In plain terms

The claim being assessed: Adding Galleri multi-cancer blood screening to usual care reduces cancer mortality in asymptomatic adults aged 50 and older.

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

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

Strongest evidence for it: The preregistered primary-subclaim rule returns Γ from SC2. Fifty-eight original reports were examined, covering Galleri/other MCED diagnostic work, prospective implementation, real mortality-based single-cancer screening comparators, and separate MRI/DTC pathways. No positive Galleri mortality crossing is counted. The current primary stage endpoint failed and a favorable secondary stage IV signal remains; neither proves a mortality benefit or a mortality null. Weakest-link returns Α from the broad companion claim, while primary-subclaim and scope-matched return Γ. The review preserves genuine detections and legitimate established screening benefits without assigning them to a different assay.

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

Galleri detects real cancers, but a cancer-mortality benefit remains unestablished; the 2026 primary stage endpoint was null and a lower stage IV rate in a secondary endpoint that could not be formally tested requires clinical follow-up.

Implications for health and treatment: The preregistered primary-subclaim rule returns Γ from SC2. Fifty-eight original reports were examined, covering Galleri/other MCED diagnostic work, prospective implementation, real mortality-based single-cancer screening comparators, and separate MRI/DTC pathways. No positive Galleri mortality crossing is counted. The current primary stage endpoint failed and a favorable secondary stage IV signal remains; neither proves a mortality benefit or a mortality null. Weakest-link returns Α from the broad companion claim, while primary-subclaim and scope-matched return Γ. The review preserves genuine detections and legitimate established screening benefits without assigning them to a different assay. The framework's 'epiphenomenon' label means a causal role in reducing cancer deaths is not established by the evidence, not that screening has no effect; no effect on cancer deaths has yet been shown in either direction.

Limits for individual care: 58 primary reports with 3 selected partial-body examinations, plus 2 records added in the 2026-11-02 adversarial correction. Confidence reflects evidence density and reproducibility, not access depth or correctness: the ceiling is well characterised because the randomized Galleri trial, a large prospective study and a real-world series all leave no Galleri mortality contrast. The grade applies to the exact registered claim; stronger findings for narrower subclaims remain separate. No complete-paper audit or blinded calibration is claimed. This educational review does not establish an individual treatment recommendation.

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

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

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

At a glance

Sub-claims assessed separately

Full assessment

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

Sources: 58 primary reports examined at the 2026-10-13 cutoff, plus 2 records added in the 2026-11-02 adversarial correction (PATHFINDER 2 and a Tier B meta-analysis, both at abstract/metadata depth); ledger: 60 sources; claims: 65; verified: 65; unverified: 0. Verification is bounded attribution. Access: 3 partial-body and 55 bounded abstract/status examinations. Literature cutoff: 2026-10-13; source examination UTC date: 2026-10-13. No complete-paper audit or blinded calibration claimed.

Adversarial review: Authorship of record: Codex (OpenAI), as recorded in the score card and provenance files; the earlier "Sol 6.1" header label is a house label and no specific model version is inferred. Adversarially reviewed by Claude Opus 5.5 (Anthropic) on 2026-10-14 (4 corrections) and again on 2026-11-02 (findings S1–S11; corrections applied 2026-11-02 at the principal investigator's authorisation; see ADVERSARIAL REVIEW AND APPROVAL). Status: Red-teamed; corrections applied.

AT A GLANCE

Galleri detects real cancers, but a cancer-mortality benefit remains unestablished; the 2026 primary stage endpoint was null and a lower stage IV rate in a secondary endpoint that could not be formally tested requires clinical follow-up.

Deciding experiment: Continue or run a randomized intention-to-screen comparison of specified Galleri plus usual care versus usual care in asymptomatic adults ≥50, preserving established screening and treatment access. Prespecify cancer-specific mortality, all-cause mortality, site-specific deaths, diagnostic procedures, treatment morbidity, interval cancers and sustained incidence differences over an adequate horizon. Use blinded death adjudication and systematic outcome linkage in both arms. Stage shifts are candidate mediators, with predefined surrogate validation rather than conversion into modeled saved lives.

Belief versus evidence: The preregistered primary-subclaim rule returns Γ from SC2. Fifty-eight original reports were examined, covering Galleri/other MCED diagnostic work, prospective implementation, real mortality-based single-cancer screening comparators, and separate MRI/DTC pathways. No positive Galleri mortality crossing is counted. The current primary stage endpoint failed and a favorable secondary stage IV signal remains; neither proves a mortality benefit or a mortality null. Weakest-link returns Α from the broad companion claim, while primary-subclaim and scope-matched return Γ. The review preserves genuine detections and legitimate established screening benefits without assigning them to a different assay.

Best-supported rival: Screening benefit is site, assay, risk and pathway specific. UKCTOCS's original randomized long-term report (PMID33991479) separates a context-specific clinical screening model from an assumption that favorable stage movement universally establishes saved lives: stage IV fell while ovarian/tubal mortality did not. Positive NLST/NELSON mortality findings preserve the possibility of genuine screening benefit. This comparator does not decide Galleri's eventual mortality result.

COMPETING MECHANISTIC MODELS

Rank 1: Screening benefit is site, assay, risk and pathway specific (Ε). Earlier detection can reduce mortality in defined screening programmes, while other apparently favorable diagnostic or stage programmes do not. NLST and NELSON show positive randomized lung-mortality effects in high-risk smokers (PMID21714641, 31995683); UKCTOCS shows a long-term ovarian stage IV/mortality dissociation, with a non-significant combined stage III/IV change (PMID33991479); across 41 screening RCTs, reductions in late-stage incidence tracked mortality reductions closely for ovarian and lung cancer but weakly for colorectal and prostate cancer (PMID38583868). Galleri sensitivity varies by burden and site (PMID34176681, 42773209). This establishes contextual screening causality, not a positive Galleri mortality result. It does not identify a universal stage surrogate or a superior test for every cancer.

Rank 2: Diagnostic cascades and overdiagnosis alter net benefit (Γ). Broad screening produces real detections and additional diagnostic work, with a net balance that can differ from analytical accuracy. PATHFINDER false-positive resolution/workup (PMID37805216), UKCTOCS unnecessary surgery (PMID37183782), SHIP biopsy/cost findings (PMID32860149, 34996801) and DTC confirmation errors (PMID29565420) document relevant burdens. Overdiagnosis is not directly quantified by every incidental finding, and higher costs or biopsies do not alone prove net harm. Observational MRI attribution and lack of exact Galleri mortality follow-up limit clinical balance claims.

Rank 3: Lead time, length selection and incomplete ascertainment inflate survival narratives (Γ). Longer survival from diagnosis or favorable detected-cancer series can arise without moving the actual time of death. UKCTOCS separates a stage IV reduction, with a non-significant combined stage III/IV change, from deaths (PMID33991479); real-world Galleri has 45% positive follow-up and lacks systematic negative follow-up (PMID41173830). Prediagnostic PanSeer/ARIC results show a potential observation interval, not intervention-caused saved lives (PMID32694610, 40402478). These biases are plausible source-specific limits rather than proof that every earlier diagnosis is useless or that aggressive cf DNA-detected cancers are all indolent. Randomized site-specific mortality benefits remain real.

Rank 4: Galleri addition reduces cancer mortality (Γ). Adding specified Galleri screening to usual care causes fewer cancer deaths in baseline asymptomatic adults ≥50. True prospective cancers, tissue-origin guidance and a favorable stage IV secondary contrast support clinical plausibility (PMID37805216, 42773209, 42776074). No positive randomized death endpoint is reported in the examined sources; the combined stage III/IV primary endpoint was null. Separate screening programmes or survival simulations cannot supply exact-assay mortality proof.

Rank 5: Shedding-biased detection (Γ). CfDNA tests preferentially detect high-shedding, biologically advanced cancers, so detections concentrate where an earlier diagnosis changes outcomes least. CCGA validation sensitivity rose from 16.8% in stage I to 90.1% in stage IV (PMID34176681); NHS all-cancer episode sensitivity was 26.7–37.2% (PMID42773209); tumor-fraction correlates link detection to tumor burden and aggressiveness (PMID34432811); and in NHS-Galleri the stage IV decrease was offset by more stage III cancers (PMID42776074). Preferential detection of advanced disease does not show that every detected cancer would be untreatable or that earlier detection never helps; the mortality consequence is untested. Added after the 2026-11-02 adversarial review (finding S10) and appended without re-ranking the original tournament.

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

Adding Galleri multi-cancer blood screening to usual care reduces cancer mortality in asymptomatic adults aged 50 and older.

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

Selection: position 20 in the registered editorial traffic-priority list; registration timestamp 2026-10-13 T 16: 06: 25.540258+00: 00. Ranked editorial traffic priority rather than measured global popularity or a preset verdict. Registered claim assessed against original sources; no independent blinded calibration is claimed.

Evidence and causal interpretation

Rubric fit: The frozen framework has no screening-mortality rubric. Cancer-rubric biological perturbation wording is quoted unchanged, with an explicit clinical equivalent: randomized addition of the specified screening pathway to usual care with a positive cancer-death effect over adequate follow-up. Diagnostic validation and stage shift are distinct from mortality; tissue-origin accuracy, sensitivity, specificity and predictive values are descriptive properties, not HSEF causal probabilities. No animal lineage experiment is required to evaluate a human screening programme. A descriptive diagnostic appendix accompanies SC1 so the real phenomenon is not graded as refuted.

The registered question is whether adding Galleri to usual care saves lives in asymptomatic adults aged 50 and older. The review preserves separate questions about cancer signals, diagnostic accuracy, stage, treatment and harms. Positive predictive value depends on the screened population; high specificity and tissue-origin accuracy do not mean that every positive is cancer, every negative is cancer-free or the programme prevents deaths. Other MCED technologies remain separate exposures.

The latest original randomized NHS-Galleri report was published online 22 September 2026. 142250 adults aged 50–77 were assigned to tested or stored blood alongside usual care. After three rounds, the primary stage III/IV rate ratio was 1.03 (95% CI 0.92–1.14; P=.63). The key secondary stage IV ratio, also among 12 prespecified cancer types, was 0.86 (0.74–1.00); because the fixed-sequence plan stopped at the failed primary endpoint, this result was not formally tested and its interval is unadjusted. The stage IV decrease (341 vs 394 participants) was offset by more stage III cancers (360 vs 287), and a prespecified secondary composite of stage III/IV or fatal cancers across routinely staged types was null (IRR 1.04, 0.97–1.12). These testing-status and count details come from the PMC author manuscript (PMC7619610), examined in the 2026-11-02 adversarial pass. Both results remain in the review, with their endpoint hierarchy; neither is a cancer-mortality analysis (PMID42776074). The official trial information states that analyses of cancer deaths continue. Publisher abstract access was verified; full-body access was not obtained.

The companion NHS performance paper reports 937 detected primary cancers across rounds. Specificity was 99.5–99.6%, but all-cancer episode sensitivity was 26.7–37.2%, and positive predictive value declined from 58.0% to 45.8% across rounds (PMID42773209). These descriptive intervention-arm estimates are not a second randomized efficacy trial. CCGA's known-cancer validation found 51.5% overall sensitivity and 16.8% in stage I, with different spectrum and prevalence (PMID34176681). Detection often increases with burden; this does not by itself establish a successful early-treatment window.

PATHFINDER found 35 confirmed cancers among 92 signal-positive participants, with 57 false positives and a median 162-day diagnostic resolution in the false-positive group (PMID37805216). That 62% false-positive share among positive signals is different from the population false-positive rate. DETECT-A used a different blood-test/PET-CT pathway in 10006 women and reported 26 blood-detected cancers, 24 routine-screening detections and 46 detected by neither (PMID32345712). Curative-intent surgery is meaningful but is not proof of cure or reduced mortality in an uncontrolled study.

PATHFINDER 2 enrolled 35,878 adults aged 50 or older without clinical suspicion of cancer; among 32,007 performance-analyzable participants, detection rate was 0.54%, PPV 60.3%, episode sensitivity 39.3% (69.8% for 12 prespecified cancers), specificity 99.64%, and 213 (0.6%) had invasive procedures after a positive test. Endpoints were descriptive and the authors state that randomized trials and longer follow-up are needed to assess clinical utility (PMID42773207).

