2017. PMID 29107330. DOI: 10.1016/j.cell.2017.10.001. Access: full_text_partial. Study: PMID29107330.

16. Puttick C, Jones TP, Leung MM et al. MHC Hammer reveals genetic and non-genetic HLA disruption in cancer evolution. Nature genetics. 2024. PMID 39358601. DOI: 10.1038/s41588-024-01883-8. Access: abstract_metadata. Study: PMID39358601.

17. Dijkstra KK, Vendramin R, Karagianni D et al. Subclonal immune evasion in non-small cell lung cancer. Cancer cell. 2025. PMID 40614739. DOI: 10.1016/j.ccell.2025.06.012. Access: abstract_metadata. Study: PMID40614739.

18. Pathria P, Gotthardt D, Prchal-Murphy M et al. Myeloid STAT3 promotes formation of colitis-associated colorectal cancer in mice. Oncoimmunology. 2015. PMID 26137415. DOI: 10.1080/2162402X.2014.998529. Access: abstract_metadata. Study: PMID26137415.

19. Hong WF, Zhang F, Wang N et al. Dynamic immunoediting by macrophages in homologous recombination deficiency-stratified pancreatic ductal adenocarcinoma. Drug resistance updates : reviews and commentaries in antimicrobial and anticancer chemotherapy. 2024. PMID 39002266. DOI: 10.1016/j.drup.2024.101115. Access: abstract_metadata. Study: PMID39002266.

20. Houlahan KE, Khan A, Greenwald NF et al. Germline-mediated immunoediting sculpts breast cancer subtypes and metastatic proclivity. Science (New York, N.Y.). 2024. PMID 38815010. DOI: 10.1126/science.adh8697. Access: abstract_metadata. Study: PMID38815010.

21. Zhou Y, Mo S, Cui H et al. Immune-tumor interaction dictates spatially directed evolution of esophageal squamous cell carcinoma. National science review. 2024. PMID 38803565. DOI: 10.1093/nsr/nwae150. Access: abstract_metadata. Study: PMID38803565.

22. Lakatos E, Gunasri V, Zapata L et al. Epigenetically driven and early immune evasion in colorectal cancer evolution. Nature genetics. 2025. PMID 41193656. DOI: 10.1038/s41588-025-02349-1. Access: abstract_metadata. Study: PMID41193656.

23. Cheung JF, Hunt BG, Wahed S et al. Distinct T cell functions enable efficient immunoediting and prevent tumor emergence of developing sarcomas. Cancer cell. 2025. PMID 41043437. DOI: 10.1016/j.ccell.2025.09.005. Access: abstract_metadata. Study: PMID41043437.

24. He S, Sun S, Liu K et al. Comprehensive assessment of computational methods for cancer immunoediting. Cell reports methods. 2025. PMID 40132544. DOI: 10.1016/j.crmeth.2025.101006. Access: abstract_metadata. Study: PMID40132544.

25. Lakatos E, Williams MJ, Schenck RO et al. Evolutionary dynamics of neoantigens in growing tumors. Nature genetics. 2020. PMID 32929288. DOI: 10.1038/s41588-020-0687-1. Access: abstract_metadata. Study: PMID32929288.

26. Roerden M, Castro AB, Cui Y et al. Neoantigen architectures define immunogenicity and drive immune evasion of tumors with heterogenous neoantigen expression. Journal for immunotherapy of cancer. 2024. PMID 39521615. DOI: 10.1136/jitc-2024-010249. Access: abstract_metadata. Study: PMID39521615.

27. AbdulJabbar K, Raza SEA, Rosenthal R et al. Geospatial immune variability illuminates differential evolution of lung adenocarcinoma. Nature medicine. 2020. PMID 32461698. DOI: 10.1038/s41591-020-0900-x. Access: abstract_metadata. Study: PMID32461698.

28. Van den Eynden J, Jiménez-Sánchez A, Miller ML et al. Lack of detectable neoantigen depletion signals in the untreated cancer genome. Nature genetics. 2019. PMID 31768072. DOI: 10.1038/s41588-019-0532-6. Access: full_text_partial. Study: PMID31768072.

29. Castro A, Ozturk K, Pyke RM et al. Elevated neoantigen levels in tumors with somatic mutations in the HLA-A, HLA-B, HLA-C and B2M genes. BMC medical genomics. 2019. PMID 31345234. DOI: 10.1186/s12920-019-0544-1. Access: abstract_metadata. Study: PMID31345234.

