In plain terms

The claim being assessed: Autoimmunity is initiated by infection-triggered molecular mimicry and bystander activation that overcome baseline genetic deficiencies in immune tolerance.

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

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

Strongest evidence for it: The preregistered weakest-link rule returns Β because the descriptive antecedence subclaim SC1 is associational. This is an aggregation limitation, not a finding that all infection causality is merely correlational: defined Campylobacter/ganglioside GBS convergence reaches Ζ, intrinsic tolerance necessity Η, and broader context-dependent initiation Ε. Primary-subclaim and scope-matched sensitivities return Ε. Human MS mimic-specific initiation remains Γ. Do not publish the overall letter without this subclaim explanation or interpret it as refuting EBV risk causality.

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

The preregistered weakest-link rule returns Β because the descriptive antecedence subclaim SC1 is associational. This is an aggregation limitation, not a finding that all infection causality is merely correlational: defined Campylobacter/ganglioside GBS convergence reaches Ζ, intrinsic tolerance necessity Η, and broader context-dependent initiation Ε. Primary-subclaim and scope-matched sensitivities return Ε. Human MS mimic-specific initiation remains Γ. Do not publish the overall letter without this subclaim explanation or interpret it as refuting EBV risk causality.

Implications for health and treatment: The preregistered weakest-link rule returns Β because the descriptive antecedence subclaim SC1 is associational. This is an aggregation limitation, not a finding that all infection causality is merely correlational: defined Campylobacter/ganglioside GBS convergence reaches Ζ, intrinsic tolerance necessity Η, and broader context-dependent initiation Ε. Primary-subclaim and scope-matched sensitivities return Ε. Human MS mimic-specific initiation remains Γ. Do not publish the overall letter without this subclaim explanation or interpret it as refuting EBV risk causality.

Limits for individual care: 70 primary reports with 6 selected partial-body examinations; density, heterogeneous controls and incomplete access set the confidence, not correctness. The grade applies to the exact registered claim; stronger findings for narrower subclaims remain separate. No complete-paper audit or blinded calibration is claimed. This educational review does not establish an individual treatment recommendation.

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

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

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

At a glance

Sub-claims assessed separately

Figures

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

Full assessment

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

Sources: 70 primary reports; claims: 77; verified: 77; unverified: 0. Verification is bounded attribution. Access: 6 partial-body and 64 complete abstract/status examinations. Literature cutoff: 2026-10-07; source examination UTC date: 2026-10-07. No complete-paper audit or blinded calibration claimed.

Adversarial review: Generated by Sol 6.1; adversarially reviewed by Claude Opus 5.5 (Anthropic) on 2026-10-14. 3 corrections applied (see ADVERSARIAL REVIEW AND APPROVAL). Status: Approved as revised.

AT A GLANCE

Infection can trigger specific autoimmune pathways; a single infection-first account does not explain autoimmunity across diseases.

Confidence basis: Moderate reflects evidence density and reproducibility, not reviewer access depth. The Campylobacter/ganglioside and inherited-tolerance literatures are replicated across independent laboratories and instrument classes; against that, much of the mimicry evidence for other diseases rests on induced or engineered systems (adjuvant immunization, ganglioside-deficient recipients, ectopic LMP1, transgenic and gnotobiotic hosts), the microbial results change sign with strain, sex, diet and housing, the abatacept cohort is nonrandomized, and recurrence of the same serorepositories and clinical isolate platforms across rows is only partly audited.

Deciding experiment: In baseline infection-negative, genotype- and tolerance-function-characterized cohorts, prevent a specific pathogen before seroconversion or selectively block a prespecified cross-reactive epitope while preserving antiviral immunity. Follow incident disease and matched antigen-specific clones, with independently replicated infection exposure and disease ascertainment; in parallel use physiological pathogen mimic-loss and restored-mimic rescue with functional pathology.

Belief versus evidence: The preregistered weakest-link rule returns Β because the descriptive antecedence subclaim SC1 is associational. This is an aggregation limitation, not a finding that all infection causality is merely correlational: defined Campylobacter/ganglioside GBS convergence reaches Ζ, intrinsic tolerance necessity Η, and broader context-dependent initiation Ε. Primary-subclaim and scope-matched sensitivities return Ε. Human MS mimic-specific initiation remains Γ. Do not publish the overall letter without this subclaim explanation or interpret it as refuting EBV risk causality.

Best-supported rival: Gene-environment threshold. The strongest discriminator between a conditional threshold model and tolerance-only primacy is Wen: germ-free MyD88-null NOD retain autoimmunity while commensal colonization protects. It establishes an interaction, not a universal microbial requirement. Human tolerance variants and GBS mimic manipulations address distinct questions.

COMPETING MECHANISTIC MODELS

Rank 1: Gene-environment threshold (Ε). Host regulatory state, tissue injury and defined microbial exposure jointly set disease-specific thresholds. Conditional infection effects, HLA presentation, innate sensing and opposite microbial signs support context dependence. A flexible integrated model must predict genotype/exposure outcomes prospectively and cannot define susceptible hosts after disease.

Rank 2: Intrinsic tolerance failure (Η). Central and peripheral regulatory pathways prevent pathogenic self-reactivity. Human FOXP3/CTLA4 natural experiments, conditional/acute depletion, reconstitution and genotype-aligned clinical therapy converge. Rare syndromic causality does not establish that all ordinary autoimmunity starts with a measurable intrinsic deficit; CTLA4 penetrance is incomplete.

Rank 3: Infection-triggered tolerance breakdown (Ε). Specific pathogens initiate selected autoimmune injury through mimicry or licensing. C. jejuni carbohydrate specificity and LOS perturbations provide the clearest route; EBV prospective data strongly constrain temporal alternatives. Defined GBS success does not establish one MS mimic or a common initiating infection across the class.

Rank 4: Bystander inflammatory licensing (Ε). Infection or tissue injury supplies antigen presentation and inflammatory help to pre-existing autoreactivity. Viral-sensor perturbations, EBV LMP1 and oligodendrocyte injury support separate licensing routes. Induced host systems and broad inflammatory changes do not identify a universal endogenous human initiating step.

Rank 5: Microbial-deprivation tolerance calibration (Ε). Some microbial exposures reduce rather than cause autoimmune injury. Germ-free/colonized MyD88-null NOD and early helminth-lupus experiments map protective effects. Protection is strain, genotype, age and housing dependent; infection deprivation is not a universal explanation.

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

Autoimmunity is initiated by infection-triggered molecular mimicry and bystander activation that overcome baseline genetic deficiencies in immune tolerance.

Use the index causal initiation claim and separate GBS from MS. Intrinsic tolerance evidence supports a rival and does not prove infection initiation. All registered components of the broad initiation proposition are required.

Selection: public-interest rank 27 from the editorial ranking compiled 2026-09-23; categorical prior verdicts were visible. Private fresh review; earlier documents are preserved.

Evidence and causal interpretation

Bjornevik 2022 prospectively places EBV seroconversion before MS and neurofilament rise with a CMV negative control. Finnish maternity and older late-infection observations add temporal coherence, while the small DOD peptidome analysis shares its repository. These constrain reverse causation strongly; the exact initiating immune mediator remains a different question (PMID35025605, PMID31226219, PMID38497939).

Campylobacter GBS combines antecedent clinical exposure, ganglioside-identical bacterial chemistry, genotype/epitope/clinical specificity and functional antibody injury. LOS sensitization produces rabbit neuropathy, while original Godschalk Results subtract sialylated mimics genetically. Its Methods bypass normal ganglioside tolerance and measure antibody induction; the rabbit phenotype and clinical evidence provide separate causal support. Defined GBS convergence is therefore stronger than an isolated biomarker experiment (PMID7477117, PMID8228822, PMID16162859, PMID15277677, PMID15578098).

Lanz original Results verify EBNA1/GlialCAM binding, phosphorylation dependence and exacerbation after deliberate PLP-induced EAE. Human antibody studies, ANO2 transfer and 2026 HLA-DR15/LMP1 work support several mechanistic routes. They do not demonstrate selective endogenous human mimic necessity before onset. Healthy-control T-cell cross-reactivity, inconsistent mimic signatures and late-lytic rather than purely EBNA1 responses remain real constraints (PMID35073561, PMID41534529, PMID41534530, PMID41534531, PMID38843296, PMID39742283, PMID42455903).

Tolerance necessity has different evidence: human FOXP3/CTLA4 variants, Treg-specific CTLA4 loss, adult acute Treg removal and timely regulatory-cell restoration. Genotype-aligned abatacept controls clinical manifestations in a nonrandomized cohort. These converge across natural and experimental instruments, but healthy CTLA4 carriers and background-specific PD1 phenotypes prevent unconditional sufficiency or class-wide genotype explanations (PMID11137993, PMID25329329, PMID25213377, PMID18845758, PMID17136045, PMID7636184, PMID37595759).

Younis original 2025 Results profile diagnosed SLE blood and integrate prior cell-line genomic datasets. EBV-positive patient B-cell antibodies bind nuclear antigens; chromatin-loaded transformed lines activate autologous helper and autoreactive B cells. This is important functional amplification evidence, with cell-line/loading controls, but it neither establishes prediagnostic initiation nor selective in vivo viral necessity (PMID41223250).

T1D exposure results are mixed and stage-dependent. TEDDY links prolonged enterovirus-B shedding to autoimmunity rather than clinical diabetes; a small independent virome cohort does not associate common enteroviruses with onset. Receptor signaling can alter the sign of the same viral challenge in NOD. The formula randomization in TRIGR did not randomize viral exposure (PMID31792456, PMID28696303, PMID34804036, PMID25422874, PMID12699416).

DiViD is a positive randomized test of residual insulin secretion after diagnosis. Original Methods prespecify repeated-measures C-peptide analysis, while Results retain null insulin-dose and late HbA1c contrasts. All participants were peripheral-PCR negative; two antiviral drugs have inseparable and potentially immunomodulatory effects. CVB vaccine phase I establishes safety/immunogenicity, not reduced autoimmune incidence. Neither result proves an initiating infection (PMID37789144, PMID38369573).

Microbial sign depends on context. Original Wen Results show diabetes in germ-free MyD88-null NOD, protection after defined colonization and retained diabetogenic cells. SFB drives K/BxN arthritis but associates with female NOD protection; early helminth exposure suppresses lupus. An HBV-transgenic pristane study cannot establish acquired infection or human protection despite its abstract wording (PMID18806780, PMID20620945, PMID21709219, PMID31954872, PMID27125291).

RA/lupus pathobiont experiments provide additional bounded routes: E. gallinarum-directed vaccination in predisposed mice, L. reuteri/diet effects, arthritogenic versus nonarthritogenic Subdoligranulum isolates and Aa toxin-induced citrullination. One Aa endocarditis case resolves after eradication, while ordinary RA microbiome normalization after immune treatment permits reverse causation. These do not establish one obligatory class-wide pathogen (PMID29590047, PMID30581114, PMID36288282, PMID27974664, PMID30459755, PMID26214836).

The formal overall Β follows the registered minimum rule including the descriptive antecedence row. It must be read with SC2a Ζ, SC3 Η and primary/scope-matched Ε sensitivities. This artifact is explicitly flagged for editorial interpretation; it is not a scientific claim that EBV or Campylobacter evidence is only correlational. The proposed next experiment targets human onset mediation rather than repeating marker discovery.

Consensus supplied useful gap and timeline leads and narrow indexed passage checks for Lanz Methods, Wen microbial protection and Younis established-disease sampling. Original sections confirm those limits. It substituted a Godschalk thesis and missed Krogvold, while correcting an unsupported full-penetrance assertion. Direct public publisher text recovered Godschalk after repository failure; its exact immune-deficient/adjuvant design supplies the original check. Provider meters and counts are not evidence grades.