The 111080-test real-world Galleri series needs denominator discipline. Original Methods/Results show voluntary clinical outcomes for 459 of 1011 positive tests, of which 411 had completed workup. The asymptomatic complete-workup subset had 128 cancers among 259, an empirical positive predictive value of 49.4%. Negative outcomes were not collected systematically, and some tests were used in symptomatic or already diagnosed patients (PMID41173830). This large dataset therefore cannot establish whole-population sensitivity, all-comer screening predictive values or a mortality effect.

Established screening comparisons demonstrate both possibility and limits. NLST and NELSON show randomized lung-mortality benefit in high-risk smokers (PMID21714641, 31995683). Conversely, actual UKCTOCS Methods/Results show 47.2% more stage I and 24.5% fewer stage IV ovarian/tubal diagnoses with multimodal screening, while the combined stage III/IV reduction (10.2%, 95% CI −21.3 to 2.4) was not significant and mortality did not fall after 16.3 years (PMID33991479). A 2024 meta-analysis of 41 screening RCTs found that reductions in stage III–IV incidence tracked mortality reductions closely for ovarian and lung cancer but weakly for colorectal and prostate cancer (PMID38583868), so stage-surrogate validity is site specific rather than universally absent. CAP's 15-year small prostate-death effect and Nord ICC's incidence benefit with nonsignificant mortality illustrate the role of disease, programme, uptake and horizon (PMID38581198, 36214590). These are actual outcome studies, not transferable proof about Galleri.

Full-body MRI detects lesions and occasional cancers, with different evidence in hereditary TP53 surveillance and average-risk health checks (PMID27501770, 40588398, 27635171). SHIP's original body shows optional MRI plus concurrent laboratory disclosure, increased biopsies and incomplete regional pathology coverage; most biopsies were nonmalignant (PMID32860149). Costs and short-term distress also occurred, while a separate long-term aggregate quality-of-life analysis was null (PMID34996801, 23239059, 26597793). A selected DTC confirmation series found 40% false variants among referred raw-data findings; that is not 40% of all consumer tests (PMID29565420). No blanket equivalence in net clinical screening benefit follows from these distinct pathways.

HYPOTHESIS STATE EVALUATION FRAMEWORK (HSEF v 5.3)

Subclaim matrix and aggregation

Preregistered aggregation: primary-subclaim. The preregistered primary-subclaim rule returns Γ from SC2. Fifty-eight original reports were examined, covering Galleri/other MCED diagnostic work, prospective implementation, real mortality-based single-cancer screening comparators, and separate MRI/DTC pathways. No positive Galleri mortality crossing is counted. The current primary stage endpoint failed and a favorable secondary stage IV signal remains; neither proves a mortality benefit or a mortality null. Weakest-link returns Α from the broad companion claim, while primary-subclaim and scope-matched return Γ. The review preserves genuine detections and legitimate established screening benefits without assigning them to a different assay. Sensitivities: weakest-link Α; primary-subclaim Γ; scope-matched Γ.

Decisive-perturbation inventory and causal tally

[1] NHS-Galleri randomized 142250 adults 50–77 to tested versus stored blood plus usual care for 3 annual rounds; primary stage III/IV incidence IRR 1.03, 95% CI 0.92–1.14, P=.63; key secondary stage IV IRR 0.86, 95% CI 0.74–1.00, not formally tested under the fixed-sequence plan and with an unadjusted interval; the stage IV decrease (341 vs 394) was offset by more stage III cancers (360 vs 287), and the secondary stage III/IV-or-fatal composite was null (IRR 1.04, 0.97–1.12; PMC7619610). Primary proxy failed, secondary signal retained; no mortality result in this publication. Neither randomized design nor a stage IV contrast supplies a positive cancer-mortality crossing. Publisher abstract verified 22 September 2026; full-body access not claimed.

[2] Same NHS trial descriptive test-performance report across 3 rounds: 937 detected primary cancers, PPV 58.0/50.4/45.8%, all-cancer episode sensitivity 26.7–37.2%, specificity 99.5–99.6%. Descriptive intervention-arm diagnosis metrics without hypothesis testing; not an independent randomized trial or mortality analysis. High NPV largely reflects low prevalence and does not mean every cancer is excluded.

[10] PATHFINDER convenience cohort ≥50 without cancer symptoms: 6621 analyzable; 92 signals, 35 confirmed, 57 false positives; false-positive diagnostic resolution median 162 days and 30% procedures. Single-arm feasibility primary endpoint is diagnostic workup, not mortality. A low overall false-positive rate and 62% false-positive share among signals are different denominators. Attempted original XML returned 500; abstract examined.

[18] Original M/R: 111080 real-world tests ≥22, 1011 signals; voluntary outcomes 459, 48 ongoing workups, 411 complete. Asymptomatic completed subset 128/259 invasive cancers, ePPV 49.4%; negatives not systematically followed. Symptomatic/active-cancer uses and incomplete positive follow-up preclude full-population sensitivity, universal PPV or mortality claims. Large test count is not complete clinical ascertainment or screening-only cohort.

[45] Original M/R SHIP 6753 participants with optional MRI 3371; linked biopsy rate 6.9 pre to 9.9 post per 100 observation-years; large absolute increase in nonmalignant results, higher with disclosed MRI/lab findings. Concurrent lab disclosures, selection and regional pathology coverage constrain MRI-specific causation; increased biopsies do not quantify overdiagnosis or prove all MRI net harm. No randomized mortality contrast.

[48] NLST 53454 high-risk smokers randomized 3 annual LDCT versus chest radiography; lung-cancer mortality 20% relative and all-cause 6.7% lower at initial follow-up, with many positive-screen false alarms. Positive disease-specific screening-mortality proof in defined risk/context, not Galleri; higher incidence and procedure harms must remain. Different modality cannot supply exact lead threshold.

[49] NELSON randomized high-risk smokers 50–74 to volumetric CT/no screening; 10 year male lung death RR 0.76, 95% CI 0.61–0.94, women RR 0.67, 95% CI 0.38–1.14. True positive clinical screen effect in defined population; sex subgroup imprecise, different test/site/care pathway. No MCED mortality crossing.

[51] Original M/R UKCTOCS 202562 average-risk women 50–74, median 16.3 years; MMS increased stage I 47.2% and reduced stage IV 24.5%, but ovarian/tubal deaths not lower P=.58; USS P=.36. Direct stage/mortality dissociation blocks universal surrogate claim; does not prove Galleri mortality null because assay/site differ. Positive clinical mortality crossing absent for ovarian programme. The combined stage III/IV reduction (10.2%, 95% CI −21.3 to 2.4) was not significant, so the dissociation is stage IV-only.

Tally: 0 positive qualifying study units across scoped subclaims; 4 screened discriminating contrast units, the randomized allocated-arm contrasts [1], [48], [49] and [51]. [45] is an observational pre/post contrast and [2], [10] and [18] are single-arm descriptive series. These are not independent causal lines or proof of full-claim necessity/sufficiency. Null and negative contrasts are discriminators, not positive threshold crossings. Reports and replications are not independent causal lines. Zero positive exact Galleri mortality study units are counted; diagnostic performance and stage signals do not become death outcomes. Single-cancer mortality trials remain comparator context, not nominal-assay threshold credit. Original validation components count as reports; simulations, reviews, protocols, notices, recruitment/status pages and provider counts do not pad observed clinical efficacy. The 58 report count is not 58 randomized MCED trials or 58 complete-paper audits.

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

The exact-assay mortality claim has no positive crossing, so stage, blood-signal or treatment-intent data remain partial support under the disclosed clinical adaptation. Diagnostic accuracy is not itself causally graded. The primary stage null and secondary stage IV signal are different endpoint results and neither is recast as a cancer-death analysis. Clinical mortality trials of other modalities score within their own populations and do not transfer a positive result to Galleri. Independently complementary same-scope causal evidence would be assessed as a set; it is not present for this lead.

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: Diagnostic accuracy is descriptive and genuine cancer detections remain credited; causal screening benefit requires a different endpoint. Stage shift and modeled survival do not substitute for randomized cancer deaths. A null primary surrogate is not a positive threshold crossing, and a favorable secondary surrogate is not a mortality result. Different assays, selected case-control cohorts and enriched hereditary or prenatal populations retain actual scope.

SC1 cap: “Biomarker change without functional/phenotypic change caps at Γ.” Prospective Galleri studies and the NHS intervention-arm performance report demonstrate genuine cancer detections with quantified false positives, missed cancers and tissue-origin guidance. Diagnostic accuracy is reported descriptively, not as a causal grade on whether detection exists. The helper-compatible Γ entry records the unbridged inference from that diagnostic signal to meaningful clinical benefit, not doubt that the test detects some cancers. Case-control sensitivity differs from screening performance and cannot be substituted into population benefit estimates.

SC3 cap: “Biomarker change without functional/phenotypic change caps at Γ.” A stage shift is insufficient by itself to establish saved lives. UKCTOCS directly demonstrates more stage I and fewer stage IV ovarian/tubal diagnoses without a cancer-mortality reduction after 16.3 years. Across 41 screening RCTs, late-stage incidence reductions tracked mortality closely for ovarian and lung cancer but weakly for colorectal and prostate cancer (PMID38583868); UKCTOCS's combined stage III/IV change was not significant, so its dissociation is a stage IV-only one, the same pattern as NHS-Galleri. NHS-Galleri's stage IV secondary signal is retained, but the combined stage III/IV primary endpoint was not met. Site-specific biology, lead time, overdiagnosis, treatment and competing mortality require direct validation; stage-to-mortality causal inference retains Γ.

SC4 cap: “For highly promoted but weakly causal narratives sustained by marker association or conceptual appeal.” General-population full-body MRI and DTC testing lack a verified comparable net mortality benefit in the examined sources. MRI detects real lesions and some cancers, while incidental findings, investigations, costs and distress also occur. Hereditary TP53 surveillance and abnormal prenatal cf DNA workup are enriched contexts and different packages. A selected DTC confirmation series identifies false variants without implying 40% of all consumer tests are wrong. Blanket clinical-benefit equivalence remains Α.

SC1 field-valid definition: “ - Endogenous genetic manipulation in autochthonous models - Patient-derived system perturbation with matched rescue AND in vivo confirmation - Clonal tracking demonstrating persistence of dependency through selection pressure” Positive full-scope result: no. Prospective Galleri studies and the NHS intervention-arm performance report demonstrate genuine cancer detections with quantified false positives, missed cancers and tissue-origin guidance. Diagnostic accuracy is reported descriptively, not as a causal grade on whether detection exists. The helper-compatible Γ entry records the unbridged inference from that diagnostic signal to meaningful clinical benefit, not doubt that the test detects some cancers. Case-control sensitivity differs from screening performance and cannot be substituted into population benefit estimates.

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. No examined primary report provides a positive randomized Galleri cancer-mortality contrast. The 2026 NHS trial randomized 142250 adults aged 50–77; its primary stage III/IV endpoint was null, while the stage IV secondary contrast was favorable. These are stage endpoints, not deaths. Official trial information states that mortality follow-up continues; the trial was not powered for mortality endpoints. Detection, feasibility, reported treatment and favorable stage subsets cannot cross the registered mortality threshold. Γ reflects plausible partial clinical system support with an unverified mortality bridge, not proof that a mortality trial has refuted Galleri.

SC3 field-valid definition: “ - Endogenous genetic manipulation in autochthonous models - Patient-derived system perturbation with matched rescue AND in vivo confirmation - Clonal tracking demonstrating persistence of dependency through selection pressure” Positive full-scope result: no. A stage shift is insufficient by itself to establish saved lives. UKCTOCS directly demonstrates more stage I and fewer stage IV ovarian/tubal diagnoses without a cancer-mortality reduction after 16.3 years. Across 41 screening RCTs, late-stage incidence reductions tracked mortality closely for ovarian and lung cancer but weakly for colorectal and prostate cancer (PMID38583868); UKCTOCS's combined stage III/IV change was not significant, so its dissociation is a stage IV-only one, the same pattern as NHS-Galleri. NHS-Galleri's stage IV secondary signal is retained, but the combined stage III/IV primary endpoint was not met. Site-specific biology, lead time, overdiagnosis, treatment and competing mortality require direct validation; stage-to-mortality causal inference retains Γ.