30. Koebel CM, Vermi W, Swann JB et al. Adaptive immunity maintains occult cancer in an equilibrium state. Nature. 2007. PMID 18026089. DOI: 10.1038/nature06309. Access: abstract_metadata. Study: PMID18026089.

31. Matsushita H, Vesely MD, Koboldt DC et al. Cancer exome analysis reveals a T-cell-dependent mechanism of cancer immunoediting. Nature. 2012. PMID 22318521. DOI: 10.1038/nature10755. Access: full_text_partial. Study: PMID22318521.

32. Diamond MS, Kinder M, Matsushita H et al. Type I interferon is selectively required by dendritic cells for immune rejection of tumors. The Journal of experimental medicine. 2011. PMID 21930769. DOI: 10.1084/jem.20101158. Access: abstract_metadata. Study: PMID21930769.

33. DuPage M, Mazumdar C, Schmidt LM et al. Expression of tumour-specific antigens underlies cancer immunoediting. Nature. 2012. PMID 22318517. DOI: 10.1038/nature10803. Access: full_text_partial. Study: PMID22318517.

34. O'Sullivan T, Dunn GP, Lacoursiere DY et al. Cancer immunoediting of the NK group 2D ligand H60a. Journal of immunology (Baltimore, Md. : 1950). 2011. PMID 21876033. DOI: 10.4049/jimmunol.1100413. Access: abstract_metadata. Study: PMID21876033.

35. Bolitho P, Street SE, Westwood JA et al. Perforin-mediated suppression of B-cell lymphoma. Proceedings of the National Academy of Sciences of the United States of America. 2009. PMID 19196996. DOI: 10.1073/pnas.0809008106. Access: abstract_metadata. Study: PMID19196996.

36. Wu T, Wang G, Wang X et al. Quantification of Neoantigen-Mediated Immunoediting in Cancer Evolution. Cancer research. 2022. PMID 35486454. DOI: 10.1158/0008-5472.CAN-21-3717. Access: abstract_metadata. Study: PMID35486454.

37. Griffin GK, Wu J, Iracheta-Vellve A et al. Epigenetic silencing by SETDB1 suppresses tumour intrinsic immunogenicity. Nature. 2021. PMID 33953401. DOI: 10.1038/s41586-021-03520-4. Access: full_text_partial. Study: PMID33953401.

38. Zhang SM, Cai WL, Liu X et al. KDM5B promotes immune evasion by recruiting SETDB1 to silence retroelements. Nature. 2021. PMID 34671158. DOI: 10.1038/s41586-021-03994-2. Access: abstract_metadata. Study: PMID34671158.

39. Ritter C, Fan K, Paschen A et al. Epigenetic priming restores the HLA class-I antigen processing machinery expression in Merkel cell carcinoma. Scientific reports. 2017. PMID 28536458. DOI: 10.1038/s41598-017-02608-0. Access: abstract_metadata. Study: PMID28536458.

40. Zhu M, Huang Y, Bender ME et al. Evasion of Innate Immunity Contributes to Small Cell Lung Cancer Progression and Metastasis. Cancer research. 2021. PMID 33495232. DOI: 10.1158/0008-5472.CAN-20-2808. Access: abstract_metadata. Study: PMID33495232.

41. Burr ML, Sparbier CE, Chan KL et al. An Evolutionarily Conserved Function of Polycomb Silences the MHC Class I Antigen Presentation Pathway and Enables Immune Evasion in Cancer. Cancer cell. 2019. PMID 31564637. DOI: 10.1016/j.ccell.2019.08.008. Access: abstract_metadata. Study: PMID31564637.

42. Maecker HL, Yun Z, Maecker HT et al. Epigenetic changes in tumor Fas levels determine immune escape and response to therapy. Cancer cell. 2002. PMID 12204534. DOI: 10.1016/s1535-6108(02)00095-8. Access: abstract_metadata. Study: PMID12204534.

43. Li Y, Goldberg EM, Chen X et al. Histone methylation antagonism drives tumor immune evasion in squamous cell carcinomas. Molecular cell. 2022. PMID 36206767. DOI: 10.1016/j.molcel.2022.09.007. Access: abstract_metadata. Study: PMID36206767.

44. Foidart P, Li Z, Cai X et al. Whole-genome doubling drives immune evasion by silencing antigen presentation. Cancer cell. 2026. PMID 42102810. DOI: 10.1016/j.ccell.2026.04.007. Access: abstract_metadata. Study: PMID42102810.