HYPOTHESIS STATE EVALUATION FRAMEWORK (HSEF v5.3)

Subclaim matrix and aggregation

Preregistered aggregation: weakest-link. The preregistered weakest-link rule returns Β because the descriptive antecedence subclaim SC1 is associational. This is an aggregation limitation, not a finding that all infection causality is merely correlational: defined Campylobacter/ganglioside GBS convergence reaches Ζ, intrinsic tolerance necessity Η, and broader context-dependent initiation Ε. Primary-subclaim and scope-matched sensitivities return Ε. Human MS mimic-specific initiation remains Γ. Do not publish the overall letter without this subclaim explanation or interpret it as refuting EBV risk causality. Sensitivities: weakest-link Β; primary-subclaim Ε; scope-matched Ε.

Decisive-perturbation inventory and causal tally

[2] Patient-derived antibody cross-binds EBNA1 and GlialCAM; peptide immunization aggravates PLP-induced EAE. Selected Results show induced EAE amplification rather than spontaneous MS initiation or selective human mimic necessity.

[5] GlialCAM/EBNA1 reciprocal blocking confirms cross-reactivity and HLA-associated risk patterns. Established-case serology and overlapping mimic cohorts cannot establish onset mediation.

[6] EBNA1/CRYAB reciprocal blocking and mouse responses support cross-reactive immunity. Risk serology and immunization do not isolate CRYAB as a necessary initiator of MS.

[7] ANO2 pre-immunization and transferred CD4 cells exacerbate EAE; human clones cross-recognize EBNA1. Induced EAE enhancement is a phenotype but does not establish spontaneous human onset necessity.

[10] ANO2/EBNA1 blocking and genotype combinations link cross-reactivity to MS risk. Case-control antibody positivity occurs in a minority and does not demonstrate individual causal initiation.

[14] EBV reprograms HLA-DR15 B-cell presentation of MBP peptides recognized by MS T cells. Human tissue alignment and infected-cell experiments support a route distinct from proving onset through molecular mimicry.

[17] LMP1 or CD40L overrides death of antigen-capturing myelin-reactive B cells and enables lesions. Engineered viral-protein signaling tests a bounded route; ectopic expression cannot establish endogenous human primacy alone.

[18] Open-label EBV-specific T-cell treatment is feasible and clinical improvement tracks product reactivity. Small uncontrolled progressive-MS phase I trial does not prove efficacy or prevention; linked erratum details unexamined.

[23] Serial conduction studies reclassify transient Campylobacter-related demyelinating patterns as axonal neuropathy. Clinical classification depends on timing; this limits broad demyelination transfer but does not refute axonal mimicry.

[24] GBS-strain LPS immunization induces GM1-cross-reactive rabbit antibodies. Antibody induction alone is not demonstrated neuropathic functional injury.

[25] Cloned glycosyltransferases synthesize defined ganglioside-like structures with linkage confirmation. Biochemical biosynthesis establishes the molecular mimic rather than human disease necessity.

[26] GBS/MFS LPS induces matched antibody specificity; control-strain LPS lacks ganglioside reactivity. Immunization establishes antigen specificity without showing spontaneous infection produces neuropathy in every host.

[28] NMR and chemical analysis identify a GM1-identical terminal bacterial oligosaccharide. Structural identity is not alone a causal neurological experiment.

[29] Anti-GD1a antibody plus complement injures high-antigen motor terminals and blocks function. Tolerance bypass and ganglioside-overexpressing recipients are engineered; normal mice resist the same exposure.

[30] LOS-sensitized rabbits develop anti-GM1 antibodies, weakness and GBS-like peripheral pathology. Patient/mouse antibody blocks coculture potentials but does not itself induce paralysis; selected full body unavailable.

[31] cst-II/orf10 loss removes sialylated mimics and anti-GD1a induction while orf11 is a negative control. Original publisher Methods use ganglioside-deficient mice and repeated adjuvant immunization; asialo-GM1 responses remain and neuropathy was not measured.

[32] Some neuropathy-associated isolates lack detectable ganglioside mimics or have frameshifted biosynthesis genes. Post-isolation phase variation may differ from infection-time expression; these are boundaries rather than definitive mimic absence at onset.

[33] Foxp3-lineage neuritin loss causes abnormal germinal-center/plasma-cell autoimmunity with reconstitution effects. Rescue primarily changes cellular/antibody endpoints and does not establish infection initiation.

[34] Treg-specific CTLA4 loss causes spontaneous fatal autoimmune disease. Conditional lineage specificity supports tolerance function; developmental/immune interactions do not establish a universal trigger.

[35] Independent human CTLA4 variants impair regulatory function and produce immune dysregulation with incomplete penetrance. Unaffected carriers invalidate complete-penetrance claims; no measured infectious mimic is required by the proposed genotype mechanism.

[36] Abatacept controls manifestations in a genotype-defined LRBA/CTLA4 cohort. Nonrandomized treatment allocation and disease-duration differences limit comparative superiority estimates.

[37] Engineered dominant AIRE mutations disrupt central tolerance and alter chromatin/transcription. Mutant-line developmental biology and expression changes do not prove ordinary sporadic autoimmunity is AIRE-deficient.

[38] Aire-deficient thymic stroma loses tissue-antigen transcription and supports defined autoimmunity. Constitutive loss has developmental confounds; thymic localization narrows but does not remove them.

[39] Human FOXP3 mutations segregate with IPEX immune dysregulation. Rare syndromic causality is not a class-wide requirement for pathogenic FOXP3 variants.

[40] Independent IPEX sequencing finds forkhead-domain FOXP3 variants matching scurfy biology. Genetic identification alone does not measure a specific infection or explain all polygenic disease.

[41] CD25-cell removal permits autoimmune disease and timely reconstitution prevents it. Athymic transfer and imperfect CD25 lineage specificity constrain physiology; adult conditional depletion is an orthogonal follow-up.

[42] Acute neonatal/adult Foxp3-Treg ablation causes autoimmunity without lifelong gene-loss dependence. Mouse induced depletion establishes ongoing regulatory necessity, not that human infection always removes Tregs.

[43] Human CTLA4 haploinsufficiency causes cellular dysregulation and wild-type complementation corrects ex vivo proliferation. Original text preserves healthy carriers and culture rescue; common disease initiation is a separate scope.

[44] PD1 loss produces background-dependent lupus-like autoimmunity accelerated by Fas loss. Constitutive immune development and genetic background limit generalization.

[45] BALB/c PD1 loss causes autoimmune cardiomyopathy while RAG2 deficiency prevents it. B/T-cell dependency and strain context do not identify an infectious initiator.

[46] Inherited human PD1 deficiency impairs antimycobacterial immunity and causes immune dysregulation. Single affected child with TB cannot independently determine whether infection or tolerance loss initiated each lesion.

[48] Established-SLE EBV-positive B cells encode nuclear autoreactivity and transformed chromatin-loaded cells activate autologous helper/B cells. Original Results use diagnosed patients and cell-line genomic datasets; no prediagnostic onset sequence or selective in vivo viral depletion is established.

[49] Oligodendrocyte ablation precedes delayed adaptive demyelination; transfer and MOG tolerization change disease. Engineered tissue injury establishes an alternative route in a model, not proof that human MS routinely begins without EBV.

[50] Sensing the same diabetogenic virus differently through MDA5/TLR3 changes diabetes risk and can protect. Sensor, local interferon and host state determine sign; this is not a universal infection-to-disease rule.

[53] TLR3 knockout protects female NOD from CB4 diabetes and reduces insulitis versus matched uninfected controls. Constitutive receptor loss and forced viral challenge establish model-specific causal contribution.

[54] Affinity-purified recent-onset T1D antibodies cross-bind viral/islet antigens and induce beta-cell apoptosis. Culture killing does not establish the infection-to-human-onset chain; SC2b comparison illustrates mimicry transfer limits, not MS evidence.

[56] Randomized new-onset antiviral combination preserves stimulated C-peptide on the primary repeated-measures analysis. Post-diagnosis functional secretion is not prevention; peripheral PCR is negative, drug effects cannot be separated, and insulin-dose/12-month HbA1c contrasts are null.

[60] Germ-free MyD88-null NOD regain diabetes and defined commensal colonization attenuates it. Original Results show protective microbiota and retained diabetogenic cells, with genotype/housing limits and no human prevention demonstration.

[61] E. gallinarum translocation induces autoimmunity in predisposed mice and pathogen-directed vaccine prevents mortality. Human liver DNA is associative; antibiotic pleiotropy and induced colonization do not establish spontaneous human initiation; correction attached.

[62] L. reuteri colonization worsens TLR7-linked lupus and resistant starch lowers translocation and disease. Strain, diet and transgenic host set context; human fecal enrichment is not onset mediation.

[63] Aa leukotoxin causes neutrophil hypercitrullination matching RA antigen patterns. Human exposure/ACPA association and ex vivo toxin action do not establish necessary RA onset.

[64] Aa endocarditis eradication coincides with sustained arthritis and ACPA resolution in one genetically at-risk patient. One RA-like infection-associated case lacks randomized withdrawal/rechallenge and cannot establish ordinary RA primacy.

[65] One Subdoligranulum isolate causes arthritic autoimmunity in germ-free mice and lymphocyte depletion eliminates disease. At-risk human antibody selection plus germ-free strain challenge proves a bounded route rather than all RA initiation.

[66] SFB colonization restores Th17/autoantibodies and arthritis in germ-free K/BxN; IL17 neutralization prevents arthritis. Transgenic host and immune background set context; one commensal can have the opposite sign in NOD diabetes.

[67] Naturally acquired SFB segregates with diabetes protection in female NOD despite persistent insulitis. Colonization is naturally incomplete and observational within the facility; sex dependence limits mechanistic certainty.

[68] Early H. microstoma infection suppresses spontaneous lupus manifestations in NZBWF1. Parasitic immune pleiotropy and unsuccessful nematode settlement limit target specificity and human transfer.

[69] HBV-transgenic mice have reduced pristane-lupus renal and antibody pathology. The actual manipulation is a transgene, not acquired HBV infection; the abstract's human-protection conclusion is unsupported by this experiment.

Tally: 47 discriminating contrasts. Null and negative contrasts are discriminators, not positive threshold crossings. Reports and replications are not independent causal lines. Shared cohorts, experimental platforms, transgenic lineages and repeated outcomes do not constitute fully independent causal lines.

Cap audit, precedence and Β/Γ/Δ boundary

The immunology threshold credits lineage-specific depletion, target-aligned rescue and hard disease outcomes in their tested settings. In vitro cross-reactivity and biomarker-only mutant antibody induction cannot alone cross the human onset boundary. GBS convergence combines clinical strain/phenotype specificity with molecular subtraction and organismal weakness, and is not re-capped because its human clinical leg is observational. The universal forced-transfer/overexpression limits apply to extending induced EAE, LMP1 and gnotobiotic models to endogenous human initiation. Familial tolerance causality remains separate from sporadic infection claims.

Gamma hard-rule audit: Ex vivo functional assays alone — applicable to human cross-reactive clones and SLE APC cultures. Cytokine profiles plus in vitro work — insufficient for endogenous onset. Broad immunosuppression — does not identify an infectious initiator. Correlational human data plus in vitro mechanism — caps the human onset link when selective disease-relevant dependency is missing; independent GBS and tolerance causal sets are assessed separately.