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. General-population full-body MRI and DTC testing lack a verified comparable net mortality benefit in the examined sources. MRI detects real lesions and some cancers, while incidental findings, investigations, costs and distress also occur. Hereditary TP53 surveillance and abnormal prenatal cf DNA workup are enriched contexts and different packages. A selected DTC confirmation series identifies false variants without implying 40% of all consumer tests are wrong. Blanket clinical-benefit equivalence remains Α.

SC1 applied rubric adaptation (distinct from the unchanged literal criterion): Diagnostic observations are described with actual denominators and ascertainment. The accompanying Γ is only the unestablished causal translation to clinical benefit; it is not a rating that detection or analytical validity is mechanistically uncertain.

SC2 applied rubric adaptation (distinct from the unchanged literal criterion): The field-valid screening equivalent is a randomized offer/addition of the exact Galleri pathway versus usual care in baseline asymptomatic adults ≥50, with retained standard screening and adequately powered cancer-mortality follow-up. Stage and detection endpoints do not supply that death endpoint.

SC3 applied rubric adaptation (distinct from the unchanged literal criterion): A stage surrogate must predict intervention-caused mortality change across prespecified cancer sites, assays and treatment contexts. Prognostic stage-survival associations and longer survival measured from earlier diagnosis are not surrogate validation.

SC4 applied rubric adaptation (distinct from the unchanged literal criterion): Each commercial MRI or DTC test requires its own defined population, diagnostic pathway, mortality/function endpoints and harm assessment. High-risk syndrome surveillance, risk genotyping and average-risk occult-cancer screening are different propositions.

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 as a cancer-stemness rule. The separate universal signal-to-clinical-benefit cap applies only to the causal translation, not diagnostic accuracy itself.

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” Not triggered: SC1 evidence is human prospective screening and randomized-arm diagnostic data; no in vitro-only class is offered.

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: directly studied human screening is evaluated through the disclosed clinical equivalent; the absent positive mortality outcome is the blocker.

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 as a cancer-stemness rule; stage-to-mortality surrogate inference remains explicitly capped separately.

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. Keep diagnostic performance descriptive; verify interval-cancer ascertainment and modern assay performance prospectively, then separately demonstrate meaningful clinical benefit rather than applying a causal letter to accuracy.

SC2 Δ: “Requires at least one meaningful causal perturbation showing the proposed mechanism contributes to initiation, maintenance, or relapse.” Satisfied: no. A prespecified, adequately powered randomized Galleri-plus-usual-care comparison must demonstrate lower cancer mortality with blinded outcome adjudication, sufficient follow-up and intact standard screening.

SC3 Δ: “Requires at least one meaningful causal perturbation showing the proposed mechanism contributes to initiation, maintenance, or relapse.” Satisfied: no. Validate cancer/site-specific intervention-caused stage changes against subsequent randomized mortality, including overdiagnosis, lead time and treatment effects; a universal bridge must survive counterexamples.

SC4 Δ: “Requires at least one meaningful causal perturbation showing the proposed mechanism contributes to initiation, maintenance, or relapse.” Satisfied: no. For each exact MRI or DTC pathway, test a baseline-defined population against usual care with clinical outcomes, downstream procedures and harm ascertainment; enriched hereditary surveillance does not supply general-population equivalence.

Upper ladder: Randomized clinical screening is a valid prophylactic perturbation equivalent, and multiple other-site trials were evaluated. None provides a positive exact Galleri mortality contrast. Dose-response, observational survival and model predictions do not independently exclude lead time, ascertainment or overdiagnosis. No exposure-specific genetic natural experiment supplies the clinical bridge. A genuinely complementary same-scope set would be considered together rather than capped by its weakest observational leg.

Independence: No subclaim or model reaches Ζ. The model-level Ε rests on contextual clinical screening effects with explicit positive/null programme boundaries, not an assertion of independently failing Galleri mortality lines. NHS, CCGA, PATHFINDER, SHIP and comparator follow-ups are grouped; different biomarkers or report dates do not create new allocation trials.

Falsification and pitfalls

A phenomenon graded as though it were a claim. True cancer detection is preserved as descriptive diagnostic evidence. The overall HSEF letter addresses the causal mortality proposition, not whether an assay can detect some cancers.

Confirmation by biomarker. Stage, cf DNA signal and predicted five-year survival are not directly observed mortality benefit.

Association dressed as mechanism. Earlier-stage prognosis and longer postdiagnosis survival cannot by themselves show screening moved the time of death.

Selective generalization. High-risk lung programmes, hereditary TP53 surveillance and selected prenatal referrals do not establish average-risk Galleri or commercial whole-body MRI mortality benefit.

Replication confused with reproduction. NHS, CCGA, PATHFINDER, SHIP, NLST, UKCTOCS and Nord ICC follow-ups are grouped; additional resolution or follow-up is not a new allocation trial.

Negative field-valid perturbation read as a crossed threshold. NHS's null primary stage result does not supply a positive clinical crossing merely because allocation was randomized.

Causal-evidence tallies that do not survive an audit. No mortality threshold credit is assigned to simulations, diagnostic metrics, other-assay trials, corrections or provider counts.

Mechanism conflation. Galleri methylation, Cancer SEEK/PET-CT, other fragmentomic/protein tests, MRI and germline DTC risk are different tests and pathways.

Adversarial steelman: above, then below

The strongest Δ case notes that Galleri now has a large randomized study, many confirmed asymptomatic detections and a favorable stage IV secondary contrast. For cancers lacking usual screening, an earlier treatable interception could plausibly improve outcomes. Positive mortality trials of other screening pathways demonstrate that the general clinical mechanism can work.

The strongest Β case observes that no Galleri mortality result is available and many supporting claims are diagnostic, prognostic or simulated. A mortality-benefit claim based solely on these correlations would remain association-bound. The further case for Α would stress promotion beyond the available clinical endpoint.

The higher case is blocked by the missing positive exact-assay cancer-death endpoint: a secondary stage IV result and curative treatment intent do not satisfy it. The lower case is rejected because the screening pathway has genuine prospective clinical implementation and randomized perturbation evidence, even though its measured endpoints do not verify mortality benefit. Γ captures partial system support without declaring that detection is false or future mortality benefit impossible.

Final score, causal direction and M/S/D

The preregistered primary-subclaim rule returns Γ from SC2. Fifty-eight original reports were examined, covering Galleri/other MCED diagnostic work, prospective implementation, real mortality-based single-cancer screening comparators, and separate MRI/DTC pathways. No positive Galleri mortality crossing is counted. The current primary stage endpoint failed and a favorable secondary stage IV signal remains; neither proves a mortality benefit or a mortality null. Weakest-link returns Α from the broad companion claim, while primary-subclaim and scope-matched return Γ. The review preserves genuine detections and legitimate established screening benefits without assigning them to a different assay.

For the precise mortality-benefit mechanism, no positive clinical mortality dependency is demonstrated in either direction, so Step2 returns epiphenomenon for that unverified causal account. This does not label the assay biologically inert or claim detections are imaginary; screening changes diagnoses and workups. The label applies to the unestablished saved-lives bridge, not the validity of every cancer signal. The label means that a causal role in reducing cancer deaths is not established by the evidence, not that screening has no effect: no effect of this screening on cancer deaths has yet been shown in either direction, so the label records an unproven link, not a proven absence of effect.

Momentum: Mechanistic only. NHS-Galleri lists cancer-specific mortality up to 8 years as a prespecified secondary outcome (NCT05611632), and the primary report states that cancer-specific and nested mortality will be analysed at 3 and 6 years after the final trial visit, but also that the trial was not powered for mortality endpoints; after the failed primary endpoint its fixed-sequence plan precludes formal testing of secondary outcomes. This follow-up is informative but cannot by itself decisively cross the SC2 threshold. Sources: https://clinicaltrials.gov/study/NCT05611632 ; PMC7619610 (PMID42776074).

Stability: Uncertain. Prespecified randomized cancer-mortality follow-up could further constrain the exact clinical boundary; current stage results are not that result, and the trial was not powered for mortality endpoints. Named ongoing programme is NHS-Galleri, not an unspecified future study.

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

PROPOSED DECISIVE EXPERIMENTS

Continue or run a randomized intention-to-screen comparison of specified Galleri plus usual care versus usual care in asymptomatic adults ≥50, preserving established screening and treatment access. Prespecify cancer-specific mortality, all-cause mortality, site-specific deaths, diagnostic procedures, treatment morbidity, interval cancers and sustained incidence differences over an adequate horizon. Use blinded death adjudication and systematic outcome linkage in both arms. Stage shifts are candidate mediators, with predefined surrogate validation rather than conversion into modeled saved lives.

Readout: Randomized cancer-death effect estimates and intervals, all-cause deaths, site-specific outcomes, interval cancers, diagnostic/treatment harms and excess incidence after sufficient follow-up. Controls: Usual-care screening retained, matched invitations/follow-up, prespecified assay versions, blinded outcome review, systematic registry linkage and intention-to-screen analysis. Supporting: A replicated lower cancer-mortality rate with net clinical benefit and explicit baseline/site boundaries; diagnostic and treatment pathways support independently validated mediation. Falsifying: No meaningful death reduction with adequate exposure, adherence and horizon, or offsetting procedure/treatment harms; favorable stage subsets cannot erase a null or adverse mortality result.

KNOWLEDGE GAPS

Exact Galleri randomized cancer-mortality follow-up rather than surrogate stage outcomes

Site-specific and assay-version-specific stage-to-mortality validation

Systematic interval-cancer and false-negative follow-up, especially in real-world use

Overdiagnosis versus beneficial interception, requiring long-term excess-incidence and outcome assessment

Diagnostic capacity, patient-level harms and procedure/treatment burden under usual implementation

Performance transportability across risk, ethnicity, prior cancer and care systems

Independent general-population MRI and DTC benefit-harm trials rather than hereditary or diagnostic-referral extrapolation

REVIEWER NOTES

Fifty-eight originals were actually presented and manually appraised. Selected original Methods/Results were read for UKCTOCS 33991479, Galleri real-world 41173830 and MRI/biopsy SHIP 32860149. PATHFINDER and DETECT-A public XML requests failed 500; their original abstracts/status were read and no body access is claimed. Latest NHS originals and official trial follow-up status were verified, with NEJM full-body 403. Attached assay corrections are separately checked and do not add primary counts. Observed Consensus output is reconciled separately, including its wrong NHS age/enrolment denominator. No exhaustive systematic review, full-paper/supplement audit, blind calibration or scientific peer review is claimed.

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

Consensus and original-source reconciliation

Consensus correctly reports a failed primary stage III/IV endpoint and a favorable secondary stage IV contrast, but uses 142924 enrolled participants aged 50–79. The original primary report actually randomized 142250 aged 50–77. It gives primary IRR 1.03, 95% CI 0.92–1.14, P=.63 and secondary stage IV IRR 0.86, CI 0.74–1.00. Neither endpoint is mortality. The publisher confirms online publication 22 September 2026; the official trial site says cancer-death analyses continue. Provider 'No' is interpreted as benefit unestablished, not a completed death trial proving no effect. Additional provider event counts not verified at original abstract depth are not adopted. Original PMID42776074.

Consensus's general high-specificity and tissue-origin discussion is supported, but original metrics are reported by round rather than as one universal percentage. PPV 58.0%, 50.4%, 45.8% and all-cancer episode sensitivity 26.7–37.2% accompany specificity 99.5–99.6%; 937 detected primary cancers are from the same NHS parent. These descriptive analyses had no hypothesis testing. A high NPV does not establish absence of every cancer or saved lives. Original PMID42773209.

Consensus accurately identifies 6621 evaluable participants, 92 signals, 35 confirmed cancers, 57 false positives and median 162-day false-positive diagnostic resolution. The original abstract reports procedures in 17 of 57 false positives. 62% is the proportion among positive signals, not among all screened people. The provider's 86 negative-screen cancers/sensitivity 29.5% were not verified from the available original abstract or body at this access depth and are not used as load-bearing source facts. Public XML returned 500; the receipt remains abstract/status. Original PMID37805216.