45. Marzo AL, Lake RA, Lo D et al. Tumor antigens are constitutively presented in the draining lymph nodes. Journal of immunology (Baltimore, Md. : 1950). 1999. PMID 10229818. DOI: none in PubMed record. Access: abstract_metadata. Study: PMID10229818.

46. Enfield KSS, Colliver E, Lee C et al. Spatial Architecture of Myeloid and T Cells Orchestrates Immune Evasion and Clinical Outcome in Lung Cancer. Cancer discovery. 2024. PMID 38581685. DOI: 10.1158/2159-8290.CD-23-1380. Access: abstract_metadata. Study: PMID38581685.

47. Van Waes C, Urban JL, Rothstein JL et al. Highly malignant tumor variants retain tumor-specific antigens recognized by T helper cells. The Journal of experimental medicine. 1986. PMID 2945891. DOI: 10.1084/jem.164.5.1547. Access: abstract_metadata. Study: PMID2945891.

48. Linette GP, Becker-Hapak M, Skidmore ZL et al. Immunological ignorance is an enabling feature of the oligo-clonal T cell response to melanoma neoantigens. Proceedings of the National Academy of Sciences of the United States of America. 2019. PMID 31685621. DOI: 10.1073/pnas.1906026116. Access: abstract_metadata. Study: PMID31685621.

49. Zaretsky JM, Garcia-Diaz A, Shin DS et al. Mutations Associated with Acquired Resistance to PD-1 Blockade in Melanoma. The New England journal of medicine. 2016. PMID 27433843. DOI: 10.1056/NEJMoa1604958. Access: abstract_metadata. Study: PMID27433843.

50. Sucker A, Zhao F, Real B et al. Genetic evolution of T-cell resistance in the course of melanoma progression. Clinical cancer research : an official journal of the American Association for Cancer Research. 2014. PMID 25294904. DOI: 10.1158/1078-0432.CCR-14-0567. Access: abstract_metadata. Study: PMID25294904.

51. Sade-Feldman M, Jiao YJ, Chen JH et al. Resistance to checkpoint blockade therapy through inactivation of antigen presentation. Nature communications. 2017. PMID 29070816. DOI: 10.1038/s41467-017-01062-w. Access: abstract_metadata. Study: PMID29070816.

52. Evans RA, Diamond MS, Rech AJ et al. Lack of immunoediting in murine pancreatic cancer reversed with neoantigen. JCI insight. 2016. PMID 27642636. DOI: 10.1172/jci.insight.88328. Access: full_text_partial. Study: PMID27642636.

53. Borden ES, Castellano D, Kantzos EA et al. Immunoediting restricts clonal neoantigens in primary, treatment-naive human tumors. Immunity. 2026. PMID 42276071. DOI: 10.1016/j.immuni.2026.05.005. Access: full_text_partial. Study: PMID42276071.

54. Smyth MJ, Thia KY, Street SE et al. Perforin-mediated cytotoxicity is critical for surveillance of spontaneous lymphoma. The Journal of experimental medicine. 2000. PMID 10974040. DOI: 10.1084/jem.192.5.755. Access: abstract_metadata. Study: PMID10974040.

55. Street SE, Zerafa N, Iezzi M et al. Host perforin reduces tumor number but does not increase survival in oncogene-driven mammary adenocarcinoma. Cancer research. 2007. PMID 17545627. DOI: 10.1158/0008-5472.CAN-06-4084. Access: abstract_metadata. Study: PMID17545627.

56. Street SE, Trapani JA, MacGregor D et al. Suppression of lymphoma and epithelial malignancies effected by interferon gamma. The Journal of experimental medicine. 2002. PMID 12093877. DOI: 10.1084/jem.20020063. Access: abstract_metadata. Study: PMID12093877.

57. Crowe NY, Smyth MJ, Godfrey DI et al. A critical role for natural killer T cells in immunosurveillance of methylcholanthrene-induced sarcomas. The Journal of experimental medicine. 2002. PMID 12093876. DOI: 10.1084/jem.20020092. Access: abstract_metadata. Study: PMID12093876.

58. Street SE, Hayakawa Y, Zhan Y et al. Innate immune surveillance of spontaneous B cell lymphomas by natural killer cells and gammadelta T cells. The Journal of experimental medicine. 2004. PMID 15007091. DOI: 10.1084/jem.20031981. Access: abstract_metadata. Study: PMID15007091.