SC1 cap: “Pure association evidence — however large, however replicated — cannot exceed Β.” Prediagnostic EBV seroconversion and C. jejuni antecedence are reproducible with temporal and exposure controls. This registered descriptive existence claim has no selective causal manipulation of its own and is graded Β. Its formal minimum must not re-cap the independent mechanistic evidence assessed separately.

SC2b cap: “Mechanistic plausibility plus ex vivo or in vitro immune functional support caps at Γ.” EBNA1/GlialCAM, ANO2 and CRYAB cross-reactivity has human and functional support, but original Lanz Results show enhancement of induced EAE. ANO2 transfer and 2026 B-cell presentation/viral-signaling studies add model support without selective endogenous human mimic necessity at onset. The registered initiation link remains Γ.

SC1 field-valid definition: “ - Cell-type-specific conditional knockout or depletion - Pathway-specific blockade (not broad immunosuppression) changing disease course - Human targeted therapy with mechanistic alignment showing disease-modifying effect” Positive full-scope result: no. Prediagnostic EBV seroconversion and C. jejuni antecedence are reproducible with temporal and exposure controls. This registered descriptive existence claim has no selective causal manipulation of its own and is graded Β. Its formal minimum must not re-cap the independent mechanistic evidence assessed separately.

SC2a field-valid definition: “ - Cell-type-specific conditional knockout or depletion - Pathway-specific blockade (not broad immunosuppression) changing disease course - Human targeted therapy with mechanistic alignment showing disease-modifying effect” Positive full-scope result: yes. Within defined bacterial/ganglioside genotypes, chemical identity, bacterial mimic subtraction, matched antibody specificity, rabbit neuropathy and clinical isolate/phenotype contrasts converge across errors. Ζ credits the causal set, not the mutant antibody biomarker alone. Host resistance, non-mimicking isolates and absent human selective prevention limit Η and broader GBS sufficiency.

SC2b field-valid definition: “ - Cell-type-specific conditional knockout or depletion - Pathway-specific blockade (not broad immunosuppression) changing disease course - Human targeted therapy with mechanistic alignment showing disease-modifying effect” Positive full-scope result: no. EBNA1/GlialCAM, ANO2 and CRYAB cross-reactivity has human and functional support, but original Lanz Results show enhancement of induced EAE. ANO2 transfer and 2026 B-cell presentation/viral-signaling studies add model support without selective endogenous human mimic necessity at onset. The registered initiation link remains Γ.

SC3 field-valid definition: “ - Cell-type-specific conditional knockout or depletion - Pathway-specific blockade (not broad immunosuppression) changing disease course - Human targeted therapy with mechanistic alignment showing disease-modifying effect” Positive full-scope result: yes. Human FOXP3/CTLA4 inborn errors, lineage-specific CTLA4 loss, acute adult Treg removal and time-dependent reconstitution establish ongoing tolerance causality. Genotype-aligned abatacept changes clinical trajectories. Different failure structures support Η; incomplete penetrance, developmental effects and multiple tolerance routes prevent Θ/Ω for all pathways.

SC4 field-valid definition: “ - Cell-type-specific conditional knockout or depletion - Pathway-specific blockade (not broad immunosuppression) changing disease course - Human targeted therapy with mechanistic alignment showing disease-modifying effect” Positive full-scope result: yes. A coherent infection-to-autoimmune injury route is established in defined C. jejuni/ganglioside settings, with rare human Aa eradication evidence and model perturbations in other diseases. EBV antecedence strongly supports an upstream MS risk role without identifying the initiating mimic. Prediagnostic baseline tolerance breakdown and onset mediation across human diseases are unresolved, so context-specific Ε is retained.

SC5 field-valid definition: “ - Cell-type-specific conditional knockout or depletion - Pathway-specific blockade (not broad immunosuppression) changing disease course - Human targeted therapy with mechanistic alignment showing disease-modifying effect” Positive full-scope result: yes. Causal microbial routes have positive and protective signs across genotypes and diseases. SFB promotes K/BxN arthritis but associates with female NOD protection; germ-free MyD88-null NOD develop diabetes. Thus bounded causality exists but the class-wide initiation generalization fails: Ε, not a denial that particular infections are causes.

Upward threshold ladder

SC1 Δ: “Requires pathway- or cell-type-resolved causal perturbation affecting disease or phenotype.” Satisfied: no. For this descriptive antecedence row, a causal restatement and prospective selective infection intervention would be required before grading it as causal. Any future aggregation amendment must be prospectively declared; this review preserves the original rule.

SC2a Δ: “Requires pathway- or cell-type-resolved causal perturbation affecting disease or phenotype.” Satisfied: yes.

SC2a Ε: “Requires context-specific causality with explicit scope restriction.” Satisfied: yes.

SC2a Ζ: “Requires multiple convergent causal manipulations and reduced plausibility of downstream-only interpretations.” Satisfied: yes.

SC2a Η: “Requires robust upstream causal evidence across systems and interventions.” Satisfied: no. Physiological infection-time mimic subtraction/restoration with neuropathic outcomes and prospective human onset prediction, rather than adjuvant immunization or antibody induction alone.

SC2b Δ: “Requires pathway- or cell-type-resolved causal perturbation affecting disease or phenotype.” Satisfied: no. Suppress the defined cross-reactive epitope or clone before onset while preserving antiviral control; demonstrate less disease and matched rescue.

SC3 Δ: “Requires pathway- or cell-type-resolved causal perturbation affecting disease or phenotype.” Satisfied: yes.

SC3 Ε: “Requires context-specific causality with explicit scope restriction.” Satisfied: yes.

SC3 Ζ: “Requires multiple convergent causal manipulations and reduced plausibility of downstream-only interpretations.” Satisfied: yes.

SC3 Η: “Requires robust upstream causal evidence across systems and interventions.” Satisfied: yes.

SC3 Θ: “Requires predictive consistency and minimal contradiction in the claimed disease scope.” Satisfied: no. Prospective genotype/functional thresholds accurately predict tolerance failure and therapeutic rescue across populations while retaining unaffected carriers.

SC4 Δ: “Requires pathway- or cell-type-resolved causal perturbation affecting disease or phenotype.” Satisfied: yes.

SC4 Ε: “Requires context-specific causality with explicit scope restriction.” Satisfied: yes.

SC4 Ζ: “Requires multiple convergent causal manipulations and reduced plausibility of downstream-only interpretations.” Satisfied: no. Selective infection prevention or clearance before disease in baseline-tolerance-characterized people, with incident autoimmune disease and mediator-specific readouts.

SC5 Δ: “Requires pathway- or cell-type-resolved causal perturbation affecting disease or phenotype.” Satisfied: yes.

SC5 Ε: “Requires context-specific causality with explicit scope restriction.” Satisfied: yes.

SC5 Ζ: “Requires multiple convergent causal manipulations and reduced plausibility of downstream-only interpretations.” Satisfied: no. Independent prespecified disease cohorts and interventions establish which pathogen, mechanism and host state explain onset; preserve negative and protective contexts.

Upper ladder: Inborn tolerance errors plus acute conditional manipulation and aligned therapeutic trajectory change make Η reachable for SC3. GBS exposure/strain differences plus molecular manipulation and neuropathy support Ζ in defined scope. EBV seroconversion is a strong temporal causal argument, and no RCT is demanded in principle. The nominal class-wide infection initiation link lacks selective pre-onset mediation; T1D C-peptide preservation, vaccine immunogenicity and uncontrolled progressive-MS cell therapy do not fill that gap.

Independence: 70 primary reports are not 70 causal lines. DOD, TEDDY, Swedish mimic cohorts and clinical Campylobacter isolate platforms recur. GBS, intrinsic tolerance and general infection initiation are different claims: evidence cannot be pooled across them to manufacture convergence for a universal initiator. The GBS Ζ and tolerance Η sets exclude different clinical, molecular and developmental errors; exact cohort overlap remains only partly audited.

Clinical pathogen-genotype/phenotype alignment [19, 27, 20]: Tests actual infection-linked clinical patterns independently of induced model adjuvants, while retaining ascertainment confounding.

Bacterial mimic subtraction and structural chemistry [31, 25, 28]: Changes and verifies the molecular epitope, excluding mere coincidental serology; mutant immunization alone measures antibodies.

Organismal neuropathic outcome [30, 29]: Links immune specificity to weakness and nerve injury, with species/antigen-density and immunization confounds distinct from patient ascertainment.

Human tolerance natural experiments [39, 40, 35, 43]: Inherited regulatory lesions exclude infection association as the sole source of self-reactivity; penetrance and rare-genotype scope remain.

Acute lineage depletion and reconstitution [42, 34, 41]: Adult loss and timed rescue narrow lifelong developmental or familial co-variant explanations.

Genotype-aligned clinical therapy [36]: CTLA4-Ig trajectory effects support regulator function through a different instrument, with nonrandomized selection retained.

Falsification and pitfalls

Microbes need not initiate autoimmune disease and can protect. Germ-free MyD88-null NOD still develop diabetes; defined colonization attenuates it. Acute tolerance loss and sterile oligodendrocyte injury provide other model routes. Additional downgrade: False.

Cross-reactivity does not uniquely identify affected individuals or onset. Healthy-control EBV T-cell autoreactivity, partial mimic seropositivity and discordant GlialCAM/CRYAB cohort signatures remain in scope. Additional downgrade: False.

EBV upstream risk substituted for a proved initiating epitope. SC1 versus SC2b/SC4; original induced-EAE design.

Mutant antibody induction treated as clinical disease prevention. Godschalk deficient recipient and adjuvant Methods.

Genetic tolerance necessity treated as proof of infectious initiation. SC3 is rival support.

Antiviral benefit after diagnosis treated as prevention. DiViD timing, PCR-negative cohort and two-drug specificity.

HBV transgene called acquired infection or human protection. PMID27125291 actual experiment.

Formal weakest-link letter obscures stronger causal subclaims. SC1 descriptive Β controls overall; primary and scope-matched Ε sensitivities must accompany publication.

Provider retrieval counts or substituted thesis treated as primary validation. Consensus Godschalk/Krogvold failed matches and penetrance correction.

Adversarial steelman: above, then below

The strongest case for Γ or above for the assembled scientific hypothesis combines GBS causal convergence, EBV temporal specificity and model infection/sensor perturbations. It is scientifically compelling for bounded contexts. Under the retained weakest-link protocol, it cannot raise the descriptive SC1 above Β merely by relabelling associations. Primary-gate Ε is a legitimate sensitivity, and the protocol limitation must remain visible rather than implying the entire causal set is associational.

An Α narrative-only assessment is untenable: prospective antecedence, defined molecular mimics, specific subtraction, neuropathic outcomes and tolerance loss/rescue are actual primary evidence. The hypothesis merits formal causal adjudication even though its broad initiating mechanism is unresolved.

Preserve the preregistered formal Β while exposing the aggregation artifact and stronger disease-specific causal conclusions. Neither universal infection primacy nor infection-as-passive-marker explains the examined evidence. A future review should prospectively treat the descriptive existence row separately from causal conjunction grading.

Final score, causal direction and M/S/D

The preregistered weakest-link rule returns Β because the descriptive antecedence subclaim SC1 is associational. This is an aggregation limitation, not a finding that all infection causality is merely correlational: defined Campylobacter/ganglioside GBS convergence reaches Ζ, intrinsic tolerance necessity Η, and broader context-dependent initiation Ε. Primary-subclaim and scope-matched sensitivities return Ε. Human MS mimic-specific initiation remains Γ. Do not publish the overall letter without this subclaim explanation or interpret it as refuting EBV risk causality.