Consensus groups prospective/ambulatory PPVs into ranges without resolving this series' incomplete outcome ascertainment. Actual original Methods/Results show 111080 test results, 1011 positives, 459 voluntary outcome reports and 411 completed workups. Among completed asymptomatic workups, 128 of 259 had invasive cancer, ePPV 49.4%; symptomatic and known-active-cancer orders are different populations. Negative outcomes were not collected systematically. Neither the overall test count nor selected reported outcomes establish complete screening sensitivity or mortality benefit. Original PMID41173830.

Consensus's approximately 17% stage I versus 90% stage IV sensitivities match the original CCGA validation abstract: 16.8% and 90.1%. The study included 4077 validation participants with 2823 known cancers and 1254 controls. That case-control spectrum and prevalence differ from prospective population screening, so these values do not replace NHS screening episode sensitivity or directly validate a mortality model. Original PMID34176681.

Consensus discusses screening biases but does not resolve this decisive original mortality comparator. Selected original UKCTOCS Methods/Results verify randomized intention-to-screen analysis, masked outcome review and 16.3-year median follow-up. Multimodal screening yielded 47.2% more stage I and 24.5% fewer stage IV ovarian/tubal cancers, while deaths did not fall (primary P=.58). This directly challenges a universal stage-surrogate bridge without deciding the eventual Galleri death endpoint. Original PMID33991479.

The prompt requested DETECT-A, but the observed provider answer does not adjudicate its original details. The original prospective report found 26 cancers by blood testing, 24 via usual screening and 46 by neither in 10006 women, with a different blood-test/PET-CT package. Nine of 15 who underwent PET-CT had surgical excision, and 0.22% of all participants had a futile invasive procedure. Curative-intent treatment is preserved but is not proven cure or a randomized death benefit. Original PMID32345712.

The provider table cites a blood-screening study for generic whole-body MRI claims. Actual SHIP original Methods/Results show optional MRI participation plus laboratory disclosures, biopsy-rate increase from 6.9 to 9.9 per 100 observation-years and predominately nonmalignant results, with regional biopsy coverage limits. This supplies MRI-specific cascade evidence, while selection/concurrent disclosure prevents a claim that every additional biopsy was caused by MRI or that net clinical harm is established. Original PMID32860149.

Consensus describes DTC genomic actionability/anxiety broadly without an exact original quantitative comparison. The original 49-sample referred confirmation series found 40% of reported variants false positive, and some increased-risk labels benign. Referral enrichment and the variant denominator are central: this is not 40% of all customers or regulated consumer test results. It also does not establish that DTC risk genotyping has the same purpose or clinical effect as MCED screening. Original PMID29565420.

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: Multi-Cancer Blood Screening and Mortality

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

2009 — Healthy-volunteer whole-body MRI shows frequent incidental findings, motivating disclosure and follow-up questions. Original PMID18818038.

2011 — NLST directly demonstrates lung-mortality benefit from an allocated screening programme in high-risk smokers; other-assay mortality proof is possible but not transferable. Original PMID21714641.

2013 — SHIP quantifies general-population MRI incidental findings and management, with many abnormalities still uncertain. Original PMID22911290.

2013 — Incidental-finding recipients report distress despite high satisfaction, distinguishing experience from outcome benefit. Original PMID23239059.

2016 — Hereditary TP53 comprehensive surveillance is associated with survival in a self-selected high-risk cohort; this is a distinct package and population. Original PMID27501770.

2018 — Cancer SEEK provides original multi-analyte case-control discrimination, requiring subsequent prospective clinical validation. Original PMID29348365.

2018 — DTC raw-data clinical confirmation reveals selected false variants and interpretation errors, with a referral-based denominator. Original PMID29565420.

2019 — Extended NLST follow-up illustrates horizon-dependent effects and why repeated reports must retain one parent trial. Original PMID31260833.

2020 — NELSON provides another true high-risk lung-mortality screening result with long outcome follow-up. Original PMID31995683.

2020 — DETECT-A prospectively combines blood testing with PET-CT and finds treatable cancers, without a randomized mortality control. Original PMID32345712.

2020 — Stored PanSeer samples show potential years-earlier signals; retrospective detection does not itself demonstrate intervention utility. Original PMID32694610.

2021 — UKCTOCS directly separates favorable stage movement from unchanged ovarian/tubal mortality after 16.3 years. Original PMID33991479.

2021 — CCGA's independent validation characterizes stage-sensitive diagnostic performance in known cases and controls. Original PMID34176681.

2022 — Nord ICC reduces colorectal incidence at 10 years while mortality remains nonsignificant, emphasizing outcome hierarchy and screening uptake. Original PMID36214590.

2023 — PATHFINDER measures real prospective detection and diagnostic workups, with false-positive delays and no mortality comparison. Original PMID37805216.

2024 — CAP's 15-year secondary prostate-death effect demonstrates the importance of long follow-up, absolute effect and disease-specific context. Original PMID38581198.

2025 — Large real-world Galleri data expose voluntary outcome reporting and symptom-specific denominators that constrain interpretation. Original PMID41173830.

2026 — NHS intervention-arm performance provides population-screening metrics, not a second randomized mortality result. Original PMID42773209.

2026 — The current NHS randomized primary combined-stage result is null, with a favorable secondary stage IV signal and mortality follow-up still needed. Original PMID42776074.

Research gaps

Exact Galleri randomized cancer-mortality follow-up rather than surrogate stage outcomes

Site-specific and assay-version-specific stage-to-mortality validation

Systematic interval-cancer and false-negative follow-up, especially in real-world use

Overdiagnosis versus beneficial interception, requiring long-term excess-incidence and outcome assessment

Diagnostic capacity, patient-level harms and procedure/treatment burden under usual implementation

Performance transportability across risk, ethnicity, prior cancer and care systems

Independent general-population MRI and DTC benefit-harm trials rather than hereditary or diagnostic-referral extrapolation

Boundary experiment

Continue or run a randomized intention-to-screen comparison of specified Galleri plus usual care versus usual care in asymptomatic adults ≥50, preserving established screening and treatment access. Prespecify cancer-specific mortality, all-cause mortality, site-specific deaths, diagnostic procedures, treatment morbidity, interval cancers and sustained incidence differences over an adequate horizon. Use blinded death adjudication and systematic outcome linkage in both arms. Stage shifts are candidate mediators, with predefined surrogate validation rather than conversion into modeled saved lives.

ADVERSARIAL REVIEW AND APPROVAL

Authorship of record: Codex (OpenAI), as recorded in the score card and provenance files (an earlier "Sol 6.1" label is a house label; no model version is inferred). Adversarial reviewer: Claude Opus 5.5 (Anthropic). Date: 2026-10-14.

Scope of checks. All 58 REFERENCES PMIDs verified against PubMed get_article_metadata: 58/58 resolve to the stated first author, journal and DOI; 0 failures; no retraction notice found. Eleven load-bearing claims were spot-checked against the cited abstracts: the NHS-Galleri randomisation total and age band, the stage III/IV primary incidence-rate ratio with its interval and P value, the stage IV secondary ratio, the 937 MCED-detected primaries, the by-round positive predictive values, specificity and episode sensitivity, the CCGA validation set composition with its stage I and stage IV sensitivities, the PATHFINDER signal, true-positive, false-positive, median-resolution and procedure counts, the UKCTOCS stage I and stage IV changes with the unchanged ovarian and tubal mortality and its P values at 16.3 years median follow-up, and the DETECT-A detection split and futile-procedure rate. Every checked figure matched its source. Scoring logic was re-derived: caps precede aggregation, the declared primary-subclaim rule is applied to SC2, the rank-1 model's E level is a model level and is correctly not used to set the headline, and the B/G/D boundary and the Z independence basis were checked. Framework validator re-run.

Findings

Rejected findings (checked, no change). (i) "Best-supported rival" label: here competing_models.models_evaluated[0] is M4 with is_nominal: false and rank_in_tournament: 1, so the label is correct and the printed order matches rank. The renderer defect seen in sibling reviews is absent here. (ii) Collective authors: references 48 and 50 render "National Lung Screening Trial Research Team" correctly, both with and without a following named author. No leading-comma defect. (iii) Print or issue year versus PubMed e-pub year differing by one, for references 9, 39, 42, 44 and 46. Declared non-error. (iv) PubMed title fields dropping italicised gene or species names. Declared non-error. (v) Causal tally: the prose "0 positive ... 5 screened discriminating contrast units" matches the scorecard. The eight numbered entries are the decisive-perturbation inventory; the five discriminating contrasts are the subset that actually contrast allocated arms. No mismatch. (vi) No spliced "The earlier adversarial approval refers to its original edition" sentence is present, and the REVIEWER NOTES do not contradict the Consensus reconciliation section: both record the provider's wrong enrolment denominator and age band. (vii) Causal role "epiphenomenon": this is the Step-2 label for the unverified saved-lives bridge, and the review states explicitly that it does not mean the detections are imaginary. A defensible judgement call, left unchanged. (viii) Levels: SC1 to SC4 and the headline follow from the quoted caps and the declared aggregation rule; no letter changed.

Scorecard change. final_score.confidence_rationale and extended_scoring_MSD.momentum.artifact and its duplicated justification only. The level, confidence, subclaim levels, tally and every other field are unchanged. Validator re-run: PASS.

Residual limitations. Three of 58 records were examined at partial-body depth; the rest are abstract or status examinations, as the banner states. PATHFINDER and DETECT-A full-text requests failed and no body access is claimed. The NHS-Galleri cancer-death endpoint had not reported as of the stated cutoff, so the review's central gap is a gap in the literature rather than in this audit.

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

Second adversarial pass (2026-11-02): corrections applied

Adversarial reviewer: Claude Opus 5.5 (Anthropic). Date: 2026-11-02. Corrections applied: 2026-11-02, at the principal investigator's authorisation. Full log: ADVERSARIAL_REVIEW.mced-cancer-mortality.md and adversarial-findings.mced-cancer-mortality.json.

Level Γ and confidence High are unchanged, as the reviewer recorded. Applied: S1 (causal-role label qualified as an unproven link, not a proven absence of effect); S2 (Momentum changed from Decisive work underway to Mechanistic only, because NHS-Galleri mortality is a secondary outcome the trial was not powered for and cannot be formally tested after the failed primary endpoint); S3 (stage IV secondary result described as untested, with the stage III offset and the null fatal-cancer composite); S4 (SC1 restated as the proposition actually graded; In vitro-only hard rule marked not triggered); S5 (discriminating contrasts corrected from 5 to 4: [1], [48], [49], [51]); S6 (UKCTOCS combined stage III/IV change not significant; Feng et al. 2024 meta-analysis added as reference 60, Tier B); S7 (PATHFINDER 2 added as reference 59); S8 (confidence rationale rests on the well-characterised ceiling, not on other-modality trials); S9 (fused words repaired in the evidence cards and reader summary); S10 (shedding-biased detection added as a fifth model, without re-ranking); S11 (authorship reconciled to the documented record, Codex).

Two statements in the 2026-10-14 section above are not sustained and are retained only as history: its rejected finding (v) describes the discriminating contrasts as the allocated-arm subset, which they were not (corrected under S5); and its M1 states that every evidence card was repaired, which it was not (corrected under S9).

The figures for stage III/IV counts, the fatal-cancer composite, the testing status of secondary endpoints and the mortality-power statement come from the PMC author manuscript of reference 1 (PMC7619610), as examined by the adversarial reviewer; the original examination of reference 1 was at abstract depth.

REFERENCES

1. Sasieni P, Johnson P, Round T et al. Effect of Screening with Multicancer Early-Detection Test on Late-Stage Cancer Diagnosis. The New England journal of medicine. 2026. PMID 42776074. DOI: 10.1056/NEJMoa2505723. Access: abstract_metadata. Study: NHS_Galleri_NCT05611632.

2. Neal RD, Dolly S, Johnson P et al. Performance of a multi-cancer early detection test in the randomized controlled NHS-Galleri trial. Nature medicine. 2026. PMID 42773209. DOI: 10.1038/s41591-026-04652-8. Access: abstract_metadata. Study: NHS_Galleri_NCT05611632.