59. Takeda K, Smyth MJ, Cretney E et al. Critical role for tumor necrosis factor-related apoptosis-inducing ligand in immune surveillance against tumor development. The Journal of experimental medicine. 2002. PMID 11805143. DOI: 10.1084/jem.20011171. Access: abstract_metadata. Study: PMID11805143.

60. O'Sullivan T, Saddawi-Konefka R, Vermi W et al. Cancer immunoediting by the innate immune system in the absence of adaptive immunity. The Journal of experimental medicine. 2012. PMID 22927549. DOI: 10.1084/jem.20112738. Access: abstract_metadata. Study: PMID22927549.

61. Mascaux C, Angelova M, Vasaturo A et al. Immune evasion before tumour invasion in early lung squamous carcinogenesis. Nature. 2019. PMID 31243362. DOI: 10.1038/s41586-019-1330-0. Access: abstract_metadata. Study: PMID31243362.

62. Laville D, Casteillo F, Yvorel V et al. Immune Escape Is an Early Event in Pre-Invasive Lesions of Lung Squamous Cell Carcinoma. Diagnostics (Basel, Switzerland). 2020. PMID 32708311. DOI: 10.3390/diagnostics10070503. Access: abstract_metadata. Study: PMID32708311.

63. Taner H, Gong W, Lanzel EA et al. IL-1α Expands SLC2A1highSPP1highIFNlow Myeloid Cells to Drive Immune Escape and Malignant Transformation of Precancers. Cancer research. 2026. PMID 42446924. DOI: 10.1158/0008-5472.CAN-25-4416. Access: abstract_metadata. Study: PMID42446924.

Source-specific evidence cards

[1] TIM3+ breast cancer cells license immune evasion during micrometastasis outbreak.

Shows: Tumor TIM3 perturbation links stemness, suppressive gamma-delta cells and metastatic seeding

Does not show: Metastatic context and multifunctional signaling; no obligatory inception sequence

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4, SC6.

[2] MHC-II neoantigens shape tumour immunity and response to immunotherapy.

Shows: Both antigen-specific CD4 and CD8 responses contribute to rejection with local class-II antigen requirements

Does not show: Scoped antigens and models; not all human tumors

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC4.

[3] Neoantigen quality predicts immunoediting in survivors of pancreatic cancer.

Shows: High-quality neoantigen loss and clonal fitness model support selection in long survivors

Does not show: Survivor selection and recurrence; inference rather than experimental untreated initiation

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3, SC6.

[4] NKG2D-NKG2D Ligand Interaction Inhibits the Outgrowth of Naturally Arising Low-Grade B Cell Lymphoma In Vivo.

Shows: NKG2D ligand recognition restrains naturally arising B-cell tumors

Does not show: Innate pathway; grade-dependent human association not universal adaptive sequence

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC6.

[5] The Genomic and Immune Landscapes of Lethal Metastatic Breast Cancer.

Shows: Multiregional HLA loss and TCR patterns accompany lethal metastatic evolution

Does not show: Heavily treated endpoint sampling; no untreated inception or phase causality

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4, SC6.

[6] IFNgamma and lymphocytes prevent primary tumour development and shape tumour immunogenicity.

Shows: Lymphocytes and IFNgamma suppress tumors and select less immunogenic variants

Does not show: Broad immune deficiency and specific etiologies; every tumor sequence not established

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

[7] Cytotoxic T lymphocytes block tumor growth both by lytic activity and IFNγ-dependent cell-cycle arrest.

Shows: IFNgamma neutralization reverses growth arrest; arrest can dominate lysis

Does not show: Adoptive therapy and differing IFNgamma sensitivity; not spontaneous occult equilibrium

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC4.

[8] Metabolic Competition in the Tumor Microenvironment Is a Driver of Cancer Progression.

Shows: Tumor glucose competition can override T-cell protection and impair effector function

Does not show: Metabolic suppression permits antigen-retaining escape; engineered pathway sufficiency

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4, SC6.

[9] Checkpoint blockade cancer immunotherapy targets tumour-specific mutant antigens.

Shows: Mutant-antigen-specific T cells reactivate and vaccination rejects tumors

Does not show: Therapeutically augmented rejection; not untreated universal selection

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC4.

[10] A critical function for type I interferons in cancer immunoediting.

Shows: Host hematopoietic IFN responses protect against primary and transplanted tumors

Does not show: Broad pathway; linked erratum notice lacks resolved PMID/details

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. ErratumIn: PMID not supplied in the PubMed record. Subclaims: SC1.

[11] Tumour ischaemia by interferon-γ resembles physiological blood vessel regression.