For the broad class, infections can amplify, license or maintain self-reactivity; evidence does not identify one required and rate-limiting initiating mechanism. C. jejuni mimicry can initiate a defined axonal neuropathy and EBV is an upstream MS risk factor, so this class-level role is not imposed on every disease-specific route. Disease in germ-free settings does not by itself mean microbes are downstream consequences.

Momentum: Mechanistic only. 2025 SLE EBV reprogramming and 2026 ANO2/HLA-DR15/LMP1 studies examine specific routes. The examined CVB vaccine phase I tests safety/immunogenicity and DiViD treats established disease; no verified ongoing incidence-result programme is claimed here.

Stability: Stable. No named verified pending result is asserted likely to move the full class-wide initiation score imminently. A selective pre-onset pathogen/epitope intervention with incident disease would change SC4/SC2b.

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

PROPOSED DECISIVE EXPERIMENTS

In baseline infection-negative, genotype- and tolerance-function-characterized cohorts, prevent a specific pathogen before seroconversion or selectively block a prespecified cross-reactive epitope while preserving antiviral immunity. Follow incident disease and matched antigen-specific clones, with independently replicated infection exposure and disease ascertainment; in parallel use physiological pathogen mimic-loss and restored-mimic rescue with functional pathology.

Readout: Incident adjudicated autoimmune disease, onset time, organ function, pathogen acquisition/clearance, antigen-specific autoreactivity and baseline-to-onset tolerance function. Controls: Genotype, age, prior infection, vaccination, baseline autoantibodies/tolerance, tissue injury, clinical stage, target engagement and antiviral competence; pathogen variants with matched fitness. Supporting: Preventing or subtracting the defined infectious mediator reduces incident disease and restored mimic rescues the matched phenotype in the predicted baseline context. Falsifying: Verified infection or mimic suppression leaves onset unchanged, or genotype/tissue injury explains disease without the proposed infectious route.

RESEARCH TIMELINE

KNOWLEDGE GAPS

Separate EBV exposure causality from which mimic, viral program or B-cell route mediates human MS onset.

Measure baseline tolerance before disease rather than infer a genetic deficiency from post-onset inflammation.

Test physiological infection-time mimic subtraction with neurological outcomes, retaining normal-host resistance and bacterial fitness controls.

Replicate stage-specific antiviral effects and link benefit to verified tissue viral engagement; new-onset preservation is distinct from prevention.

Map protective and harmful microbial effects by baseline genotype, sex, diet and disease; do not treat all commensals as pathogens.

Reconcile recurring cohort and isolate-platform overlap before claiming independent convergence.

Before public comparison, state the weakest-link sensitivity created by a descriptive existence subclaim; prospectively improve future decomposition rather than silently adjust this review.

METHODS AND CONSENSUS CONTRIBUTION

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

REVIEWER NOTES

70 primary reports examined, including six partial original-body examinations. Formal Β is highly aggregation-sensitive because SC1 is descriptive; primary/scope-matched Ε is reported. Original Godschalk publisher text recovered after failed PMC/Europe PMC access. No complete-paper/supplement audit; linked erratum details remain unexamined. Consensus two Pro browser queries, zero API/Deep. Positive controls carried forward, not fresh blinded calibration.

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

ADVERSARIAL REVIEW AND APPROVAL

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

Scope of checks. PMIDs verified 70/70 against PubMed get_article_metadata: first author, title, journal and DOI matched for every reference; 0 unresolvable, 0 wrong-record substitutions. Retraction / expression-of-concern screen run on all 70 records: none found. The three linked corrections recorded in the ledger were confirmed to exist and to attach to the right parents — PMID 33055421 (Published Erratum, to Pender [18]), PMID 30455315 (Correction for Zhao et al., to [52]) and PMID 29724927 (Erratum for the Report, to Manfredo Vieira [61]). 12 load-bearing claims spot-checked against primary abstracts or the cited original passages (EBV seroconversion preceding MS with a CMV negative control; the Finnish maternity cohort; Lanz EBNA1/GlialCAM with PLP-induced EAE aggravation rather than spontaneous onset; Yuki 2004 LOS rabbit sensitization; Godschalk cst-II/orf10 subtraction with orf11 as control; Goodfellow anti-GD1a complement injury in ganglioside-overexpressing terminals; Wen germ-free MyD88-null NOD diabetes with protective colonization; Wu SFB-driven K/BxN arthritis versus Kriegel SFB-associated female NOD protection; Krogvold DiViD repeated-measures C-peptide primary with null insulin-dose and HbA1c contrasts; Hyöty PRV-101 phase I safety/immunogenicity only; Liu 2016 HBV transgene rather than acquired infection; Younis 2025 diagnosed-SLE sampling with cell-line genomic datasets) — all matched, including the review's corrections of two source abstracts' own overreach. Scoring logic re-derived against hsef-hypothesis-review-v5/SKILL.md and references/hsef-framework.md. Internal consistency checked across review.md, scorecard.json and validation-report.json (70 ledger rows, 77 claims, 6 partial-body + 64 abstract/status = 70).

Scoring logic — checked and upheld. Caps are applied per subclaim before thresholds, and the cap quotations match the framework and rubric text. The declared weakest-link rule is applied as written: SC1 is Β under universal cap 2 (pure association cannot exceed Β) and therefore sets the overall letter, with the mandatory alternatives reported and arithmetically correct (primary-subclaim Ε, scope-matched Ε). The cap-precedence rule is applied correctly in both directions — the GBS causal set is not re-capped because its clinical leg is observational, and the universal forced-transfer and overexpression caps do bind the extension of induced EAE, ectopic LMP1 and gnotobiotic results to endogenous human initiation. The Ζ on SC2a and the Η on SC3 each carry a written independence basis naming the distinct systematic errors the lines exclude. Momentum "Mechanistic only" is correct and was not inflated: the named artifacts are completed publications and a completed phase I, which under lead policy can never satisfy "Decisive work underway". Divergence Contested is permitted without implied-level arithmetic where the field is genuinely split, which the opposite-sign microbial literature establishes here.

Rejected findings (checked, no change). 1. The overall Β. This is the most arguable call in the review and it was argued against before being upheld. SC1 is a descriptive antecedence row rather than a causal conjunct, so a weakest-link conjunction that includes it drives the headline below the weakest causal subclaim — the review itself names this an aggregation artifact. Three things make retention correct rather than convenient: the rule was pre-registered before scoring, SKILL.md Step 3 exists precisely to prevent rule-shopping after the results are visible, and both alternatives are published (Ε/Ε) in the banner, the matrix, the steelman and the final-score section. Changing the letter here would be the error the step was written to prevent. The review's own instruction not to publish the letter without the subclaim explanation is retained and reinforced. 2. AT A GLANCE headline reading stronger than Β. Checked: the sentence reports what the matrix contains (bounded infection causality plus failed class-wide generalization) and the adjacent "Belief versus evidence" paragraph states the aggregation basis. Accurate, not inflated. No change. 3. "Best-supported rival" in AT A GLANCE naming the gene-environment threshold model. Score card confirms rank_in_tournament: 1 with is_nominal: false; the review's own nominal (M1, infection-triggered tolerance breakdown) ranks 3. The renderer defect seen elsewhere did not occur here. 4. Six references whose print year differs by one from the PubMed e-pub year (Kim 2007/2006; Kamitani 2025/2024; Traka 2016/2015; McCall 2015/2014; Lindfors 2020/2019; Zegarra-Ruiz 2019/2018). Standing convention. 5. Three apparent title mismatches against PubMed — "CD4+ T cells" and "CD8+ T cells" rendered by PubMed as "CD4T" and "CD8T", and the dropped italicised species name in "Due to Aggregatibacter actinomycetemcomitans and Its Resolution". The review's text is correct and PubMed's field is the lossy one. Standing convention. 6. v5.claim_form "phenomenon" warning. Standing convention: the registered claim states initiation plus the tolerance-breakdown consequence. 7. REVIEWER NOTES vs the spliced Consensus section. No contradiction: the notes record two Consensus Pro browser queries and zero API/Deep calls, consistent with the Consensus section and with the Methods statement. 8. Collective authors. No reference in this list has a consortium first author; no leading-comma defect present.

Residual limitations. Access depth is unchanged by this pass: six partial-body and 64 abstract/status examinations, no complete-paper or supplement audit. The three linked errata remain unexamined in substance. The figures attributed to partial-body readings of Godschalk, Lanz, Wen, Krogvold and Younis could not be re-verified beyond their abstracts and are not contradicted by any record examined. Cohort and isolate-platform overlap remains only partly audited, as the review states.

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

REFERENCES

1. Bjornevik K, Cortese M, Healy BC et al. Longitudinal analysis reveals high prevalence of Epstein-Barr virus associated with multiple sclerosis. Science (New York, N.Y.). 2022. PMID 35025605. DOI: 10.1126/science.abj8222. Access: abstract_metadata. Study: DOD-serorepository-EBV-MS.

2. Lanz TV, Brewer RC, Ho PP et al. Clonally expanded B cells in multiple sclerosis bind EBV EBNA1 and GlialCAM. Nature. 2022. PMID 35073561. DOI: 10.1038/s41586-022-04432-7. Access: full_text_partial. Study: PMID35073561.

3. Cortese M, Leng Y, Bjornevik K et al. Serologic Response to the Epstein-Barr Virus Peptidome and the Risk for Multiple Sclerosis. JAMA neurology. 2024. PMID 38497939. DOI: 10.1001/jamaneurol.2024.0272. Access: abstract_metadata. Study: DOD-serorepository-EBV-MS-peptidome.

4. Wang J, Jelcic I, Mühlenbruch L et al. HLA-DR15 Molecules Jointly Shape an Autoreactive T Cell Repertoire in Multiple Sclerosis. Cell. 2020. PMID 33091337. DOI: 10.1016/j.cell.2020.09.054. Access: abstract_metadata. Study: PMID33091337.

5. Sattarnezhad N, Kockum I, Thomas OG et al. Antibody reactivity against EBNA1 and GlialCAM differentiates multiple sclerosis patients from healthy controls. Proceedings of the National Academy of Sciences of the United States of America. 2025. PMID 40063790. DOI: 10.1073/pnas.2424986122. Access: abstract_metadata. Study: Swedish-EBNA1-mimic-cohorts.

6. Thomas OG, Bronge M, Tengvall K et al. Cross-reactive EBNA1 immunity targets alpha-crystallin B and is associated with multiple sclerosis. Science advances. 2023. PMID 37196088. DOI: 10.1126/sciadv.adg3032. Access: abstract_metadata. Study: Swedish-EBNA1-mimic-cohorts.

7. Thomas OG, Rykaczewska U, Galešić M et al. Anoctamin-2-specific T cells link Epstein-Barr virus to multiple sclerosis. Cell. 2026. PMID 41534529. DOI: 10.1016/j.cell.2025.12.032. Access: abstract_metadata. Study: PMID41534529.

8. Thomas OG, Haigh TA, Croom-Carter D et al. Heightened Epstein-Barr virus immunity and potential cross-reactivities in multiple sclerosis. PLoS pathogens. 2024. PMID 38843296. DOI: 10.1371/journal.ppat.1012177. Access: abstract_metadata. Study: PMID38843296.

9. Munger KL, Fitzgerald KC, Freedman MS et al. No association of multiple sclerosis activity and progression with EBV or tobacco use in BENEFIT. Neurology. 2015. PMID 26453645. DOI: 10.1212/WNL.0000000000002099. Access: abstract_metadata. Study: PMID26453645.