3. Swanton C, Bachtiar V, Mathews C et al. NHS-Galleri trial: Enriched enrolment approaches and sociodemographic characteristics of enrolled participants. Clinical trials (London, England). 2025. PMID 39862108. DOI: 10.1177/17407745241302477. Access: abstract_metadata. Study: NHS_Galleri_NCT05611632.

4. Lowenhoff I, Dolly S, Dowinton Smith R et al. Clinical referral to the NHS following multi-cancer early detection test results from the NHS-Galleri trial. Frontiers in oncology. 2025. PMID 40406252. DOI: 10.3389/fonc.2025.1511816. Access: abstract_metadata. Study: NHS_Galleri_NCT05611632.

5. Liu MC, Oxnard GR, Klein EA et al. Sensitive and specific multi-cancer detection and localization using methylation signatures in cell-free DNA. Annals of oncology : official journal of the European Society for Medical Oncology. 2020. PMID 33506766. DOI: 10.1016/j.annonc.2020.02.011. Access: abstract_metadata. Study: CCGA_STRIVE_assay_development.

6. Klein EA, Richards D, Cohn A et al. Clinical validation of a targeted methylation-based multi-cancer early detection test using an independent validation set. Annals of oncology : official journal of the European Society for Medical Oncology. 2021. PMID 34176681. DOI: 10.1016/j.annonc.2021.05.806. Access: abstract_metadata. Study: CCGA3_NCT02889978.

7. Jamshidi A, Liu MC, Klein EA et al. Evaluation of cell-free DNA approaches for multi-cancer early detection. Cancer cell. 2022. PMID 36400018. DOI: 10.1016/j.ccell.2022.10.022. Access: abstract_metadata. Study: CCGA1_NCT02889978.

8. Bredno J, Lipson J, Venn O et al. Clinical correlates of circulating cell-free DNA tumor fraction. PloS one. 2021. PMID 34432811. DOI: 10.1371/journal.pone.0256436. Access: abstract_metadata. Study: CCGA1_and2_NCT02889978.

9. Tang WHW, Yimer H, Tummala M et al. Performance of a targeted methylation-based multi-cancer early detection test by race and ethnicity. Preventive medicine. 2023. PMID 36495927. DOI: 10.1016/j.ypmed.2022.107384. Access: abstract_metadata. Study: CCGA3_NCT02889978.

10. Schrag D, Beer TM, Mc Donnell CH et al. Blood-based tests for multicancer early detection (PATHFINDER): a prospective cohort study. Lancet (London, England). 2023. PMID 37805216. DOI: 10.1016/S0140-6736(23)01700-2. Access: abstract_metadata. Study: PATHFINDER_NCT04241796.

11. Lennon AM, Buchanan AH, Kinde I et al. Feasibility of blood testing combined with PET-CT to screen for cancer and guide intervention. Science (New York, N.Y.). 2020. PMID 32345712. DOI: 10.1126/science.abb9601. Access: abstract_metadata. Study: PMID32345712.

12. Bao H, Yang S, Chen X et al. Early detection of multiple cancer types using multidimensional cell-free DNA fragmentomics. Nature medicine. 2025. PMID 40425843. DOI: 10.1038/s41591-025-03735-2. Access: abstract_metadata. Study: PMID40425843.

13. Taylor MS, Wu C, Fridy PC et al. Ultrasensitive Detection of Circulating LINE-1 ORF1p as a Specific Multicancer Biomarker. Cancer discovery. 2023. PMID 37698949. DOI: 10.1158/2159-8290.CD-23-0313. Access: abstract_metadata. Study: PMID37698949.

14. Gao Q, Lin YP, Li BS et al. Unintrusive multi-cancer detection by circulating cell-free DNA methylation sequencing (THUNDER): development and independent validation studies. Annals of oncology : official journal of the European Society for Medical Oncology. 2023. PMID 36849097. DOI: 10.1016/j.annonc.2023.02.010. Access: abstract_metadata. Study: PMID36849097.

15. Jin F, Yang P, Xu R et al. Phosphatidylserine-positive extracellular vesicles for detecting multiple operable cancers. Cell reports. 2025. PMID 40934080. DOI: 10.1016/j.celrep.2025.116259. Access: abstract_metadata. Study: PMID40934080.

16. Wang Y, Joshu CE, Curtis SD et al. Detection of Cancers Three Years prior to Diagnosis Using Plasma Cell-Free DNA. Cancer discovery. 2025. PMID 40402478. DOI: 10.1158/2159-8290.CD-25-0375. Access: abstract_metadata. Study: PMID40402478.

17. Dong W, Lau CH, Li J et al. Pan-cancer methylation analysis of circulating cell-free DNA. Cancer genetics. 2025. PMID 40737714. DOI: 10.1016/j.cancergen.2025.07.014. Access: abstract_metadata. Study: PMID40737714.

18. Matrana M, Shukla V, Kingsbury D et al. Real-world data and clinical experience from over 100, 000 multi-cancer early detection tests. Nature communications. 2025. PMID 41173830. DOI: 10.1038/s41467-025-64094-7. Access: full_text_partial. Study: PMID41173830.

19. Nguyen VTC, Nguyen TH, Doan NNT et al. Multimodal analysis of methylomics and fragmentomics in plasma cell-free DNA for multi-cancer early detection and localization. e Life. 2023. PMID 37819044. DOI: 10.7554/eLife.89083. Access: abstract_metadata. Study: SPOTMAS_case_control_development.

20. Michel M, Heidary M, Mechri A et al. Noninvasive Multicancer Detection Using DNA Hypomethylation of LINE-1 Retrotransposons. Clinical cancer research : an official journal of the American Association for Cancer Research. 2025. PMID 39620930. DOI: 10.1158/1078-0432.CCR-24-2669. Access: abstract_metadata. Study: PMID39620930.

21. Zhang J, Rui H, Hu H. Noninvasive multi-cancer detection using blood-based cell-free micro RNAs. Scientific reports. 2024. PMID 39333750. DOI: 10.1038/s41598-024-73783-0. Access: abstract_metadata. Study: PMID39333750.

22. Fang J, Luo S, Li S et al. Detection of primary cancer types via fragment size selection in circulating cell-free extrachromosomal circular DNA. Genome medicine. 2026. PMID 41495807. DOI: 10.1186/s13073-025-01595-6. Access: abstract_metadata. Study: PMID41495807.

23. Bratulic S, Limeta A, Dabestani S et al. Noninvasive detection of any-stage cancer using free glycosaminoglycans. Proceedings of the National Academy of Sciences of the United States of America. 2022. PMID 36469776. DOI: 10.1073/pnas.2115328119. Access: abstract_metadata. Study: PMID36469776.

24. Brito-Rocha T, Constâncio V, Leite-Silva P et al. Multi-cancer early detection via a DNA methylation multiplex dd PCR-based blood test. International journal of cancer. 2025. PMID 40317486. DOI: 10.1002/ijc.35467. Access: abstract_metadata. Study: PMID40317486.

25. Huang A, Guo DZ, Su ZX et al. GUIDE: a prospective cohort study for blood-based early detection of gastrointestinal cancers using targeted DNA methylation and fragmentomics sequencing. Molecular cancer. 2025. PMID 40468355. DOI: 10.1186/s12943-025-02367-x. Access: abstract_metadata. Study: PMID40468355.

26. Nguyen LHD, Nguyen THH, Le VH et al. Prospective validation study: a non-invasive circulating tumor DNA-based assay for simultaneous early detection of multiple cancers in asymptomatic adults. BMC medicine. 2025. PMID 39948555. DOI: 10.1186/s12916-025-03929-y. Access: abstract_metadata. Study: K_DETEK_NCT05227261.

27. Curtis SD, Liu T, Bai Y et al. Fragmentation signatures in cancer patients resemble those of patients with vascular or autoimmune diseases. Proceedings of the National Academy of Sciences of the United States of America. 2025. PMID 40833414. DOI: 10.1073/pnas.2426890122. Access: abstract_metadata. Study: PMID40833414.

28. Held D, Bolland S, Freese R et al. A Protein-Based Blood Test for Multi-Cancer Diagnostics. Biomedicines. 2025. PMID 41153790. DOI: 10.3390/biomedicines13102510. Access: abstract_metadata. Study: PMID41153790.

29. Hinestrosa JP, Kurzrock R, Lewis JM et al. Early-stage multi-cancer detection using an extracellular vesicle protein-based blood test. Communications medicine. 2022. PMID 35603292. DOI: 10.1038/s43856-022-00088-6. Access: abstract_metadata. Study: PMID35603292.

30. Marinac CR, O'Donnell EK, Clarke CA et al. Multicancer Early Detection Blood Test Performance in Cancer Survivors From the PATHFINDER Study. JCO precision oncology. 2025. PMID 41411611. DOI: 10.1200/PO-25-00498. Access: abstract_metadata. Study: PATHFINDER_NCT04241796.

31. Nadauld L, Mc Donnell CH, Dilaveri CA et al. Psychosocial impact associated with a multicancer early detection test (PATHFINDER): a prospective, multicentre, cohort study. The Lancet. Oncology. 2025. PMID 39818231. DOI: 10.1016/S1470-2045(24)00645-4. Access: abstract_metadata. Study: PATHFINDER_NCT04241796.

32. Marlow LAV, Schmeising-Barnes N, Waller J. The impact of cancer expectations on psychological responses following a cancer signal detected result in asymptomatic multi-cancer detection (MCED) testing. British journal of cancer. 2025. PMID 40926064. DOI: 10.1038/s41416-025-03180-5. Access: abstract_metadata. Study: NHS_Galleri_NCT05611632.

33. Schmeising-Barnes N, Waller J, Marlow LAV. Attitudes to multi-cancer early detection (MCED) blood tests for population-based screening: A qualitative study in Great Britain. Social science & medicine (1982). 2024. PMID 38520825. DOI: 10.1016/j.socscimed.2024.116762. Access: abstract_metadata. Study: PMID38520825.

34. Turriff AE, Annunziata CM, Malayeri AA et al. Prenatal cf DNA Sequencing and Incidental Detection of Maternal Cancer. The New England journal of medicine. 2024. PMID 39774314. DOI: 10.1056/NEJMoa2401029. Access: abstract_metadata. Study: PMID39774314.

35. Cohen JD, Li L, Wang Y et al. Detection and localization of surgically resectable cancers with a multi-analyte blood test. Science (New York, N.Y.). 2018. PMID 29348365. DOI: 10.1126/science.aar3247. Access: abstract_metadata. Study: PMID29348365.

36. Chen X, Gole J, Gore A et al. Non-invasive early detection of cancer four years before conventional diagnosis using a blood test. Nature communications. 2020. PMID 32694610. DOI: 10.1038/s41467-020-17316-z. Access: abstract_metadata. Study: PMID32694610.

37. Villani A, Shore A, Wasserman JD et al. Biochemical and imaging surveillance in germline TP53 mutation carriers with Li-Fraumeni syndrome: 11 year follow-up of a prospective observational study. The Lancet. Oncology. 2016. PMID 27501770. DOI: 10.1016/S1470-2045(16)30249-2. Access: abstract_metadata. Study: TP53_comprehensive_surveillance_89.

38. Omran M, Tham E, Brandberg Y et al. Whole-Body MRI Surveillance-Baseline Findings in the Swedish Multicentre Hereditary TP53-Related Cancer Syndrome Study (SWEP53). Cancers. 2022. PMID 35053544. DOI: 10.3390/cancers14020380. Access: abstract_metadata. Study: SWEP53_TP53_baseline.

39. Schmidt CO, Sierocinski E, Hegenscheid K et al. Impact of whole-body MRI in a general population study. European journal of epidemiology. 2016. PMID 26597793. DOI: 10.1007/s10654-015-0101-y. Access: abstract_metadata. Study: SHIP_Trend_whole_body_MRI.

40. Schmidt CO, Sierocinski E, Baumeister S et al. Effects of whole-body MRI on outpatient health service costs: a general-population prospective cohort study in Mecklenburg-Vorpommern, Germany. BMJ open. 2022. PMID 34996801. DOI: 10.1136/bmjopen-2021-056572. Access: abstract_metadata. Study: SHIP_and_SHIP-Trend_MRI.