Shows: Endothelial responsiveness is necessary and sufficient for induced ischemic regression

Does not show: Induced IFNgamma in established tumors; not nascent elimination or equilibrium

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC4.

[12] Neoantigen-directed immune escape in lung cancer evolution.

Shows: Immune context associates with copy-number antigen loss, HLA dysfunction and neoantigen promoter methylation

Does not show: Shared TRACERx; methylation association does not demonstrate demethylation/recognition rescue

Access/status: Complete primary abstract and linked-status metadata examined. Selected body sections: 1:Results:paragraph9, 1:Results:paragraph10, 1:Results:paragraph13, 1:Results:paragraph15. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3, SC4, SC5, SC6.

[13] Negative selection in tumor genome evolution acts on essential cellular functions and the immunopeptidome.

Shows: Empirical sequence analyses suggest negative selection in immune-exposed regions

Does not show: Null mutation context, expression and prediction biases matter; no immune intervention

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

[14] Characterizing MHC-I Genotype Predictive Power for Oncogenic Mutation Probability in Cancer Patients.

Shows: MHC genotype presentation scores predict mutation probabilities

Does not show: Computational association; genotype and mutation processes may confound selection

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3, SC6.

[15] Allele-Specific HLA Loss and Immune Escape in Lung Cancer Evolution.

Shows: Allele-specific HLA loss is focal, often subclonal, and linked to lost-allele-binding neoantigens

Does not show: Loss of presenter does not delete antigen mutation; cohort overlap and CIN passenger alternatives

Access/status: Complete primary abstract and linked-status metadata examined. Selected body sections: 1:Results:paragraph29, 1:Results:paragraph30, 1:Results:paragraph65. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4, SC6.

[16] MHC Hammer reveals genetic and non-genetic HLA disruption in cancer evolution.

Shows: HLA repression/splicing and LOH refine effective neoantigen burden and metastasis associations

Does not show: Shared cohorts and transcriptomic disruption; not universal sequence or functional restoration

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

[17] Subclonal immune evasion in non-small cell lung cancer.

Shows: Functional immune evasion differs among subclones within each tumor

Does not show: Ex vivo propagation and three patients; linked 2026 correction details not audited

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. ErratumIn: PMID42155448. Subclaims: SC4.

[18] Myeloid STAT3 promotes formation of colitis-associated colorectal cancer in mice.

Shows: Myeloid deletion reduces tumor formation with stromal T-cell activation and residual escape

Does not show: Inflammatory carcinogenesis and pleiotropy; adaptive mechanism inferred partly from transplantation

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC4.

[19] Dynamic immunoediting by macrophages in homologous recombination deficiency-stratified pancreatic ductal adenocarcinoma.

Shows: Spatial macrophage SIGLEC10/CD52 axis associates with repair context

Does not show: Spatial multiomics generates mechanism candidates; no obligatory sequence or definitive causal editing

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

[20] Germline-mediated immunoediting sculpts breast cancer subtypes and metastatic proclivity.

Shows: Germline epitope burden associates with amplified subtype incidence and cold aggressive tumors

Does not show: Germline selection inference; no prospective phase transitions

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. UpdateOf: PMID36993286. Subclaims: SC3, SC6.

[21] Immune-tumor interaction dictates spatially directed evolution of esophageal squamous cell carcinoma.

Shows: Spatial immune context and inferred subclone fitness vary along tumors

Does not show: Retrospective inferred editing and expression; not phase necessity

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3, SC6.

[22] Epigenetically driven and early immune evasion in colorectal cancer evolution.

Shows: Chromatin changes, antigen silencing and exclusion are inferred near transformation with little later editing

Does not show: Timing inferred from sampled extant lesions; early escape contradicts assuming a long mandatory equilibrium

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC5, SC6.

[23] Distinct T cell functions enable efficient immunoediting and prevent tumor emergence of developing sarcomas.

Shows: Timed antigen removal and T-cell/IFNgamma interruption separate direct and bystander early elimination

Does not show: Engineered antigens; antigen-negative cells also exist without T cells; silencing alone not sufficient

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC3, SC5.

[24] Comprehensive assessment of computational methods for cancer immunoediting.

Shows: Editing estimates vary widely with method, neoantigen criteria and reference data

Does not show: Original method comparison not causal immune editing; union of estimates is not truth validation

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

[25] Evolutionary dynamics of neoantigens in growing tumors.