10. Tengvall K, Huang J, Hellström C et al. Molecular mimicry between Anoctamin 2 and Epstein-Barr virus nuclear antigen 1 associates with multiple sclerosis risk. Proceedings of the National Academy of Sciences of the United States of America. 2019. PMID 31375628. DOI: 10.1073/pnas.1902623116. Access: abstract_metadata. Study: Swedish-MS-ANO2-EIMS-GEMS.

11. Vasilenko N, Tieck MP, Michel T et al. In-depth analysis of serum antibodies against Epstein-Barr virus lifecycle proteins, and EBNA1, ANO2, GlialCAM and CRYAB peptides in patients with multiple sclerosis. Frontiers in immunology. 2024. PMID 39742283. DOI: 10.3389/fimmu.2024.1487523. Access: abstract_metadata. Study: PMID39742283.

12. Haahr S, Plesner AM, Vestergaard BF et al. A role of late Epstein-Barr virus infection in multiple sclerosis. Acta neurologica Scandinavica. 2004. PMID 15016009. DOI: 10.1046/j.1600-0404.2003.00221.x. Access: abstract_metadata. Study: PMID15016009.

13. Munger KL, Hongell K, Cortese M et al. Epstein-barr virus and multiple sclerosis risk in the finnish maternity cohort. Annals of neurology. 2019. PMID 31226219. DOI: 10.1002/ana.25532. Access: abstract_metadata. Study: PMID31226219.

14. Wang J, Qiu Y, Marti Z et al. EBV infection and HLA-DR15 jointly drive multiple sclerosis by myelin peptide presentation. Cell. 2026. PMID 41534530. DOI: 10.1016/j.cell.2025.12.046. Access: abstract_metadata. Study: PMID41534530.

15. Bjornevik K, Mahler JV, Bilodeau PA et al. CD4+ T cells reactive to Epstein-Barr virus late lytic antigens are enriched in individuals with multiple sclerosis. Science translational medicine. 2026. PMID 42455903. DOI: 10.1126/scitranslmed.adz6566. Access: abstract_metadata. Study: PMID42455903.

16. Hayashi F, Mittl K, Dandekar R et al. Antigen specificity of clonally enriched CD8+ T cells in multiple sclerosis. Nature immunology. 2026. PMID 41644766. DOI: 10.1038/s41590-025-02412-3. Access: abstract_metadata. Study: PMID41644766.

17. Kim H, Schneider M, Raach Y et al. Myelin antigen capture in the CNS by B cells expressing EBV latent membrane protein 1 leads to demyelinating lesion formation. Cell. 2026. PMID 41534531. DOI: 10.1016/j.cell.2025.12.031. Access: abstract_metadata. Study: PMID41534531.

18. Pender MP, Csurhes PA, Smith C et al. Epstein-Barr virus-specific T cell therapy for progressive multiple sclerosis. JCI insight. 2018. PMID 30429369. DOI: 10.1172/jci.insight.124714. Access: abstract_metadata. Study: EBV-T-cell-MS-phase1.

19. Rees JH, Soudain SE, Gregson NA et al. Campylobacter jejuni infection and Guillain-Barré syndrome. The New England journal of medicine. 1995. PMID 7477117. DOI: 10.1056/NEJM199511233332102. Access: abstract_metadata. Study: PMID7477117.

20. Koga M, Gilbert M, Takahashi M et al. Comprehensive analysis of bacterial risk factors for the development of Guillain-Barre syndrome after Campylobacter jejuni enteritis. The Journal of infectious diseases. 2006. PMID 16425134. DOI: 10.1086/499969. Access: abstract_metadata. Study: Japanese-GBS-Campylobacter-isolate-platform.

21. Nachamkin I, Liu J, Li M et al. Campylobacter jejuni from patients with Guillain-Barré syndrome preferentially expresses a GD(1a)-like epitope. Infection and immunity. 2002. PMID 12183587. DOI: 10.1128/IAI.70.9.5299-5303.2002. Access: abstract_metadata. Study: Dutch-GBS-Campylobacter-isolate-platform.

22. Jacobs BC, van Doorn PA, Schmitz PI et al. Campylobacter jejuni infections and anti-GM1 antibodies in Guillain-Barré syndrome. Annals of neurology. 1996. PMID 8773599. DOI: 10.1002/ana.410400209. Access: abstract_metadata. Study: PMID8773599.

23. Kuwabara S, Ogawara K, Misawa S et al. Does Campylobacter jejuni infection elicit "demyelinating" Guillain-Barre syndrome? Neurology. 2004. PMID 15304587. DOI: 10.1212/01.wnl.0000133205.05169.04. Access: abstract_metadata. Study: PMID15304587.

24. Ang CW, Endtz HP, Jacobs BC et al. Campylobacter jejuni lipopolysaccharides from Guillain-Barré syndrome patients induce IgG anti-GM1 antibodies in rabbits. Journal of neuroimmunology. 2000. PMID 10713352. DOI: 10.1016/s0165-5728(99)00279-9. Access: abstract_metadata. Study: PMID10713352.

25. Gilbert M, Brisson JR, Karwaski MF et al. Biosynthesis of ganglioside mimics in Campylobacter jejuni OH4384. Identification of the glycosyltransferase genes, enzymatic synthesis of model compounds, and characterization of nanomole amounts by 600-mhz (1)h and (13)c NMR analysis. The Journal of biological chemistry. 2000. PMID 10660542. DOI: 10.1074/jbc.275.6.3896. Access: abstract_metadata. Study: PMID10660542.

26. Ang CW, De Klerk MA, Endtz HP et al. Guillain-Barré syndrome- and Miller Fisher syndrome-associated Campylobacter jejuni lipopolysaccharides induce anti-GM1 and anti-GQ1b Antibodies in rabbits. Infection and immunity. 2001. PMID 11254608. DOI: 10.1128/IAI.69.4.2462-2469.2001. Access: abstract_metadata. Study: PMID11254608.

27. Koga M, Takahashi M, Masuda M et al. Campylobacter gene polymorphism as a determinant of clinical features of Guillain-Barré syndrome. Neurology. 2005. PMID 16162859. DOI: 10.1212/01.wnl.0000176914.70893.14. Access: abstract_metadata. Study: Japanese-GBS-Campylobacter-isolate-platform.

28. Yuki N, Taki T, Inagaki F et al. A bacterium lipopolysaccharide that elicits Guillain-Barré syndrome has a GM1 ganglioside-like structure. The Journal of experimental medicine. 1993. PMID 8228822. DOI: 10.1084/jem.178.5.1771. Access: abstract_metadata. Study: PMID8228822.

29. Goodfellow JA, Bowes T, Sheikh K et al. Overexpression of GD1a ganglioside sensitizes motor nerve terminals to anti-GD1a antibody-mediated injury in a model of acute motor axonal neuropathy. The Journal of neuroscience : the official journal of the Society for Neuroscience. 2005. PMID 15716397. DOI: 10.1523/JNEUROSCI.4279-04.2005. Access: abstract_metadata. Study: PMID15716397.

30. Yuki N, Susuki K, Koga M et al. Carbohydrate mimicry between human ganglioside GM1 and Campylobacter jejuni lipooligosaccharide causes Guillain-Barre syndrome. Proceedings of the National Academy of Sciences of the United States of America. 2004. PMID 15277677. DOI: 10.1073/pnas.0402391101. Access: abstract_metadata. Study: PMID15277677.

31. Godschalk PC, Heikema AP, Gilbert M et al. The crucial role of Campylobacter jejuni genes in anti-ganglioside antibody induction in Guillain-Barre syndrome. The Journal of clinical investigation. 2004. PMID 15578098. DOI: 10.1172/JCI15707. Access: full_text_partial. Study: Dutch-GBS-Campylobacter-isolate-platform.

32. Godschalk PC, Kuijf ML, Li J et al. Structural characterization of Campylobacter jejuni lipooligosaccharide outer cores associated with Guillain-Barre and Miller Fisher syndromes. Infection and immunity. 2007. PMID 17261613. DOI: 10.1128/IAI.00872-06. Access: abstract_metadata. Study: Dutch-GBS-Campylobacter-isolate-platform.

33. Gonzalez-Figueroa P, Roco JA, Papa I et al. Follicular regulatory T cells produce neuritin to regulate B cells. Cell. 2021. PMID 33711260. DOI: 10.1016/j.cell.2021.02.027. Access: abstract_metadata. Study: PMID33711260.

34. Wing K, Onishi Y, Prieto-Martin P et al. CTLA-4 control over Foxp3+ regulatory T cell function. Science (New York, N.Y.). 2008. PMID 18845758. DOI: 10.1126/science.1160062. Access: abstract_metadata. Study: PMID18845758.

35. Schubert D, Bode C, Kenefeck R et al. Autosomal dominant immune dysregulation syndrome in humans with CTLA4 mutations. Nature medicine. 2014. PMID 25329329. DOI: 10.1038/nm.3746. Access: abstract_metadata. Study: PMID25329329.

36. Taghizade N, Babayeva R, Kara A et al. Therapeutic modalities and clinical outcomes in a large cohort with LRBA deficiency and CTLA4 insufficiency. The Journal of allergy and clinical immunology. 2023. PMID 37595759. DOI: 10.1016/j.jaci.2023.08.004. Access: abstract_metadata. Study: PMID37595759.

37. Goldfarb Y, Givony T, Kadouri N et al. Mechanistic dissection of dominant AIRE mutations in mouse models reveals AIRE autoregulation. The Journal of experimental medicine. 2021. PMID 34477806. DOI: 10.1084/jem.20201076. Access: abstract_metadata. Study: PMID34477806.

38. Anderson MS, Venanzi ES, Klein L et al. Projection of an immunological self shadow within the thymus by the aire protein. Science (New York, N.Y.). 2002. PMID 12376594. DOI: 10.1126/science.1075958. Access: abstract_metadata. Study: PMID12376594.

39. Bennett CL, Christie J, Ramsdell F et al. The immune dysregulation, polyendocrinopathy, enteropathy, X-linked syndrome (IPEX) is caused by mutations of FOXP3. Nature genetics. 2001. PMID 11137993. DOI: 10.1038/83713. Access: abstract_metadata. Study: PMID11137993.

40. Wildin RS, Ramsdell F, Peake J et al. X-linked neonatal diabetes mellitus, enteropathy and endocrinopathy syndrome is the human equivalent of mouse scurfy. Nature genetics. 2001. PMID 11137992. DOI: 10.1038/83707. Access: abstract_metadata. Study: PMID11137992.

41. Sakaguchi S, Sakaguchi N, Asano M et al. Immunologic self-tolerance maintained by activated T cells expressing IL-2 receptor alpha-chains (CD25). Breakdown of a single mechanism of self-tolerance causes various autoimmune diseases. Journal of immunology (Baltimore, Md. : 1950). 1995. PMID 7636184. DOI: none assigned (pre-DOI Journal of Immunology record). Access: abstract_metadata. Study: PMID7636184.

42. Kim JM, Rasmussen JP, Rudensky AY et al. Regulatory T cells prevent catastrophic autoimmunity throughout the lifespan of mice. Nature immunology. 2007. PMID 17136045. DOI: 10.1038/ni1428. Access: abstract_metadata. Study: PMID17136045.

43. Kuehn HS, Ouyang W, Lo B et al. Immune dysregulation in human subjects with heterozygous germline mutations in CTLA4. Science (New York, N.Y.). 2014. PMID 25213377. DOI: 10.1126/science.1255904. Access: full_text_partial. Study: PMID25213377.

