Pneumonic Plague (MONDO:0001024): A Comprehensive Disease Characterization

Autonomous discovery report — 5 iterations, 14 confirmed findings, 53 papers reviewed


Summary

Pneumonic plague is a rapidly progressive, near-uniformly fatal infection of the lung caused by the Gram-negative bacterium Yersinia pestis (NCBITaxon:632). It is the most dangerous of the three clinical forms of plague (bubonic, septicemic, pneumonic) because it is the only form that is directly transmissible from person to person by respiratory droplets, and because — untreated — its case-fatality approaches 100% within 3–7 days. Two routes produce it: primary pneumonic plague, from inhaling infectious aerosols, and secondary pneumonic plague, when bubonic or septicemic infection seeds the lungs hematogenously. The disease is fundamentally an infectious, not a genetic, disorder; there are no causal human genes, and the "genetic" story of the disease lies almost entirely in the pathogen's plasmid- and chromosome-encoded virulence arsenal.

The mechanistic core of pneumonic plague is a stealth-then-storm immunopathology. In the first 24–36 hours, Y. pestis deploys a coordinated set of virulence factors — the Pla plasminogen-activating protease, the F1 capsular antigen, and the type III secretion system (T3SS) with its Yop effectors and the LcrV (V) antigen — to actively suppress innate immunity. LcrV signals through TLR2/CD14 to induce IL-10, dampening TNFα and IFNγ; the T3SS blocks leukocyte synthesis of the chemoattractant leukotriene B4 (LTB4); and F1 plus the T3SS render the organism resistant to phagocytosis. This anti-inflammatory window lets bacteria replicate freely in the airways. By ~48 hours the disease flips to an overwhelming pro-inflammatory state, producing a purulent, multifocal, necrotizing exudative bronchopneumonia and death by day 3 in the mouse model that faithfully mirrors human disease.

Clinically, the outlook is entirely dictated by the speed of antibiotic therapy. US surveillance across 1942–2018 showed mortality of 9% with high-efficacy antimicrobials (aminoglycosides, tetracyclines, fluoroquinolones) versus 51% with only limited-efficacy therapy; a 2020–2024 Madagascar randomized trial established that oral ciprofloxacin monotherapy is noninferior to aminoglycoside-ciprofloxacin combination. The disease is a flea-borne rodent zoonosis maintained in sylvatic foci, with Madagascar accounting for ~75% of global cases; the 2017 urban Madagascar epidemic (78% pneumonic) demonstrated the epidemic potential of person-to-person spread. There is no licensed vaccine, though F1/LcrV-based candidates (subunit, adenoviral-vectored, mRNA-LNP, nanolipoprotein) confer 90–100% protection in animal models. Early antibiotics and post-exposure prophylaxis (doxycycline or ciprofloxacin for 7 days) remain the mainstays of control.


Disease Information

Overview. Pneumonic plague is a fulminant bacterial pneumonia caused by Yersinia pestis, a non-motile, Gram-negative coccobacillus of the family Yersiniaceae. It is one of three overlapping clinical presentations of plague. Pneumonic plague is distinguished by (i) its respiratory localization, (ii) its capacity for direct human-to-human aerosol transmission, and (iii) the highest case-fatality of the three forms.

Key identifiers. - MONDO: MONDO:0001024 (pneumonic plague) - Causative organism: Yersinia pestis — NCBITaxon:632 - ICD-10: A20.2 (pneumonic plague); parent A20 (plague) - ICD-11: 1B93.1 (pneumonic plague) - MeSH: D010930 ("Plague"); pneumonic form indexed under plague - SNOMED CT: Pneumonic plague (disorder)

Synonyms / alternative names. Pulmonary plague; plague pneumonia; lung plague; "the Black Death" (historical, non-specific); pest (older usage). Primary vs. secondary pneumonic plague denote route of lung involvement.

Data source type. This report is derived from aggregated disease-level resources — peer-reviewed primary literature, outbreak epidemiology (WHO/Madagascar/CDC), and controlled animal-model studies — rather than individual EHR patient data.


Etiology

Primary cause — infectious. The sole cause of pneumonic plague is infection with Yersinia pestis. This is not a genetic or multifactorial disease; there is no human causal gene. Primary pneumonic plague follows inhalation of infectious respiratory aerosols (from an infected human or, rarely, an animal such as a domestic cat); secondary pneumonic plague arises when bubonic or septicemic Y. pestis seeds the lungs.