Shows: Selection/escape model predicts neoantigen-frequency patterns and matches cohort data

Does not show: Model fit can be nonidentifying; no intervention; shared public datasets

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

[26] Neoantigen architectures define immunogenicity and drive immune evasion of tumors with heterogenous neoantigen expression.

Shows: Antigen concurrency/clonality shapes priming competition and therapy resistance; vaccination can restore control

Does not show: Engineered antigen architectures and treatment; no mandatory human natural-history phases

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC4.

[27] Geospatial immune variability illuminates differential evolution of lung adenocarcinoma.

Shows: Immune-cold spatial regions associate with evolution and relapse

Does not show: Geometry/immune associations; shared TRACERx and no immune intervention

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4, SC6.

[28] Lack of detectable neoantigen depletion signals in the untreated cancer genome.

Shows: Apparent depletion is reproduced by context-matched null mutations and diminishes after correction

Does not show: Weak detectable depletion does not exclude editing via presentation loss, low power or early escape

Access/status: Complete primary abstract and linked-status metadata examined. Selected body sections: 0:Results:paragraph4, 0:Results:paragraph13, 0:Results:paragraph14, 0:Results:paragraph17. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3, SC6.

[29] Elevated neoantigen levels in tumors with somatic mutations in the HLA-A, HLA-B, HLA-C and B2M genes.

Shows: Presentation-defective tumors carry more mutation/neoantigen burden

Does not show: Association and timing inference; no restored immune recognition or universal sequence

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3, SC4.

[30] Adaptive immunity maintains occult cancer in an equilibrium state.

Shows: Abstract reports prolonged immune-controlled occult transformed cells and edited spontaneous escape

Does not show: Full body unavailable in this examination; no animal-level depletion/spontaneous-frequency audit; phases can act independently

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC6.

[31] Cancer exome analysis reveals a T-cell-dependent mechanism of cancer immunoediting.

Shows: Mutant spectrin-beta2 recognition, antigen restoration and mixed-clone tracking establish T-cell-dependent selection

Does not show: Transplanted hypermutated mouse clones; not every untreated human tumor

Access/status: Complete primary abstract and linked-status metadata examined. Selected body sections: 2:Main text:paragraph2, 2:Main text:paragraph6, 2:Main text:paragraph7, 2:Main text:paragraph8. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC3.

[32] Type I interferon is selectively required by dendritic cells for immune rejection of tumors.

Shows: DC type-I IFN response is required for cross-priming and tumor rejection; NK/myeloid contrasts exclude broad effector attribution

Does not show: Highly immunogenic challenge models; no universal initiation phases

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

[33] Expression of tumour-specific antigens underlies cancer immunoediting.

Shows: Immune/tolerance contrasts and antigen or presentation loss establish selection; Aza restores reporter antigen expression

Does not show: Engineered antigens and RNA/reporter restoration; no universal endogenous neoantigen rescue across human tumors

Access/status: Complete primary abstract and linked-status metadata examined. Selected body sections: 3:Main text:paragraph0, 3:Main text:paragraph1, 3:Main text:paragraph5, 3:Main text:paragraph7. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC3, SC4, SC5.

[34] Cancer immunoediting of the NK group 2D ligand H60a.

Shows: NK/NKG2D selects against H60a-high tumor cells even without adaptive lymphocytes

Does not show: Innate ligand selection and transplantation; not obligate adaptive three phases

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4, SC6.

[35] Perforin-mediated suppression of B-cell lymphoma.

Shows: Perforin protects in v-Abl/bcl2 but not aggressive Myc lymphoma, defining context dependence

Does not show: Perforin spans innate/adaptive effectors; no general immune-null claim

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC6.

[36] Quantification of Neoantigen-Mediated Immunoediting in Cancer Evolution.

Shows: Reported neoantigen-negative-selection metrics distinguish elimination and escape scores

Does not show: Author labels are not observed phases; null-model dispute and gene-expression/prefiltering biases remain

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3, SC6.

[37] Epigenetic silencing by SETDB1 suppresses tumour intrinsic immunogenicity.

Shows: SETDB1 loss derepresses TE antigens and treatment sensitivity requires MHC-I/CD8 cells

Does not show: Untreated loss has little effect in tested models; no broad untreated necessary escape mechanism

Access/status: Complete primary abstract and linked-status metadata examined. Selected body sections: 1:Results:paragraph3, 1:Results:paragraph18, 1:Results:paragraph22. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4, SC5.

[38] KDM5B promotes immune evasion by recruiting SETDB1 to silence retroelements.