44. Nishimura H, Nose M, Hiai H et al. Development of lupus-like autoimmune diseases by disruption of the PD-1 gene encoding an ITIM motif-carrying immunoreceptor. Immunity. 1999. PMID 10485649. DOI: 10.1016/s1074-7613(00)80089-8. Access: abstract_metadata. Study: PMID10485649.

45. Nishimura H, Okazaki T, Tanaka Y et al. Autoimmune dilated cardiomyopathy in PD-1 receptor-deficient mice. Science (New York, N.Y.). 2001. PMID 11209085. DOI: 10.1126/science.291.5502.319. Access: abstract_metadata. Study: PMID11209085.

46. Ogishi M, Yang R, Aytekin C et al. Inherited PD-1 deficiency underlies tuberculosis and autoimmunity in a child. Nature medicine. 2021. PMID 34183838. DOI: 10.1038/s41591-021-01388-5. Access: abstract_metadata. Study: PMID34183838.

47. Kamitani F, Nishioka Y, Koizumi M et al. Immune checkpoint inhibitor-related type 1 diabetes incidence, risk, and survival association. Journal of diabetes investigation. 2025. PMID 39569589. DOI: 10.1111/jdi.14362. Access: abstract_metadata. Study: PMID39569589.

48. Younis S, Moutusy SI, Rasouli S et al. Epstein-Barr virus reprograms autoreactive B cells as antigen-presenting cells in systemic lupus erythematosus. Science translational medicine. 2025. PMID 41223250. DOI: 10.1126/scitranslmed.ady0210. Access: full_text_partial. Study: PMID41223250.

49. Traka M, Podojil JR, McCarthy DP et al. Oligodendrocyte death results in immune-mediated CNS demyelination. Nature neuroscience. 2016. PMID 26656646. DOI: 10.1038/nn.4193. Access: abstract_metadata. Study: PMID26656646.

50. Lincez PJ, Shanina I, Horwitz MS et al. Changes in MDA5 and TLR3 Sensing of the Same Diabetogenic Virus Result in Different Autoimmune Disease Outcomes. Frontiers in immunology. 2021. PMID 34804036. DOI: 10.3389/fimmu.2021.751341. Access: abstract_metadata. Study: PMID34804036.

51. Vehik K, Lynch KF, Wong MC et al. Prospective virome analyses in young children at increased genetic risk for type 1 diabetes. Nature medicine. 2019. PMID 31792456. DOI: 10.1038/s41591-019-0667-0. Access: abstract_metadata. Study: TEDDY-virome.

52. Zhao G, Vatanen T, Droit L et al. Intestinal virome changes precede autoimmunity in type I diabetes-susceptible children. Proceedings of the National Academy of Sciences of the United States of America. 2017. PMID 28696303. DOI: 10.1073/pnas.1706359114. Access: abstract_metadata. Study: PMID28696303.

53. McCall KD, Thuma JR, Courreges MC et al. Toll-like receptor 3 is critical for coxsackievirus B4-induced type 1 diabetes in female NOD mice. Endocrinology. 2015. PMID 25422874. DOI: 10.1210/en.2013-2006. Access: abstract_metadata. Study: PMID25422874.

54. Bason C, Lorini R, Lunardi C et al. In type 1 diabetes a subset of anti-coxsackievirus B4 antibodies recognize autoantigens and induce apoptosis of pancreatic beta cells. PloS one. 2013. PMID 23469060. DOI: 10.1371/journal.pone.0057729. Access: abstract_metadata. Study: PMID23469060.

55. Sadeharju K, Hämäläinen AM, Knip M et al. Enterovirus infections as a risk factor for type I diabetes: virus analyses in a dietary intervention trial. Clinical and experimental immunology. 2003. PMID 12699416. DOI: 10.1046/j.1365-2249.2003.02147.x. Access: abstract_metadata. Study: PMID12699416.

56. Krogvold L, Mynarek IM, Ponzi E et al. Pleconaril and ribavirin in new-onset type 1 diabetes: a phase 2 randomized trial. Nature medicine. 2023. PMID 37789144. DOI: 10.1038/s41591-023-02576-1. Access: full_text_partial. Study: DiViD-intervention-phase2.

57. Hyöty H, Kääriäinen S, Laiho JE et al. Safety, tolerability and immunogenicity of PRV-101, a multivalent vaccine targeting coxsackie B viruses (CVBs) associated with type 1 diabetes: a double-blind randomised placebo-controlled Phase I trial. Diabetologia. 2024. PMID 38369573. DOI: 10.1007/s00125-024-06092-w. Access: abstract_metadata. Study: PMID38369573.

58. Lin J, Moradi E, Salenius K et al. Distinct transcriptomic profiles in children prior to the appearance of type 1 diabetes-linked islet autoantibodies and following enterovirus infection. Nature communications. 2023. PMID 37993433. DOI: 10.1038/s41467-023-42763-9. Access: abstract_metadata. Study: PMID37993433.

59. Lindfors K, Lin J, Lee HS et al. Metagenomics of the faecal virome indicate a cumulative effect of enterovirus and gluten amount on the risk of coeliac disease autoimmunity in genetically at risk children: the TEDDY study. Gut. 2020. PMID 31744911. DOI: 10.1136/gutjnl-2019-319809. Access: abstract_metadata. Study: TEDDY-virome-celiac.

60. Wen L, Ley RE, Volchkov PY et al. Innate immunity and intestinal microbiota in the development of Type 1 diabetes. Nature. 2008. PMID 18806780. DOI: 10.1038/nature07336. Access: full_text_partial. Study: PMID18806780.

61. Manfredo Vieira S, Hiltensperger M, Kumar V et al. Translocation of a gut pathobiont drives autoimmunity in mice and humans. Science (New York, N.Y.). 2018. PMID 29590047. DOI: 10.1126/science.aar7201. Access: abstract_metadata. Study: PMID29590047.

62. Zegarra-Ruiz DF, El Beidaq A, Iñiguez AJ et al. A Diet-Sensitive Commensal Lactobacillus Strain Mediates TLR7-Dependent Systemic Autoimmunity. Cell host & microbe. 2019. PMID 30581114. DOI: 10.1016/j.chom.2018.11.009. Access: abstract_metadata. Study: PMID30581114.

63. Konig MF, Abusleme L, Reinholdt J et al. Aggregatibacter actinomycetemcomitans-induced hypercitrullination links periodontal infection to autoimmunity in rheumatoid arthritis. Science translational medicine. 2016. PMID 27974664. DOI: 10.1126/scitranslmed.aaj1921. Access: abstract_metadata. Study: PMID27974664.

64. Mukherjee A, Jantsch V, Khan R et al. Rheumatoid Arthritis-Associated Autoimmunity Due to Aggregatibacter actinomycetemcomitans and Its Resolution With Antibiotic Therapy. Frontiers in immunology. 2018. PMID 30459755. DOI: 10.3389/fimmu.2018.02352. Access: abstract_metadata. Study: PMID30459755.

65. Chriswell ME, Lefferts AR, Clay MR et al. Clonal IgA and IgG autoantibodies from individuals at risk for rheumatoid arthritis identify an arthritogenic strain of Subdoligranulum. Science translational medicine. 2022. PMID 36288282. DOI: 10.1126/scitranslmed.abn5166. Access: abstract_metadata. Study: PMID36288282.

66. Wu HJ, Ivanov II, Darce J et al. Gut-residing segmented filamentous bacteria drive autoimmune arthritis via T helper 17 cells. Immunity. 2010. PMID 20620945. DOI: 10.1016/j.immuni.2010.06.001. Access: abstract_metadata. Study: PMID20620945.

67. Kriegel MA, Sefik E, Hill JA et al. Naturally transmitted segmented filamentous bacteria segregate with diabetes protection in nonobese diabetic mice. Proceedings of the National Academy of Sciences of the United States of America. 2011. PMID 21709219. DOI: 10.1073/pnas.1108924108. Access: abstract_metadata. Study: PMID21709219.

68. Olia A, Shimokawa C, Imai T et al. Suppression of systemic lupus erythematosus in NZBWF1 mice infected with Hymenolepis microstoma. Parasitology international. 2020. PMID 31954872. DOI: 10.1016/j.parint.2020.102057. Access: abstract_metadata. Study: PMID31954872.

69. Liu X, Jiao Y, Cui B et al. The potential protective role of hepatitis B virus infection in pristane-induced lupus in mice. Lupus. 2016. PMID 27125291. DOI: 10.1177/0961203316631637. Access: abstract_metadata. Study: PMID27125291.

70. Zhang X, Zhang D, Jia H et al. The oral and gut microbiomes are perturbed in rheumatoid arthritis and partly normalized after treatment. Nature medicine. 2015. PMID 26214836. DOI: 10.1038/nm.3914. Access: abstract_metadata. Study: PMID26214836.

Source-specific evidence cards

[1] Longitudinal analysis reveals high prevalence of Epstein-Barr virus associated with multiple sclerosis.

Shows: EBV seroconversion precedes incident MS and neurofilament rise; CMV provides an exposure negative control.

Does not show: Prospective natural exposure strongly constrains reverse causation but does not isolate mimicry or baseline tolerance mediation.

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

[2] Clonally expanded B cells in multiple sclerosis bind EBV EBNA1 and GlialCAM.

Shows: Patient-derived antibody cross-binds EBNA1 and GlialCAM; peptide immunization aggravates PLP-induced EAE.

Does not show: Selected Results show induced EAE amplification rather than spontaneous MS initiation or selective human mimic necessity.

Access/status: Complete primary abstract and linked-status metadata examined. Selected body sections: 4:Molecular mimicry between EBNA1AA386–405 and GlialCAM, 5:Phosphorylation of GlialCAMAA370–389 at pSer376 enables MS CSF-derived MS39p2w174 binding, 7:Immunization with EBNA1AA386–405 aggravates EAE. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2b, SC4.

[3] Serologic Response to the Epstein-Barr Virus Peptidome and the Risk for Multiple Sclerosis.

Shows: Prediagnostic EBV peptidome responses are increased without one selectively targeted peptide.

Does not show: Small military nested cohort shares a repository with other reports; adjusted estimates and wide intervals limit unique epitope inference.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2b.

[4] HLA-DR15 Molecules Jointly Shape an Autoreactive T Cell Repertoire in Multiple Sclerosis.

Shows: HLA-DR15 presents self and foreign peptides to cross-reactive patient T-cell clones.

Does not show: Immunopeptidomics and ex vivo recognition do not establish which antigen first caused human onset.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2b, SC3.

[5] Antibody reactivity against EBNA1 and GlialCAM differentiates multiple sclerosis patients from healthy controls.

Shows: GlialCAM/EBNA1 reciprocal blocking confirms cross-reactivity and HLA-associated risk patterns.

Does not show: Established-case serology and overlapping mimic cohorts cannot establish onset mediation.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2b.

[6] Cross-reactive EBNA1 immunity targets alpha-crystallin B and is associated with multiple sclerosis.

Shows: EBNA1/CRYAB reciprocal blocking and mouse responses support cross-reactive immunity.

Does not show: Risk serology and immunization do not isolate CRYAB as a necessary initiator of MS.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2b.

[7] Anoctamin-2-specific T cells link Epstein-Barr virus to multiple sclerosis.

Shows: ANO2 pre-immunization and transferred CD4 cells exacerbate EAE; human clones cross-recognize EBNA1.

Does not show: Induced EAE enhancement is a phenotype but does not establish spontaneous human onset necessity.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2b.

[8] Heightened Epstein-Barr virus immunity and potential cross-reactivities in multiple sclerosis.

Shows: EBV-expanded T-cell lines recognize CNS antigens in both MS and healthy controls.