41. Sodde P, Hyder Z, Pugh S et al. Evaluation of whole-body MRI for cancer early detection in Li-Fraumeni syndrome. Journal of medical genetics. 2025. PMID 40588398. DOI: 10.1136/jmg-2025-110704. Access: abstract_metadata. Study: PMID40588398.

42. Morin SH, Cobbold JF, Lim AK et al. Incidental findings in healthy control research subjects using whole-body MRI. European journal of radiology. 2009. PMID 18818038. DOI: 10.1016/j.ejrad.2008.08.006. Access: abstract_metadata. Study: PMID18818038.

43. Ulus S, Suleyman E, Ozcan UA et al. Whole-Body MRI Screening in Asymptomatic Subjects; Preliminary Experience and Long-Term Follow-Up Findings. Polish journal of radiology. 2016. PMID 27635171. DOI: 10.12659/PJR.897570. Access: abstract_metadata. Study: PMID27635171.

44. Hegenscheid K, Seipel R, Schmidt CO et al. Potentially relevant incidental findings on research whole-body MRI in the general adult population: frequencies and management. European radiology. 2013. PMID 22911290. DOI: 10.1007/s00330-012-2636-6. Access: abstract_metadata. Study: SHIP_whole_body_MRI.

45. Richter A, Sierocinski E, Singer S et al. The effects of incidental findings from whole-body MRI on the frequency of biopsies and detected malignancies or benign conditions in a general population cohort study. European journal of epidemiology. 2020. PMID 32860149. DOI: 10.1007/s10654-020-00679-4. Access: full_text_partial. Study: SHIP_and_SHIP-Trend_MRI.

46. Schmidt CO, Hegenscheid K, Erdmann P et al. Psychosocial consequences and severity of disclosed incidental findings from whole-body MRI in a general population study. European radiology. 2013. PMID 23239059. DOI: 10.1007/s00330-012-2723-8. Access: abstract_metadata. Study: SHIP_whole_body_MRI.

47. Tandy-Connor S, Guiltinan J, Krempely K et al. False-positive results released by direct-to-consumer genetic tests highlight the importance of clinical confirmation testing for appropriate patient care. Genetics in medicine : official journal of the American College of Medical Genetics. 2018. PMID 29565420. DOI: 10.1038/gim.2018.38. Access: abstract_metadata. Study: PMID29565420.

48. National Lung Screening Trial Research Team, Aberle DR, Adams AM et al. Reduced lung-cancer mortality with low-dose computed tomographic screening. The New England journal of medicine. 2011. PMID 21714641. DOI: 10.1056/NEJMoa1102873. Access: abstract_metadata. Study: NLST_NCT00047385.

49. de Koning HJ, van der Aalst CM, de Jong PA et al. Reduced Lung-Cancer Mortality with Volume CT Screening in a Randomized Trial. The New England journal of medicine. 2020. PMID 31995683. DOI: 10.1056/NEJMoa1911793. Access: abstract_metadata. Study: NELSON_NL580.

50. National Lung Screening Trial Research Team. Lung Cancer Incidence and Mortality with Extended Follow-up in the National Lung Screening Trial. Journal of thoracic oncology : official publication of the International Association for the Study of Lung Cancer. 2019. PMID 31260833. DOI: 10.1016/j.jtho.2019.05.044. Access: abstract_metadata. Study: NLST_NCT00047385.

51. Menon U, Gentry-Maharaj A, Burnell M et al. Ovarian cancer population screening and mortality after long-term follow-up in the UK Collaborative Trial of Ovarian Cancer Screening (UKCTOCS): a randomised controlled trial. Lancet (London, England). 2021. PMID 33991479. DOI: 10.1016/S0140-6736(21)00731-5. Access: full_text_partial. Study: UKCTOCS_NCT00058032.

52. Menon U, Gentry-Maharaj A, Burnell M et al. Mortality impact, risks, and benefits of general population screening for ovarian cancer: the UKCTOCS randomised controlled trial. Health technology assessment (Winchester, England). 2025. PMID 37183782. DOI: 10.3310/BHBR5832. Access: abstract_metadata. Study: UKCTOCS_NCT00058032.

53. Martin RM, Turner EL, Young GJ et al. Prostate-Specific Antigen Screening and 15-Year Prostate Cancer Mortality: A Secondary Analysis of the CAP Randomized Clinical Trial. JAMA. 2024. PMID 38581198. DOI: 10.1001/jama.2024.4011. Access: abstract_metadata. Study: CAP_ISRCTN92187251.

54. Bretthauer M, Løberg M, Wieszczy P et al. Effect of Colonoscopy Screening on Risks of Colorectal Cancer and Related Death. The New England journal of medicine. 2022. PMID 36214590. DOI: 10.1056/NEJMoa2208375. Access: abstract_metadata. Study: Nord ICC_NCT00883792.

55. Kaminski MF, Kalager M, Løberg M et al. Long-term effects of colonoscopy screening on colorectal cancer incidence and mortality: a multicountry, population-based randomised controlled trial. Lancet (London, England). 2026. PMID 42102826. DOI: 10.1016/S0140-6736(26)00508-8. Access: abstract_metadata. Study: Nord ICC_NCT00883792.

56. Roobol MJ, de Vos II, Månsson M et al. European Study of Prostate Cancer Screening - 23-Year Follow-up. The New England journal of medicine. 2025. PMID 41160819. DOI: 10.1056/NEJMoa2503223. Access: abstract_metadata. Study: ERSPC_ISRCTN49127736.

57. Duffy SW, Vulkan D, Cuckle H et al. Effect of mammographic screening from age 40 years on breast cancer mortality (UK Age trial): final results of a randomised, controlled trial. The Lancet. Oncology. 2020. PMID 32800099. DOI: 10.1016/S1470-2045(20)30398-3. Access: abstract_metadata. Study: PMID32800099.

58. Field JK, Duffy SW, Baldwin DR et al. The UK Lung Cancer Screening Trial: a pilot randomised controlled trial of low-dose computed tomography screening for the early detection of lung cancer. Health technology assessment (Winchester, England). 2016. PMID 27224642. DOI: 10.3310/hta20400. Access: abstract_metadata. Study: UKLS_ISRCTN78513845.

59. Nabavizadeh N, McDonnell C, Kurbegov D et al. Performance and safety of a multi-cancer early detection test: the PATHFINDER 2 study. Nature medicine. 2026. PMID 42773207. DOI: 10.1038/s41591-026-04618-w. Access: abstract_metadata. Study: PATHFINDER2_NCT05155605. Added 2026-11-02 (adversarial finding S7); PubMed record verified 2026-11-02.

60. Feng X, Zahed H, Onwuka J et al. Cancer Stage Compared With Mortality as End Points in Randomized Clinical Trials of Cancer Screening: A Systematic Review and Meta-Analysis. JAMA. 2024. PMID 38583868. DOI: 10.1001/jama.2024.5814. Access: abstract_metadata. Tier B (systematic review and meta-analysis; frames, does not carry experimental numbers). Study: PMID38583868. Added 2026-11-02 (adversarial finding S6); PubMed record verified 2026-11-02.

Source-specific evidence cards

[1] Effect of Screening with Multicancer Early-Detection Test on Late-Stage Cancer Diagnosis.

Shows: NHS-Galleri randomized 142250 adults 50–77 to tested versus stored blood plus usual care for 3 annual rounds; primary stage III/IV incidence IRR 1.03, 95% CI 0.92–1.14, P=.63; key secondary stage IV IRR 0.86, 95% CI 0.74–1.00, not formally tested under the fixed-sequence plan and with an unadjusted interval; the stage IV decrease (341 vs 394) was offset by more stage III cancers (360 vs 287), and the secondary stage III/IV-or-fatal composite was null (IRR 1.04, 0.97–1.12; PMC7619610).

Does not show: Primary proxy failed, secondary signal retained; no mortality result in this publication. Neither randomized design nor a stage IV contrast supplies a positive cancer-mortality crossing. Publisher abstract verified 22 September 2026; full-body access not claimed.

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.

[2] Performance of a multi-cancer early detection test in the randomized controlled NHS-Galleri trial.

Shows: Same NHS trial descriptive test-performance report across 3 rounds: 937 detected primary cancers, PPV 58.0/50.4/45.8%, all-cancer episode sensitivity 26.7–37.2%, specificity 99.5–99.6%.

Does not show: Descriptive intervention-arm diagnosis metrics without hypothesis testing; not an independent randomized trial or mortality analysis. High NPV largely reflects low prevalence and does not mean every cancer is excluded.

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.

[3] NHS-Galleri trial: Enriched enrolment approaches and sociodemographic characteristics of enrolled participants.

Shows: Original NHS recruitment analysis 1.5 million invitations/142924 enrolled; older/deprived groups enriched but Black/Asian groups underrepresented despite targeting.

Does not show: Same parent trial, recruitment denominators differ from final randomized 142250. Sociodemographic reach does not establish efficacy, and enrollment composition limits transportability.

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.

[4] Clinical referral to the NHS following multi-cancer early detection test results from the NHS-Galleri trial.

Shows: Original NHS operational report describes nurse referral into 56 Trusts with CSO-guided pathways and safety-netting.

Does not show: Implementation feasibility and coordination are meaningful but not mortality benefit, diagnostic accuracy estimates or a separate efficacy 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.

[5] Sensitive and specific multi-cancer detection and localization using methylation signatures in cell-free DNA.

Shows: CCGA/STRIVE case-control training/validation cf DNA methylation; specificity 99.3%, all-cancer stage I sensitivity 18% rising to 93% stage IV.

Does not show: Known-cancer versus control design and version differ from population screening; sensitivity increases with burden. Detection feasibility does not validate stage-to-mortality 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, SC3.

[6] Clinical validation of a targeted methylation-based multi-cancer early detection test using an independent validation set.

Shows: CCGA independent validation 4077 participants including 2823 known cancers; specificity 99.5%, overall sensitivity 51.5%, stage I 16.8%, tissue origin 88.7%.

Does not show: Case-control spectrum/prevalence differs from asymptomatic screening; independent validation set is not an independent mortality perturbation. Stage-specific sensitivity is descriptive.

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.

[7] Evaluation of cell-free DNA approaches for multi-cancer early detection.

Shows: CCGA1 compares 10 cf DNA classifiers at 98% specificity; tumor allele fraction predicts detection and methylation localizes origin.

Does not show: Same assay-development samples; classifier optimization and limit of detection do not show clinical utility, treatment benefit or prospective-screening accuracy.

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.

[8] Clinical correlates of circulating cell-free DNA tumor fraction.

Shows: CCGA tumor-fraction models relate breast/lung tumor mitotic or metabolic volume and colorectal invasion to cf DNA and validate in CCGA2.

Does not show: Biological correlation may explain preferential detection of aggressive cancers; it does not establish that detected cancers are treatable earlier or that overdiagnosis is 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, SC3.

[9] Performance of a targeted methylation-based multi-cancer early detection test by race and ethnicity.

Shows: CCGA3 descriptive racial/ethnic analysis uses same 4077 validation participants; specificity high across groups with some small subgroup samples.

Does not show: Exploratory subgroup intervals and predominantly White cohort limit broad equivalence claims; no independent efficacy trial or mortality result.

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.

[10] Blood-based tests for multicancer early detection (PATHFINDER): a prospective cohort study.

Shows: PATHFINDER convenience cohort ≥50 without cancer symptoms: 6621 analyzable; 92 signals, 35 confirmed, 57 false positives; false-positive diagnostic resolution median 162 days and 30% procedures.

Does not show: Single-arm feasibility primary endpoint is diagnostic workup, not mortality. A low overall false-positive rate and 62% false-positive share among signals are different denominators. Attempted original XML returned 500; abstract examined.

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

[11] Feasibility of blood testing combined with PET-CT to screen for cancer and guide intervention.

Shows: DETECT-A prospective blood-test/PET-CT pathway 10006 women found 26 cancers by blood; 24 by routine screening and 46 neither; 0.22% futile invasive procedures.