Risk factors (environmental / behavioral). - Exposure to sylvatic reservoirs and vectors — living or working in endemic rural foci with rodent–flea cycles. Y. pestis is "primarily a rodent-associated, flea-borne zoonosis maintained in sylvatic foci throughout western North America" (PMID: 23590319). - Close contact with a pneumonic plague case — "Human-to-human transmission of the pathogen occurs primarily through aerosol droplets" (PMID: 25643450), so household members and healthcare workers are at elevated risk. - Handling infected animals, including hunting/skinning rodents and lagomorphs, and exposure to sick domestic cats. - Crowding and urbanization, which amplified the 2017 Madagascar epidemic (burial practices, movement of people, overcrowding) (PMID: 30632956). - Pregnancy as a state of increased infection severity requiring special management (PMID: 32435804). - Deliberate release (bioterrorism) — aerosolized Y. pestis is a Tier-1 select agent.

Genetic risk / protective factors (human host). No validated human susceptibility or protective loci are established for pneumonic plague. One speculative, unproven hypothesis in the recent literature considers whether immune-tuning variants such as TYK2 P1104A could influence responses to pneumonic plague, but this is explicitly framed as speculation (PMID: 42382739). This should be treated as a knowledge gap, not an established fact.

Gene–environment interactions. Not applicable in the classical host-genetics sense. The operative "gene–environment" axis is the pathogen's genome × host environment: temperature-regulated virulence gene expression (F1 and T3SS are induced at 37°C, the mammalian host temperature) is the key switch that converts a flea-adapted organism into a mammalian pathogen.


Phenotypes

Pneumonic plague presents as an acute, fulminant febrile respiratory illness. Onset is adult and all-age (no age restriction), acute, and severe/progressive in essentially all untreated patients. Frequencies below are qualitative/clinical-series–based.

Phenotype Type HPO suggestion Characteristics
High fever, chills Symptom/sign HP:0001945 (Fever) Near-universal; abrupt onset
Cough Symptom HP:0012735 (Cough) Common; progresses rapidly
Hemoptysis / bloody sputum Sign HP:0002105 (Hemoptysis) Classic; blood-tinged/watery sputum
Dyspnea Symptom HP:0002094 (Dyspnea) Rapidly worsening respiratory distress
Chest pain Symptom HP:0100749 (Chest pain) Frequent
Pneumonia / bronchopneumonia Clinical sign HP:0002090 (Pneumonia) Purulent, multifocal, exudative
Headache, malaise Symptom HP:0002315 (Headache) Early prodrome
Sepsis / shock Sign HP:0100806 (Sepsis) Terminal, with multi-organ failure
Altered consciousness Sign HP:0011446 (Abnormal consciousness) Late/terminal
Leukocytosis Lab abnormality HP:0001974 (Leukocytosis) With neutrophilia
Disseminated intravascular coagulation Lab/clinical HP:0005521 (DIC) Terminal complication

Severity and progression. Uniformly severe and progressive. The classic course runs from a nonspecific febrile prodrome to fulminant pneumonia with respiratory failure and death within 3–7 days if untreated. The mouse intranasal model produces "a purulent multifocal severe exudative bronchopneumonia that closely resembles the disease observed in humans" (PMID: 16306265).

Atypical presentations. During the 2017 Madagascar urban outbreak, atypical features (prolonged illness, prominent upper-respiratory symptoms) were reported, complicating recognition (PMID: 32274983).

Quality-of-life impact. As an acute, life-threatening illness measured in days, pneumonic plague's "QoL" burden is dominated by acute mortality and, in survivors treated early, generally full recovery. Chronic disability is not a characteristic feature; there are no established EQ-5D/SF-36 datasets specific to plague survivors (knowledge gap).


Genetic / Molecular Information

Human genetics: not applicable. Pneumonic plague has no causal human genes, no pathogenic germline/somatic variants, no modifier genes, no chromosomal abnormalities, and no disease-defining epigenetic signature in the host. It is an acquired infectious disease. Sections that would apply to a Mendelian disorder (ACMG variant classification, gnomAD allele frequencies, COSMIC somatic mutations, karyotyping) are not applicable.

Pathogen genetics (the operative "molecular information"). Virulence is encoded across the Y. pestis chromosome and three plasmids:

Locus / gene Location Product & role
pla pPCP1 / pPst (~9.5 kb) Plasminogen-activator protease; essential for primary pneumonic plague
caf1 (+ caf1M/caf1A) pFra / pMT1 (~100 kb) F1 capsular antigen (Caf1, 15.5 kDa) via chaperone–usher assembly; antiphagocytic
ymt pFra / pMT1 Murine toxin / phospholipase D; flea-gut survival
lcrV, yop genes, ysc pCD1 / pYV (~70 kb) T3SS injectisome, LcrV (V antigen), Yop effectors
hms locus (hmsHFRS, hmsT/P) Chromosome (pgm/pigmentation) Biofilm; flea proventricular blockage
pgm locus / yop-ysc Chromosome / pCD1 Upregulated in vivo during lung infection