Does not show: Cross-reactivity also in controls limits disease-specific sufficiency; peripheral viral load is unchanged.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2b, SC4.

[9] No association of multiple sclerosis activity and progression with EBV or tobacco use in BENEFIT.

Shows: BENEFIT antibody titres do not predict CIS conversion or subsequent clinical/MRI activity.

Does not show: A post-CIS titre null does not test preventing EBV infection before initiation.

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

[10] Molecular mimicry between Anoctamin 2 and Epstein-Barr virus nuclear antigen 1 associates with multiple sclerosis risk.

Shows: ANO2/EBNA1 blocking and genotype combinations link cross-reactivity to MS risk.

Does not show: Case-control antibody positivity occurs in a minority and does not demonstrate individual causal initiation.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2b.

[11] In-depth analysis of serum antibodies against Epstein-Barr virus lifecycle proteins, and EBNA1, ANO2, GlialCAM and CRYAB peptides in patients with multiple sclerosis.

Shows: ANO2 differs in MS but GlialCAM/CRYAB responses and disease-phase signatures do not distinguish consistently.

Does not show: Cross-sectional assay/cohort heterogeneity qualifies universal mimic or reactivation narratives.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2b, SC4.

[12] A role of late Epstein-Barr virus infection in multiple sclerosis.

Shows: Late EBV exposure and recalled mononucleosis associate with MS in a high-prevalence region.

Does not show: Recall and geographic contrasts do not isolate infection timing from correlated exposures.

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

[13] Epstein-barr virus and multiple sclerosis risk in the finnish maternity cohort.

Shows: Prediagnostic maternal/offspring EBV antibody associations recur with CMV comparison.

Does not show: Maternal antibodies are not the child's measured infection; observational confounding remains.

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

[14] EBV infection and HLA-DR15 jointly drive multiple sclerosis by myelin peptide presentation.

Shows: EBV reprograms HLA-DR15 B-cell presentation of MBP peptides recognized by MS T cells.

Does not show: Human tissue alignment and infected-cell experiments support a route distinct from proving onset through molecular mimicry.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2b, SC4.

[15] CD4+ T cells reactive to Epstein-Barr virus late lytic antigens are enriched in individuals with multiple sclerosis.

Shows: MS CD4 responses preferentially target late-lytic EBV antigens and fall after anti-CD20 treatment.

Does not show: B-cell depletion changes multiple immune functions; reduced responses or shedding do not isolate viral initiation.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2b, SC4.

[16] Antigen specificity of clonally enriched CD8+ T cells in multiple sclerosis.

Shows: Expanded CSF CD8 clonotypes include functionally EBV-reactive receptors; many mimotopes fail functional response.

Does not show: Established-disease antigen specificity and the updated preprint are one report lineage, not prospective onset proof.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. UpdateOf: PMID39282370. Subclaims: SC2b, SC4.

[17] Myelin antigen capture in the CNS by B cells expressing EBV latent membrane protein 1 leads to demyelinating lesion formation.

Shows: LMP1 or CD40L overrides death of antigen-capturing myelin-reactive B cells and enables lesions.

Does not show: Engineered viral-protein signaling tests a bounded route; ectopic expression cannot establish endogenous human primacy alone.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2b, SC4.

[18] Epstein-Barr virus-specific T cell therapy for progressive multiple sclerosis.

Shows: Open-label EBV-specific T-cell treatment is feasible and clinical improvement tracks product reactivity.

Does not show: Small uncontrolled progressive-MS phase I trial does not prove efficacy or prevention; linked erratum details unexamined.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. ErratumIn: PMID33055421. Subclaims: SC4.

[19] Campylobacter jejuni infection and Guillain-Barré syndrome.

Shows: Recent Campylobacter infection is enriched in incident GBS and links to motor/axonal severity.

Does not show: Case-control antecedence does not alone prove mimicry or explain non-Campylobacter GBS.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2a, SC4.

[20] Comprehensive analysis of bacterial risk factors for the development of Guillain-Barre syndrome after Campylobacter jejuni enteritis.

Shows: Bacterial LOS class and sialyltransferase genotype distinguish neuropathy from uncomplicated enteritis isolates.

Does not show: Isolate ascertainment and linked strain features limit attributing clinical risk to one gene alone.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2a.

[21] Campylobacter jejuni from patients with Guillain-Barré syndrome preferentially expresses a GD(1a)-like epitope.

Shows: GBS isolates preferentially express GD1a-like mimicry with relevant biosynthesis genes.

Does not show: Strain association is not a selective onset intervention.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2a.

[22] Campylobacter jejuni infections and anti-GM1 antibodies in Guillain-Barré syndrome.

Shows: Anti-GM1 positivity associates with recent Campylobacter and a severe motor phenotype.

Does not show: Antibody-positive disease subgroups are not all GBS and observational treatment differences are not randomized mediation.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC2a.

[23] Does Campylobacter jejuni infection elicit "demyelinating" Guillain-Barre syndrome?

Shows: Serial conduction studies reclassify transient Campylobacter-related demyelinating patterns as axonal neuropathy.

Does not show: Clinical classification depends on timing; this limits broad demyelination transfer but does not refute axonal mimicry.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2a, SC5.

[24] Campylobacter jejuni lipopolysaccharides from Guillain-Barré syndrome patients induce IgG anti-GM1 antibodies in rabbits.

Shows: GBS-strain LPS immunization induces GM1-cross-reactive rabbit antibodies.

Does not show: Antibody induction alone is not demonstrated neuropathic functional injury.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2a.

[25] Biosynthesis of ganglioside mimics in Campylobacter jejuni OH4384. Identification of the glycosyltransferase genes, enzymatic synthesis of model compounds, and characterization of nanomole amounts by 600-mhz (1)h and (13)c NMR analysis.

Shows: Cloned glycosyltransferases synthesize defined ganglioside-like structures with linkage confirmation.

Does not show: Biochemical biosynthesis establishes the molecular mimic rather than human disease necessity.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2a.

[26] Guillain-Barré syndrome- and Miller Fisher syndrome-associated Campylobacter jejuni lipopolysaccharides induce anti-GM1 and anti-GQ1b Antibodies in rabbits.

Shows: GBS/MFS LPS induces matched antibody specificity; control-strain LPS lacks ganglioside reactivity.

Does not show: Immunization establishes antigen specificity without showing spontaneous infection produces neuropathy in every host.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2a.

[27] Campylobacter gene polymorphism as a determinant of clinical features of Guillain-Barré syndrome.

Shows: cst-II polymorphism aligns bacterial mimic, patient autoantibody and neurological phenotype.

Does not show: Same clinical isolate platform can share failure modes with other genotype association reports.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2a.

[28] A bacterium lipopolysaccharide that elicits Guillain-Barré syndrome has a GM1 ganglioside-like structure.

Shows: NMR and chemical analysis identify a GM1-identical terminal bacterial oligosaccharide.

Does not show: Structural identity is not alone a causal neurological experiment.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2a.

[29] Overexpression of GD1a ganglioside sensitizes motor nerve terminals to anti-GD1a antibody-mediated injury in a model of acute motor axonal neuropathy.

Shows: Anti-GD1a antibody plus complement injures high-antigen motor terminals and blocks function.

Does not show: Tolerance bypass and ganglioside-overexpressing recipients are engineered; normal mice resist the same exposure.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2a.

[30] Carbohydrate mimicry between human ganglioside GM1 and Campylobacter jejuni lipooligosaccharide causes Guillain-Barre syndrome.

Shows: LOS-sensitized rabbits develop anti-GM1 antibodies, weakness and GBS-like peripheral pathology.

Does not show: Patient/mouse antibody blocks coculture potentials but does not itself induce paralysis; selected full body unavailable.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2a, SC4.

[31] The crucial role of Campylobacter jejuni genes in anti-ganglioside antibody induction in Guillain-Barre syndrome.

Shows: cst-II/orf10 loss removes sialylated mimics and anti-GD1a induction while orf11 is a negative control.

Does not show: Original publisher Methods use ganglioside-deficient mice and repeated adjuvant immunization; asialo-GM1 responses remain and neuropathy was not measured.

Access/status: Complete primary abstract and linked-status metadata examined. Selected body sections: 0:Publisher HTML paragraphs:paragraph62, 0:Publisher HTML paragraphs:paragraph65, 0:Publisher HTML paragraphs:paragraph70, 0:Publisher HTML paragraphs:paragraph82, 0:Publisher HTML paragraphs:paragraph92. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2a.

[32] Structural characterization of Campylobacter jejuni lipooligosaccharide outer cores associated with Guillain-Barre and Miller Fisher syndromes.

Shows: Some neuropathy-associated isolates lack detectable ganglioside mimics or have frameshifted biosynthesis genes.

Does not show: Post-isolation phase variation may differ from infection-time expression; these are boundaries rather than definitive mimic absence at onset.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2a, SC5.

[33] Follicular regulatory T cells produce neuritin to regulate B cells.

Shows: Foxp3-lineage neuritin loss causes abnormal germinal-center/plasma-cell autoimmunity with reconstitution effects.

Does not show: Rescue primarily changes cellular/antibody endpoints and does not establish infection initiation.

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

[34] CTLA-4 control over Foxp3+ regulatory T cell function.

Shows: Treg-specific CTLA4 loss causes spontaneous fatal autoimmune disease.

Does not show: Conditional lineage specificity supports tolerance function; developmental/immune interactions do not establish a universal trigger.

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

[35] Autosomal dominant immune dysregulation syndrome in humans with CTLA4 mutations.

Shows: Independent human CTLA4 variants impair regulatory function and produce immune dysregulation with incomplete penetrance.

Does not show: Unaffected carriers invalidate complete-penetrance claims; no measured infectious mimic is required by the proposed genotype mechanism.

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

[36] Therapeutic modalities and clinical outcomes in a large cohort with LRBA deficiency and CTLA4 insufficiency.

Shows: Abatacept controls manifestations in a genotype-defined LRBA/CTLA4 cohort.

Does not show: Nonrandomized treatment allocation and disease-duration differences limit comparative superiority estimates.

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

[37] Mechanistic dissection of dominant AIRE mutations in mouse models reveals AIRE autoregulation.

Shows: Engineered dominant AIRE mutations disrupt central tolerance and alter chromatin/transcription.

Does not show: Mutant-line developmental biology and expression changes do not prove ordinary sporadic autoimmunity is AIRE-deficient.

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

[38] Projection of an immunological self shadow within the thymus by the aire protein.

Shows: Aire-deficient thymic stroma loses tissue-antigen transcription and supports defined autoimmunity.

Does not show: Constitutive loss has developmental confounds; thymic localization narrows but does not remove them.

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

[39] The immune dysregulation, polyendocrinopathy, enteropathy, X-linked syndrome (IPEX) is caused by mutations of FOXP3.

Shows: Human FOXP3 mutations segregate with IPEX immune dysregulation.

Does not show: Rare syndromic causality is not a class-wide requirement for pathogenic FOXP3 variants.

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

[40] X-linked neonatal diabetes mellitus, enteropathy and endocrinopathy syndrome is the human equivalent of mouse scurfy.

Shows: Independent IPEX sequencing finds forkhead-domain FOXP3 variants matching scurfy biology.

Does not show: Genetic identification alone does not measure a specific infection or explain all polygenic disease.

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

[41] Immunologic self-tolerance maintained by activated T cells expressing IL-2 receptor alpha-chains (CD25). Breakdown of a single mechanism of self-tolerance causes various autoimmune diseases.

Shows: CD25-cell removal permits autoimmune disease and timely reconstitution prevents it.