Does not show: Different assay plus PET-CT package, no randomized mortality control. Curative-intent surgery is not proven cure or mortality reduction; routine screening remains complementary. Public XML failed 500.

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

[12] Early detection of multiple cancer types using multidimensional cell-free DNA fragmentomics.

Shows: Alternative fragmentomics assay validated in separate known-cancer cohorts and preliminary 3724 asymptomatic cohort with 53.5% sensitivity/98.1% specificity.

Does not show: Not Galleri; case-control high sensitivity falls in screening context. Clinical decision potential and cancer detection do not prove survival 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.

[13] Ultrasensitive Detection of Circulating LINE-1 ORF1p as a Specific Multicancer Biomarker.

Shows: ORF1p ultrasensitive plasma protein assay identifies multicancer and precursor signals, with prognostic associations and response monitoring.

Does not show: Alternative biomarker, not Galleri or randomized early detection. Original correction 39228301 restores a compositor-omitted ovarian tissue-expression pie chart in Fig1D; no changed detection result is specified. Biomarker detection is not a mortality 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. Update Of: PMID36747644; Erratum In: PMID39228301. Subclaims: SC1.

[14] Unintrusive multi-cancer detection by circulating cell-free DNA methylation sequencing (THUNDER): development and independent validation studies.

Shows: THUNDER original case-control independent 505 cancer/505 control validation of 6 cancer methylation panel showed 69.1% sensitivity/98.9% specificity.

Does not show: Predicted late-stage decline and survival gains are simulations, not observed intervention outcomes; new original diagnostic component alone earns report credit. Assay differs from Galleri.

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.

[15] Phosphatidylserine-positive extracellular vesicles for detecting multiple operable cancers.

Shows: Alternative PS-positive vesicle assay 1869 clinical samples and 3 validation cohorts; early-stage sensitivity 74.7% with 89.8% specificity; cells/mice support marker biology.

Does not show: Known-cancer cohort spectrum and relatively lower specificity may matter at low prevalence; no Galleri inference, asymptomatic mortality trial or demonstrated curative 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.

[16] Detection of Cancers Three Years prior to Diagnosis Using Plasma Cell-Free DNA.

Shows: ARIC 26 cases/26 controls with stored serial plasma; 4 of 6 assessable earlier samples contained same mutations 3.1–3.5 years before diagnosis at much lower fraction.

Does not show: Very small nested case-control detection benchmark; retrospective earlier-sample interrogation is not prospective clinical intervention or proof that years-earlier treatment prevents 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: SC1, SC3.

[17] Pan-cancer methylation analysis of circulating cell-free DNA.

Shows: Alternative PCR methylation panel 103 diverse cancer cases/40 healthy controls showed 47.6% sensitivity/90% specificity.

Does not show: Background sentence asserting all-cause mortality reduction is not a study result; no mortality endpoint or population screening. Small proof-of-concept rather than Galleri 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, SC2.

[18] Real-world data and clinical experience from over 100, 000 multi-cancer early detection tests.

Shows: Original M/R: 111080 real-world tests ≥22, 1011 signals; voluntary outcomes 459, 48 ongoing workups, 411 complete. Asymptomatic completed subset 128/259 invasive cancers, ePPV 49.4%; negatives not systematically followed.

Does not show: Symptomatic/active-cancer uses and incomplete positive follow-up preclude full-population sensitivity, universal PPV or mortality claims. Large test count is not complete clinical ascertainment or screening-only cohort.

Access/status: Bounded original primary abstract Results/Findings/conclusion passages and linked-status metadata examined; no complete-abstract audit asserted. Selected body sections: 1:Results:paragraph5, 1:Results:paragraph6, 1:Results:paragraph12, 1:Results:paragraph13, 3:Methods:paragraph1, 3:Methods:paragraph2, 3:Methods:paragraph3, 3:Methods:paragraph4. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2.

[19] Multimodal analysis of methylomics and fragmentomics in plasma cell-free DNA for multi-cancer early detection and localization.

Shows: SPOTMAS methylomics/fragmentomics 738 nonmetastatic cancer cases/1550 healthy controls; 72.4% sensitivity at 97% specificity; stage-dependent results.

Does not show: Alternative assay case-control training/validation, not population predictive values or mortality benefit. Update Of notices carry no supplied PMIDs and are retained as version-history uncertainty.

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. Update Of: PMIDnot supplied; Update Of: PMIDnot supplied; Update Of: PMIDnot supplied. Subclaims: SC1.

[20] Noninvasive Multicancer Detection Using DNA Hypomethylation of LINE-1 Retrotransposons.

Shows: LINE-1 hypomethylation classifiers 747 samples in 2 independent cohorts distinguish 6 cancer plasmas with high AUC.

Does not show: Different reference-free assay; AUC across selected clinical cohorts is not low prevalence screening PPV or clinical mortality 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.

[21] Noninvasive multi-cancer detection using blood-based cell-free micro RNAs.

Shows: Serum miRNA training 2800 plus and independent large case-control validation yields very high reported site-specific sensitivity/specificity.

Does not show: Platform, batch/spectrum and control selection need prospective transportability checks; independent assay validation is not assigned screening benefit or Galleri performance.

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.

[22] Detection of primary cancer types via fragment size selection in circulating cell-free extrachromosomal circular DNA.

Shows: 413 cancer/239 healthy plasma eccDNA data plus a PDX fragment model train ScanTecc; cancer classification AUC 0.92 and origin AUC 0.79.

Does not show: Exploratory assay and in-silico fragment selection cannot establish clinical utility, population PPV or benefit; PDX mechanism does not supply screening mortality evidence.

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.

[23] Noninvasive detection of any-stage cancer using free glycosaminoglycans.

Shows: Free GAG profiles original 1260 subjects and validation screening-like 121/49 cases show lower first-stage detection at ≥99% specificity; undetected cancers have better prognosis.

Does not show: Detection correlates with risk but does not show screening reduces deaths; prognostic selection and known-case sampling constrain survival claims. Alternative assay not Galleri.

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.

[24] Multi-cancer early detection via a DNA methylation multiplex dd PCR-based blood test.

Shows: Novel dd PCR promoter panel 485 plasmas and tissue data achieves 60.1% sensitivity/87.4% specificity across 4 major cancers.

Does not show: Low prevalence false positive implications, small proof-of-concept and other-assay identity limit population screening; no clinical death 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.

[25] GUIDE: a prospective cohort study for blood-based early detection of gastrointestinal cancers using targeted DNA methylation and fragmentomics sequencing.

Shows: GUIDE GutSeer original training/validation and locked 846 person independent GI cohort shows ~81.5% sensitivity/94.4% specificity and precancer detection.

Does not show: Mixed in/outpatient known GI cases and new assay, not Galleri or mortality screening trial; stage detection not benefit proof.

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.

[26] Prospective validation study: a non-invasive circulating tumor DNA-based assay for simultaneous early detection of multiple cancers in asymptomatic adults.

Shows: KDETEK 9057 asymptomatic ≥40 Vietnam; 9024 eligible, 43 signals, 17 malignancies, PPV 39.5%, one-year negative follow-up; TOO 52.9%.

Does not show: Small cancer count and 12 month ascertainment constrain estimates; alternative SPOTMAS and age 40 scope differ. Original calls utility but does not measure mortality utility.

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.

[27] Fragmentation signatures in cancer patients resemble those of patients with vascular or autoimmune diseases.

Shows: Original fragmentomics study shows venous thromboembolism/autoimmune profiles resemble cancer and produce false positives under binary classifiers; adding proteins/multiclass approach improves discrimination.

Does not show: Specificity challenge to this assay class, not direct evidence against Galleri methylation specificity. Original correction 41628341 adds funding acknowledgments rather than numerical results. No mortality endpoint was measured.

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. Erratum In: PMID41628341. Subclaims: SC1.

[28] A Protein-Based Blood Test for Multi-Cancer Diagnostics.

Shows: Alternative 16 protein classifier 141 known cancer cases/119 healthy controls reports 100% sensitivity and 97% specificity.

Does not show: Small case-control model development cannot establish prospective screening 100% sensitivity or survival benefit. Original correction 41595793 changes male breast percentage to 3.3% and specificity 95% CI to 90–99% for breast and 92–99% for lung; authors state conclusions unchanged. This is not Galleri.

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. Erratum In: PMID41595793. Subclaims: SC1.

[29] Early-stage multi-cancer detection using an extracellular vesicle protein-based blood test.

Shows: EV protein pilot 139 pathology confirmed stage I/II cases versus 184 controls; sensitivity 71.2% at 99.5% specificity with large cancer-site differences.

Does not show: Selected case-control pilot, different 3 cancer assay; early-stage discrimination is not clinical screening benefit or all-cancer performance.

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.

[30] Multicancer Early Detection Blood Test Performance in Cancer Survivors From the PATHFINDER Study.

Shows: PATHFINDER 1609 survivor subgroup, 20 signals and 10 new diagnoses including second primaries/recurrences; test metrics described similar.

Does not show: Same parent cohort; survivors differ from average-risk first screening and no random survival comparison. Added subgroups do not create efficacy replication.

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] Psychosocial impact associated with a multicancer early detection test (PATHFINDER): a prospective, multicentre, cohort study.

Shows: PATHFINDER PRO secondary study found transient higher anxiety after signals, mean return toward baseline by 12 months and high satisfaction; response counts differ by group/time.

Does not show: Same parent trial and missing questionnaires; averages/satisfaction cannot exclude individual distress or prove net mortality benefit. No random screening psychosocial 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, SC4.

[32] The impact of cancer expectations on psychological responses following a cancer signal detected result in asymptomatic multi-cancer detection (MCED) testing.

Shows: 41 selected NHS signal-positive interviewees describe expectations, uncertainty and reassurance differences with/without cancer.

Does not show: Purposive qualitative subset not incidence of harm or causal mortality evidence; same NHS parent, self-reported cancer 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, SC4.

[33] Attitudes to multi-cancer early detection (MCED) blood tests for population-based screening: A qualitative study in Great Britain.

Shows: 53 British adults 50–77 focus groups express enthusiasm for easy blood screening and concern about false positives/follow-up procedures.

Does not show: Beliefs about reduced deaths are participants' views rather than evidence; hypothetical acceptability not actual screening benefit/harm rate.

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

[34] Prenatal cf DNA Sequencing and Incidental Detection of Maternal Cancer.

Shows: IDENTIFY selected 107 pregnant/postpartum unusual prenatal cf DNA results; 52 occult cancers, WBMRI sensitivity 98% and specificity 88.5%; certain copy-number pattern strongly predictive.

Does not show: Extremely enriched abnormal-test referred population and pregnancy context; not general-population MRI/Galleri screening performance or mortality reduction.

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

[35] Detection and localization of surgically resectable cancers with a multi-analyte blood test.

Shows: Cancer SEEK 1005 clinically detected nonmetastatic cancers/812 healthy controls showed median 70% detection and >99% specificity.

Does not show: Different mutation/protein assay and known case-control design; no asymptomatic prospective or mortality comparison. Detection/localization not cure.

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.

[36] Non-invasive early detection of cancer four years before conventional diagnosis using a blood test.

Shows: PanSeer stored TZL plasma 605 asymptomatic subjects including 191 subsequent cancers showed prediagnostic signals up to 4 years; 223 postdiagnosis cases also assayed.

Does not show: Nested selected stored samples do not establish live clinical screening intervention or mortality benefit; prospective confirmation required, not Galleri.

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.

[37] Biochemical and imaging surveillance in germline TP53 mutation carriers with Li-Fraumeni syndrome: 11 year follow-up of a prospective observational study.

Shows: 89 TP53 carriers chose comprehensive surveillance or not; 19 crossed over; 59 under surveillance, 5 year postdiagnosis survival 88.8% vs 59.6%, false positives/negatives occurred.

Does not show: As-treated self-selection, lead time and multi-modality bundle in high-risk hereditary families; not isolated MRI causal benefit or general average-risk commercial screening.

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

[38] Whole-Body MRI Surveillance-Baseline Findings in the Swedish Multicentre Hereditary TP53-Related Cancer Syndrome Study (SWEP53).

Shows: SWEP53 baseline 61 adult TP53 carriers; 19 further workups, 3 asymptomatic malignancies; dedicated brain/breast imaging included.