Shows: KDM5B recruits SETDB1 independently of demethylation; retroelements activate innate sensing and rejection

Does not show: Specific melanoma machinery, pleiotropy and multiple antigen/sensing effects

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4, SC5.

[39] Epigenetic priming restores the HLA class-I antigen processing machinery expression in Merkel cell carcinoma.

Shows: HDAC inhibition and peptide provision restore HLA surface expression

Does not show: APM restoration not coding neoantigen rescue; immunodeficient xenograft does not show functional host T-cell causality

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4, SC5.

[40] Evasion of Innate Immunity Contributes to Small Cell Lung Cancer Progression and Metastasis.

Shows: Restoring NKG2D ligands suppresses growth/metastasis via NK cells

Does not show: Lineage-linked innate avoidance; high TMB does not guarantee adaptive engagement

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4, SC6.

[41] An Evolutionarily Conserved Function of Polycomb Silences the MHC Class I Antigen Presentation Pathway and Enables Immune Evasion in Cancer.

Shows: Conserved developmental PRC2 program silences antigen-presentation pathway and supports immune escape

Does not show: Physiological lineage inheritance can precede selection; presentation silencing not necessarily antigen mutation loss

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4, SC5, SC6.

[42] Epigenetic changes in tumor Fas levels determine immune escape and response to therapy.

Shows: Fas rescue and host NK dependence restore tumor control/chemosensitivity

Does not show: Fas susceptibility rather than neoantigen loss; scope-specific broad HDAC effects

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4, SC5.

[43] Histone methylation antagonism drives tumor immune evasion in squamous cell carcinomas.

Shows: EZH2 inhibition restores immune infiltration and impairs mutant tumor growth despite retroelement expression

Does not show: Viral mimicry is not always immunity; chromatin cross-talk context-dependent

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4, SC5.

[44] Whole-genome doubling drives immune evasion by silencing antigen presentation.

Shows: WGD alters metabolism/chromatin and MHC regulation; PRC2 inhibition restores presentation and CD8 infiltration

Does not show: Engineered WGD and many changes; no inference of obligatory prior equilibrium

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4, SC5, SC6.

[45] Tumor antigens are constitutively presented in the draining lymph nodes.

Shows: Antigen is continuously presented in draining nodes despite absence of strong CTL control

Does not show: Proliferative priming is not elimination; engineered TCR/antigen and inoculation model

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC6.

[46] Spatial Architecture of Myeloid and T Cells Orchestrates Immune Evasion and Clinical Outcome in Lung Cancer.

Shows: Neutrophil, fibroblast and vascular spatial barriers associate with metastasis and immune exclusion

Does not show: Shared cohort; mechanistic hypotheses rather than intervention or traced first phase

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4, SC6.

[47] Highly malignant tumor variants retain tumor-specific antigens recognized by T helper cells.

Shows: Malignant variants retain helper-T-cell antigens while losing CTL targets; immunization discriminates protection

Does not show: Specific antigen classes; retained antigens do not refute all immune selection

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3, SC6.

[48] Immunological ignorance is an enabling feature of the oligo-clonal T cell response to melanoma neoantigens.

Shows: Expressed shared neoantigens can lack reactive clones; vaccination broadens responses

Does not show: Small metastatic series and vaccination; ignorance is contextual not universal absence of immunity

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3, SC6.

[49] Mutations Associated with Acquired Resistance to PD-1 Blockade in Melanoma.

Shows: JAK1/JAK2/B2M alterations accompany acquired resistance and functional signaling/presentation defects

Does not show: Exogenous checkpoint pressure; acquired resistance cannot prove untreated tumor initiation

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

[50] Genetic evolution of T-cell resistance in the course of melanoma progression.

Shows: Early B2M allele loss plus later second hit tracks progressive loss of recognition

Does not show: Few lesions and established metastatic course; no mandatory equilibrium

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

[51] Resistance to checkpoint blockade therapy through inactivation of antigen presentation.

Shows: B2M defects enrich in progression/nonresponders and predict poor outcome

Does not show: Treatment and observational enrichment; presenter loss rather than antigen mutation depletion

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

[52] Lack of immunoediting in murine pancreatic cancer reversed with neoantigen.

Shows: Juvenile sustained validated T-cell depletion leaves spontaneous tumor timing/survival intact; OVA addition restores CD8 rejection and selection

Does not show: Scoped murine counterexample; predicted neoepitope absence not all possible antigens; OVA engineered positive control

Access/status: Complete primary abstract and linked-status metadata examined. Selected body sections: 0:Publisher HTML paragraphs:paragraph57, 0:Publisher HTML paragraphs:paragraph62, 0:Publisher HTML paragraphs:paragraph67, 0:Publisher HTML paragraphs:paragraph73, 0:Publisher HTML paragraphs:paragraph86, 0:Publisher HTML paragraphs:paragraph88. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC3, SC6.