Does not show: Athymic transfer and imperfect CD25 lineage specificity constrain physiology; adult conditional depletion is an orthogonal follow-up.

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

[42] Regulatory T cells prevent catastrophic autoimmunity throughout the lifespan of mice.

Shows: Acute neonatal/adult Foxp3-Treg ablation causes autoimmunity without lifelong gene-loss dependence.

Does not show: Mouse induced depletion establishes ongoing regulatory necessity, not that human infection always removes Tregs.

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

[43] Immune dysregulation in human subjects with heterozygous germline mutations in CTLA4.

Shows: Human CTLA4 haploinsufficiency causes cellular dysregulation and wild-type complementation corrects ex vivo proliferation.

Does not show: Original text preserves healthy carriers and culture rescue; common disease initiation is a separate scope.

Access/status: Complete primary abstract and linked-status metadata examined. Selected body sections: 1:Main text:paragraph3, 1:Main text:paragraph5, 1:Main text:paragraph6, 1:Main text:paragraph10. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3, SC5.

[44] Development of lupus-like autoimmune diseases by disruption of the PD-1 gene encoding an ITIM motif-carrying immunoreceptor.

Shows: PD1 loss produces background-dependent lupus-like autoimmunity accelerated by Fas loss.

Does not show: Constitutive immune development and genetic background limit generalization.

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

[45] Autoimmune dilated cardiomyopathy in PD-1 receptor-deficient mice.

Shows: BALB/c PD1 loss causes autoimmune cardiomyopathy while RAG2 deficiency prevents it.

Does not show: B/T-cell dependency and strain context do not identify an infectious initiator.

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

[46] Inherited PD-1 deficiency underlies tuberculosis and autoimmunity in a child.

Shows: Inherited human PD1 deficiency impairs antimycobacterial immunity and causes immune dysregulation.

Does not show: Single affected child with TB cannot independently determine whether infection or tolerance loss initiated each lesion.

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

[47] Immune checkpoint inhibitor-related type 1 diabetes incidence, risk, and survival association.

Shows: Incident T1D follows checkpoint therapy with regimen and baseline-disease risk contrasts.

Does not show: Administrative coding and nonrandomized therapy comparisons constrain incidence attribution; survival analysis is vulnerable to time-dependent selection.

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

[48] Epstein-Barr virus reprograms autoreactive B cells as antigen-presenting cells in systemic lupus erythematosus.

Shows: Established-SLE EBV-positive B cells encode nuclear autoreactivity and transformed chromatin-loaded cells activate autologous helper/B cells.

Does not show: Original Results use diagnosed patients and cell-line genomic datasets; no prediagnostic onset sequence or selective in vivo viral depletion is established.

Access/status: Complete primary abstract and linked-status metadata examined. Selected body sections: 0:Results:paragraph1, 0:Results:paragraph10, 0:Results:paragraph11, 0:Results:paragraph17, 0:Results:paragraph18, 0:Results:paragraph19, 0:Results:paragraph20. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4, SC5.

[49] Oligodendrocyte death results in immune-mediated CNS demyelination.

Shows: Oligodendrocyte ablation precedes delayed adaptive demyelination; transfer and MOG tolerization change disease.

Does not show: Engineered tissue injury establishes an alternative route in a model, not proof that human MS routinely begins without EBV.

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

[50] Changes in MDA5 and TLR3 Sensing of the Same Diabetogenic Virus Result in Different Autoimmune Disease Outcomes.

Shows: Sensing the same diabetogenic virus differently through MDA5/TLR3 changes diabetes risk and can protect.

Does not show: Sensor, local interferon and host state determine sign; this is not a universal infection-to-disease rule.

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

[51] Prospective virome analyses in young children at increased genetic risk for type 1 diabetes.

Shows: TEDDY prolonged enterovirus-B shedding associates with islet autoimmunity but not T1D incidence.

Does not show: Repeated stools measure exposure patterns and surrogate seroconversion, not necessary clinical initiation.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC4, SC5.

[52] Intestinal virome changes precede autoimmunity in type I diabetes-susceptible children.

Shows: Virome changes precede autoimmunity but common enterovirus sequences do not associate in this small cohort.

Does not show: Small sample and linked correction details unexamined limit precise diversity inference.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. ErratumIn: PMID30455315. Subclaims: SC1, SC4, SC5.

[53] Toll-like receptor 3 is critical for coxsackievirus B4-induced type 1 diabetes in female NOD mice.

Shows: TLR3 knockout protects female NOD from CB4 diabetes and reduces insulitis versus matched uninfected controls.

Does not show: Constitutive receptor loss and forced viral challenge establish model-specific causal contribution.

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

[54] In type 1 diabetes a subset of anti-coxsackievirus B4 antibodies recognize autoantigens and induce apoptosis of pancreatic beta cells.

Shows: Affinity-purified recent-onset T1D antibodies cross-bind viral/islet antigens and induce beta-cell apoptosis.

Does not show: Culture killing does not establish the infection-to-human-onset chain; SC2b comparison illustrates mimicry transfer limits, not MS evidence.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2b, SC4, SC5.

[55] Enterovirus infections as a risk factor for type I diabetes: virus analyses in a dietary intervention trial.

Shows: Enterovirus infections precede islet autoantibodies in genetically at-risk children.

Does not show: Formula allocation was randomized; viral infection exposure was not randomized and clinical T1D causality is unproven.

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

[56] Pleconaril and ribavirin in new-onset type 1 diabetes: a phase 2 randomized trial.

Shows: Randomized new-onset antiviral combination preserves stimulated C-peptide on the primary repeated-measures analysis.

Does not show: Post-diagnosis functional secretion is not prevention; peripheral PCR is negative, drug effects cannot be separated, and insulin-dose/12-month HbA1c contrasts are null.

Access/status: Complete primary abstract and linked-status metadata examined. Selected body sections: 1:Results:paragraph0, 1:Results:paragraph5, 1:Results:paragraph8, 1:Results:paragraph9, 1:Results:paragraph14, 1:Results:paragraph15, 2:Discussion:paragraph2, 2:Discussion:paragraph3, 2:Discussion:paragraph5, 2:Discussion:paragraph8, 3:Methods:paragraph11, 3:Methods:paragraph12, 3:Methods:paragraph15, 3:Methods:paragraph16. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4, SC5.

[57] Safety, tolerability and immunogenicity of PRV-101, a multivalent vaccine targeting coxsackie B viruses (CVBs) associated with type 1 diabetes: a double-blind randomised placebo-controlled Phase I trial.

Shows: Phase-I CVB vaccine is tolerable and induces neutralizing responses.

Does not show: Healthy-adult safety/immunogenicity without autoimmune incidence outcomes does not prove prevention.

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

[58] Distinct transcriptomic profiles in children prior to the appearance of type 1 diabetes-linked islet autoantibodies and following enterovirus infection.

Shows: Prediagnostic islet-autoantibody subgroups differ in transcription and antiviral-response signatures.

Does not show: Cell composition, genotype and response correlates do not isolate infectious initiation.

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

[59] Metagenomics of the faecal virome indicate a cumulative effect of enterovirus and gluten amount on the risk of coeliac disease autoimmunity in genetically at risk children: the TEDDY study.

Shows: TEDDY enterovirus exposure interacts with gluten intake in celiac autoimmunity risk.

Does not show: Nested observational analysis with autoantibody endpoint shares TEDDY data and does not remove the causal dietary antigen.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC4, SC5.

[60] Innate immunity and intestinal microbiota in the development of Type 1 diabetes.

Shows: Germ-free MyD88-null NOD regain diabetes and defined commensal colonization attenuates it.

Does not show: Original Results show protective microbiota and retained diabetogenic cells, with genotype/housing limits and no human prevention demonstration.

Access/status: Complete primary abstract and linked-status metadata examined. Selected body sections: 2:Main text:paragraph1, 2:Main text:paragraph2, 2:Main text:paragraph4, 2:Main text:paragraph5, 2:Main text:paragraph12, 2:Main text:paragraph13. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4, SC5.

[61] Translocation of a gut pathobiont drives autoimmunity in mice and humans.

Shows: E. gallinarum translocation induces autoimmunity in predisposed mice and pathogen-directed vaccine prevents mortality.

Does not show: Human liver DNA is associative; antibiotic pleiotropy and induced colonization do not establish spontaneous human initiation; correction attached.

Access/status: Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. ErratumIn: PMID29724927. Subclaims: SC4, SC5.

[62] A Diet-Sensitive Commensal Lactobacillus Strain Mediates TLR7-Dependent Systemic Autoimmunity.

Shows: L. reuteri colonization worsens TLR7-linked lupus and resistant starch lowers translocation and disease.

Does not show: Strain, diet and transgenic host set context; human fecal enrichment is not onset mediation.

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

[63] Aggregatibacter actinomycetemcomitans-induced hypercitrullination links periodontal infection to autoimmunity in rheumatoid arthritis.

Shows: Aa leukotoxin causes neutrophil hypercitrullination matching RA antigen patterns.

Does not show: Human exposure/ACPA association and ex vivo toxin action do not establish necessary RA onset.

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

[64] Rheumatoid Arthritis-Associated Autoimmunity Due to Aggregatibacter actinomycetemcomitans and Its Resolution With Antibiotic Therapy.

Shows: Aa endocarditis eradication coincides with sustained arthritis and ACPA resolution in one genetically at-risk patient.

Does not show: One RA-like infection-associated case lacks randomized withdrawal/rechallenge and cannot establish ordinary RA primacy.

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

[65] Clonal IgA and IgG autoantibodies from individuals at risk for rheumatoid arthritis identify an arthritogenic strain of Subdoligranulum.

Shows: One Subdoligranulum isolate causes arthritic autoimmunity in germ-free mice and lymphocyte depletion eliminates disease.

Does not show: At-risk human antibody selection plus germ-free strain challenge proves a bounded route rather than all RA initiation.

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

[66] Gut-residing segmented filamentous bacteria drive autoimmune arthritis via T helper 17 cells.

Shows: SFB colonization restores Th17/autoantibodies and arthritis in germ-free K/BxN; IL17 neutralization prevents arthritis.

Does not show: Transgenic host and immune background set context; one commensal can have the opposite sign in NOD diabetes.

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

[67] Naturally transmitted segmented filamentous bacteria segregate with diabetes protection in nonobese diabetic mice.

Shows: Naturally acquired SFB segregates with diabetes protection in female NOD despite persistent insulitis.

Does not show: Colonization is naturally incomplete and observational within the facility; sex dependence limits mechanistic certainty.

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

[68] Suppression of systemic lupus erythematosus in NZBWF1 mice infected with Hymenolepis microstoma.

Shows: Early H. microstoma infection suppresses spontaneous lupus manifestations in NZBWF1.

Does not show: Parasitic immune pleiotropy and unsuccessful nematode settlement limit target specificity and human transfer.

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

[69] The potential protective role of hepatitis B virus infection in pristane-induced lupus in mice.

Shows: HBV-transgenic mice have reduced pristane-lupus renal and antibody pathology.

Does not show: The actual manipulation is a transgene, not acquired HBV infection; the abstract's human-protection conclusion is unsupported by this experiment.

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

[70] The oral and gut microbiomes are perturbed in rheumatoid arthritis and partly normalized after treatment.

Shows: RA oral/gut microbiota vary with disease and partly normalize after treatment.

Does not show: Treatment-associated normalization permits reverse causation and does not show dysbiosis first initiated RA.

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

Documentary record

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

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

Sources: 70 · load-bearing claims: 77 · verified: 77 · unverified: 0 · contradicted: 0 · schema 5.3