Does not show: High-risk hereditary surveillance and single baseline yield; no assigned mortality contrast or all-comer commercial MRI equivalence.

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.

[39] Impact of whole-body MRI in a general population study.

Shows: SHIP-Trend 4420 participants/2188 MRI; 2.5 year follow-up found negligible overall QoL/depression differences with MRI or incidental disclosures.

Does not show: Optional MRI participation and attrition weighting; overall psychosocial null does not establish clinical benefit or rule out short-term/individual distress.

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.

[40] Effects of whole-body MRI on outpatient health service costs: a general-population prospective cohort study in Mecklenburg-Vorpommern, Germany.

Shows: SHIP 5019 with insurance, 2969 MRI, propensity weighted nonparticipants; incidental findings 32%, outpatient costs about€295/11.6% higher over 2 years.

Does not show: Quasi-experimental, not randomized; MRI plus clinical workup and selection may confound; costs are not automatically medical harms and no cancer mortality benefit comparison.

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.

[41] Evaluation of whole-body MRI for cancer early detection in Li-Fraumeni syndrome.

Shows: Retrospective 75 LFS adults, total 325 WB scans; 9 of 17 cancers MRI-detected, 89 scans benign incidental findings prompted 53 investigations; sensitivity 42.9%.

Does not show: Highly selected TP53 surveillance rather than general population, DTC or Galleri; repeated scans/subjects not independent cases and no controlled mortality 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: SC4.

[42] Incidental findings in healthy control research subjects using whole-body MRI.

Shows: 148 healthy research volunteers, 29.1% abnormal WB scans, 12.8% clinically significant findings; prevalence higher with age/BMI.

Does not show: Research cohort median 36 and single hospital; clinical significance label not proven net benefit or mortality prevention. No comparison intervention.

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

[43] Whole-Body MRI Screening in Asymptomatic Subjects; Preliminary Experience and Long-Term Follow-Up Findings.

Shows: 118 consecutive asymptomatic health check participants; 103 benign lesions in 83 subjects, 2 malignant and 1 precancerous lesion found.

Does not show: Detection yield and anatomic coverage not mortality/net benefit; selected health check cohort and small malignancy denominator.

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

[44] Potentially relevant incidental findings on research whole-body MRI in the general adult population: frequencies and management.

Shows: SHIP 2500 WBMRI; 36.2% potentially relevant incidental findings, 31.5% disclosed, most nature unclear; 9 urgent referrals.

Does not show: Reader/advisory classification is not fully confirmed malignancy; extra organ imaging in subsets and shared SHIP participants require grouping. No mortality 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: SC4.

[45] The effects of incidental findings from whole-body MRI on the frequency of biopsies and detected malignancies or benign conditions in a general population cohort study.

Shows: Original M/R SHIP 6753 participants with optional MRI 3371; linked biopsy rate 6.9 pre to 9.9 post per 100 observation-years; large absolute increase in nonmalignant results, higher with disclosed MRI/lab findings.

Does not show: Concurrent lab disclosures, selection and regional pathology coverage constrain MRI-specific causation; increased biopsies do not quantify overdiagnosis or prove all MRI net harm. No randomized mortality contrast.

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:paragraph3, 1:Methods:paragraph4, 1:Methods:paragraph10, 1:Methods:paragraph23, 2:Results:paragraph1, 2:Results:paragraph7, 2:Results:paragraph14. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

[46] Psychosocial consequences and severity of disclosed incidental findings from whole-body MRI in a general population study.

Shows: 405 SHIP incidental-finding survey respondents; 28.6% moderate/severe distress after disclosure despite 96% satisfaction; subjective/radiological severity agreement low.

Does not show: Selected disclosed respondents, no random counterfactual and same SHIP parent; acute harm/satisfaction are not universal or long term net benefit metrics.

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.

[47] False-positive results released by direct-to-consumer genetic tests highlight the importance of clinical confirmation testing for appropriate patient care.

Shows: 49 clinical confirmation referrals from DTC raw data/third party interpretation; 40% reported variants false positive and some increased-risk labels clinically benign.

Does not show: Referral-enriched variant denominator; does not mean 40% of all DTC customers or tests are wrong. Genetic risk interpretation, not MCED diagnosis and not a clinical mortality 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: SC4.

[48] Reduced lung-cancer mortality with low-dose computed tomographic screening.

Shows: NLST 53454 high-risk smokers randomized 3 annual LDCT versus chest radiography; lung-cancer mortality 20% relative and all-cause 6.7% lower at initial follow-up, with many positive-screen false alarms.

Does not show: Positive disease-specific screening-mortality proof in defined risk/context, not Galleri; higher incidence and procedure harms must remain. Different modality cannot supply exact lead threshold.

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.

[49] Reduced Lung-Cancer Mortality with Volume CT Screening in a Randomized Trial.

Shows: NELSON randomized high-risk smokers 50–74 to volumetric CT/no screening; 10 year male lung death RR 0.76, 95% CI 0.61–0.94, women RR 0.67, 95% CI 0.38–1.14.

Does not show: True positive clinical screen effect in defined population; sex subgroup imprecise, different test/site/care pathway. No MCED mortality 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: SC2, SC3.

[50] Lung Cancer Incidence and Mortality with Extended Follow-up in the National Lung Screening Trial.

Shows: Same NLST extended-mortality median 12.3 years; lung death RR 0.92 CI 0.85–1.00, dilution adjusted 0.89; NNS 303, all cause interval includes null.

Does not show: Longer horizon alters estimates; repeated parent follow-up, not an independent trial or evidence for all cancer blood screens.

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.

[51] Ovarian cancer population screening and mortality after long-term follow-up in the UK Collaborative Trial of Ovarian Cancer Screening (UKCTOCS): a randomised controlled trial.

Shows: Original M/R UKCTOCS 202562 average-risk women 50–74, median 16.3 years; MMS increased stage I 47.2% and reduced stage IV 24.5%, but ovarian/tubal deaths not lower P=.58; USS P=.36.

Does not show: Direct stage/mortality dissociation blocks universal surrogate claim; does not prove Galleri mortality null because assay/site differ. Positive clinical mortality crossing absent for ovarian programme. The combined stage III/IV reduction (10.2%, 95% CI −21.3 to 2.4) was not significant, so the dissociation is stage IV-only.

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:paragraph4, 1:Methods:paragraph8, 1:Methods:paragraph11, 2:Results:paragraph3, 2:Results:paragraph11, 2:Results:paragraph16. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC3.

[52] Mortality impact, risks, and benefits of general population screening for ovarian cancer: the UKCTOCS randomised controlled trial.

Shows: UKCTOCS original comprehensive follow-up retains mortality null and stage changes; unnecessary false positive surgery 2.3 per MMS screen-detected case and 10 per USS-detected case.

Does not show: Same parent trial and denominators, not per-person risk; no independent RCT or direct Galleri harms rates. This is original trial report, not protocol.

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, SC4.

[53] Prostate-Specific Antigen Screening and 15-Year Prostate Cancer Mortality: A Secondary Analysis of the CAP Randomized Clinical Trial.

Shows: CAP 415357 men 50–69 cluster invitation PSA; 15 year secondary prostate-death RR 0.92 CI 0.85–0.99, absolute risk 0.69% vs 0.78%; low-grade/localized detection increased, all cause effect nonsignificant.

Does not show: Original 10 year primary null and 15 year secondary result both matter; different organ, tests and invitation not Galleri. Absolute effect small and diagnostic/treatment deaths reported.

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.

[54] Effect of Colonoscopy Screening on Risks of Colorectal Cancer and Related Death.

Shows: Nord ICC 84585 adults 55–64 randomized invitation colonoscopy/usual care; 42% uptake; 10 year incidence RR 0.82 positive, death RR 0.90 CI 0.64–1.16 null; 15 major bleeds.

Does not show: Invitation shared-care effect differs from procedure efficacy among compliers; incidence benefit does not by itself establish mortality benefit. No Galleri inference.

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

[55] Long-term effects of colonoscopy screening on colorectal cancer incidence and mortality: a multicountry, population-based randomised controlled trial.

Shows: Same Nord ICC 13 year follow-up incidence benefit and cancer death RR 0.88 CI 0.68–1.08; lower control mortality than expected may reduce power.

Does not show: No significant mortality contrast, per protocol nonrandom adherence sensitive; Erratum 42119586 attached. Avoid the uncorrected fine-grain subgroup CI, which the abstract prints with a negative upper bound.

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. Erratum In: PMID42119586. Subclaims: SC2, SC3.

[56] European Study of Prostate Cancer Screening - 23-Year Follow-up.

Shows: ERSPC 23 year follow-up core 162236 men 55–69 randomized PSA invitation; prostate-death RR 0.87 CI 0.80–0.95, absolute reduction 0.22%, incidence higher 1.30.

Does not show: True positive particular screening context but overdiagnosis remains, not Galleri or every cancer. Attached Erratum 41812212 needs detail check; not direct MCED mortality evidence.

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. Erratum In: PMID41812212. Subclaims: SC2, SC3.

[57] Effect of mammographic screening from age 40 years on breast cancer mortality (UK Age trial): final results of a randomised, controlled trial.

Shows: UK Age 160921 women randomized mammography 39–41 through 48 or usual age 50 screen; 10 year breast death RR 0.75 positive, later interval RR 0.98 null.

Does not show: Below registered Galleri age at baseline and different assay; timing and disease-specific boundary context only, not exact lead clinical crossing or proof all stage changes 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: SC2, SC3.

[58] The UK Lung Cancer Screening Trial: a pilot randomised controlled trial of low-dose computed tomography screening for the early detection of lung cancer.

Shows: UKLS pilot 4055 high risk adults 50–75 randomized CT/usual care; 42 detected cancers, 85.7% stage 1/2; temporary distress and cost model reported.

Does not show: Pilot feasibility with surrogate/cost model, not actual mortality comparison at this report; don't borrow NLST death benefit or this stage yield for Galleri.

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.

[59] Performance and safety of a multi-cancer early detection test: the PATHFINDER 2 study.

Shows: PATHFINDER 2 prospective single-arm study, 35,878 enrolled adults ≥50 without clinical suspicion of cancer; among 32,007 performance-analyzable participants, detection rate 0.54%, PPV 60.3%, episode sensitivity 39.3% (69.8% for 12 prespecified cancers), specificity 99.64%; 213 of 35,335 safety-analyzable participants (0.6%) had invasive procedures after a positive test.

Does not show: Descriptive endpoints without hypothesis testing and no comparator arm; no mortality endpoint. The authors state that randomized trials and longer follow-up are needed to assess clinical utility. Strengthens the SC1 and harms record; does not address SC2.

Access/status: Bounded original abstract and PubMed metadata examined 2026-11-02 during the adversarial correction; no complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC4.

[60] Cancer Stage Compared With Mortality as End Points in Randomized Clinical Trials of Cancer Screening: A Systematic Review and Meta-Analysis.

Shows: Systematic review and meta-analysis of 41 cancer-screening RCTs comparing stage III–IV incidence with cancer-specific mortality as end points; correlation between the two reductions varied by cancer type, high for ovarian and lung and weak for colorectal and prostate.

Does not show: Tier B synthesis: frames SC3 but does not carry an experimental number; does not include a Galleri or MCED mortality result and does not validate a multi-cancer stage surrogate.

Access/status: Bounded original abstract and PubMed metadata examined 2026-11-02 during the adversarial correction; no complete-paper or supplement audit claimed. none_found. Subclaims: SC3.

Documentary record

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

• ADVERSARIAL_REVIEW.mced-cancer-mortality.md
• CONSENSUS_CROSSCHECK.html
• CONSENSUS_CROSSCHECK.md
• RESEARCH_TIMELINE.html
• RESEARCH_TIMELINE.md
• SHA256.txt
• adversarial-findings.mced-cancer-mortality.json
• codex-correction-provenance.json
• consensus-validation.json
• edition-approval-audit.json
• mced-cancer-mortality.review.md
• mced-cancer-mortality.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: 60 · load-bearing claims: 65 · verified: 65 · unverified: 0 · contradicted: 0 · schema 5.3