[53] Immunoediting restricts clonal neoantigens in primary, treatment-naive human tumors.

Shows: High-infiltrate intact-host tumors restrict strong-binding/immunogenic neoantigens to lower clonal frequency with purity/growth controls

Does not show: Cross-sectional immunosuppression and age differences; not every cancer, not functional destruction of each predicted epitope

Access/status: Complete primary abstract and linked-status metadata examined. Selected body sections: 1:RESULTS::paragraph3, 1:RESULTS::paragraph7, 1:RESULTS::paragraph10, 1:RESULTS::paragraph13, 1:RESULTS::paragraph20. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3, SC6.

[54] Perforin-mediated cytotoxicity is critical for surveillance of spontaneous lymphoma.

Shows: Perforin loss predisposes lymphoma and CD8 rejection differs strongly

Does not show: Lymphoma-specific; p53 sarcoma incidence/onset unaffected; innate/adaptive substrate not separated globally

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC6.

[55] Host perforin reduces tumor number but does not increase survival in oncogene-driven mammary adenocarcinoma.

Shows: Perforin/NK reduces tumor number and delays onset without survival gain; CD8 depletion does not reproduce effect

Does not show: Positive narrow innate surveillance cannot upgrade universal adaptive elimination

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC6.

[56] Suppression of lymphoma and epithelial malignancies effected by interferon gamma.

Shows: IFNgamma/perforin restrain lymphoma and some late epithelial cancers

Does not show: Strain-specific protection; IL12/18/TNF single loss not sufficient; broad pathways

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC6.

[57] A critical role for natural killer T cells in immunosurveillance of methylcholanthrene-induced sarcomas.

Shows: Physiological NKT transfer protection requires CD1d, downstream NK/CD8, IFNgamma and perforin

Does not show: Specific carcinogen challenge and pathway; NKT is not proof of general conventional-T sequence

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

[58] Innate immune surveillance of spontaneous B cell lymphomas by natural killer cells and gammadelta T cells.

Shows: NK and gamma-delta effectors protect/reject tumors without conventional MHC-I context

Does not show: Complex beta2m affects MHC/CD1d/CD16; no isolated adaptive obligatory phase

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC6.

[59] Critical role for tumor necrosis factor-related apoptosis-inducing ligand in immune surveillance against tumor development.

Shows: TRAIL blockade increases susceptible tumors and interacts with NK/IFNgamma

Does not show: Specific effectors and susceptible lineages; no cross-etiology necessity

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

[60] Cancer immunoediting by the innate immune system in the absence of adaptive immunity.

Shows: NK IFNgamma and M1 macrophages edit tumors without adaptive immunity; CD40 restores editing

Does not show: Innate editing refutes an exclusively adaptive definition but remains carcinogen-model scoped

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC4, SC6.

[61] Immune evasion before tumour invasion in early lung squamous carcinogenesis.

Shows: Immune activation and escape markers arise before invasion

Does not show: Morphological trajectories and biomarkers rather than experimentally identified universal equilibrium

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4, SC6.

[62] Immune Escape Is an Early Event in Pre-Invasive Lesions of Lung Squamous Cell Carcinoma.

Shows: PD-L1 is present in a subset of dysplasias and absent in controls

Does not show: PD-L1 staining not functional immune escape or phase sequence

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4, SC6.

[63] IL-1α Expands SLC2A1highSPP1highIFNlow Myeloid Cells to Drive Immune Escape and Malignant Transformation of Precancers.

Shows: Early pharmacologic/genetic IL1 interruption reduces suppressive myeloid subsets, progression and mortality

Does not show: Defined myeloid pre-emption model; no lineage proof that no brief equilibrium ever occurred

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4, SC6.

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
• SHA256.txt
• adversarial-findings.json
• cancer-immunoediting.review.md
• cancer-immunoediting.scorecard.json
• codex-correction-provenance.json
• consensus-addendum.md
• consensus-validation.json
• edition-approval-audit.json
• reader-summary.json
• validation-report.json
• validation-round2-2026-10-14.json
• validation.txt

Sources: 63 · load-bearing claims: 70 · verified: 70 · unverified: 0 · contradicted: 0 · schema 5.3