Environmental Information


Mechanism / Pathophysiology

Ordered causal chain (initiating event → clinical manifestation)

  1. Inhalation of Y. pestis aerosol (primary) — or hematogenous seeding of the lung from bubonic/septicemic infection (secondary) — delivers bacteria to the alveolar space. [established]
  2. Shift to 37°C host temperature induces expression of F1 capsule and the T3SS/LcrV system. [established]
  3. The T3SS injects Yop effectors into host leukocytes, which normally recognize the T3SS and respond by synthesizing leukotriene B4 (LTB4); Yop effectors actively block LTB4 synthesis, leading to failure of neutrophil chemoattraction. [established — P38271464]
  4. Secreted LcrV signals via TLR2/CD14 to induce IL-10, which results in suppression of pro-inflammatory TNFα and IFNγ. [established — P12391013 P11801671]
  5. F1 capsule + T3SS together confer resistance to phagocytosis, allowing extracellular bacterial survival and replication. [established — P11854232]
  6. Pla protease remodels the host fibrinolytic/hemostatic environment in the airway, enabling rapid bacterial replication and fulminant pneumonia; without Pla, inflammation aborts and lung repair activates. [established — P17255510]
  7. Steps 3–6 jointly produce an early anti-inflammatory "stealth" window (0–36 h) with unchecked bacterial proliferation. [established — P16306265]
  8. Rising bacterial burden crosses a threshold that flips the response to a highly pro-inflammatory state (~48 h), causing massive neutrophil influx, purulent multifocal exudative bronchopneumonia, and alveolar destruction. [established — P16306265]
  9. Bacteremic dissemination → septicemia, endotoxin/cytokine-driven shock, DIC, and multi-organ failure, resulting in death by ~day 3 (model) / 3–7 days (human) if untreated. [established]
 Aerosol inhalation
        │  (37°C induces F1, T3SS/LcrV)
        ▼
 ┌─────────── EARLY "STEALTH" PHASE (0–36 h) ───────────┐
 │  T3SS/Yop ──┤ blocks LTB4  ──► no neutrophil recruit  │
 │  LcrV ─TLR2/CD14─► IL-10 ──► ↓TNFα ↓IFNγ              │
 │  F1 + T3SS ──► antiphagocytic ──► extracellular growth│
 │  Pla ──► fibrinolysis, rapid airway replication       │
 └───────────────────────────┬──────────────────────────┘
                              │  bacterial burden threshold
                              ▼
 ┌────────── LATE "STORM" PHASE (~48 h → death) ─────────┐
 │  Overwhelming pro-inflammatory response               │
 │  Purulent multifocal exudative bronchopneumonia       │
 │  Dissemination ► septicemia ► shock ► DIC ► MOF ► death│
 └───────────────────────────────────────────────────────┘

Detail by category


Anatomical Structures Affected


Temporal Development


Inheritance and Population (Epidemiology)


Diagnostics

Clinical/microbiological. Definitive diagnosis rests on isolating Y. pestis or detecting its antigens/DNA from sputum, blood, or bronchoalveolar specimens.

Differential diagnosis: community-acquired bacterial pneumonia, inhalational anthrax, tularemia pneumonia, influenza/severe viral pneumonia, hantavirus pulmonary syndrome, melioidosis, and Q fever. Distinguishing features: rapid progression, hemoptysis, epidemiologic exposure, and F1 antigen positivity. Co-infection can occur — a case of pneumonic plague with nosocomial MDR Stenotrophomonas maltophilia has been reported (PMID: 29843675).

Omics/genetic testing of the host: not applicable diagnostically.


Outcome / Prognosis

Mortality is the defining outcome and is dominated by treatment timing.

Population / setting Case-fatality Source
Untreated pneumonic/septicemic ~100% PMID: 25643450
US, high-efficacy antimicrobials 9% PMID: 32435801
US, limited-efficacy therapy only 51% PMID: 32435801
2017 Madagascar, treated (observed) ~25% PMID: 32274983
2017 Madagascar, confirmed cases 25% (8/32) PMID: 30930106

Prognostic factors: time from symptom onset to effective antibiotic (single most important), antimicrobial class (aminoglycosides/tetracyclines/fluoroquinolones favorable), pregnancy, co-infection, and access to care. Recovery potential: with early appropriate antibiotics and supportive care, full recovery is expected; chronic sequelae are not characteristic. Complications: ARDS, septic shock, DIC, multi-organ failure, secondary/opportunistic infection.


Treatment

Antibiotics are the definitive therapy; speed is decisive. NCIT term suggestions in brackets.

Pharmacogenomics: no plague-specific host pharmacogenomic guidance. Standard aminoglycoside ototoxicity considerations (e.g., MT-RNR1 variants) are general, not plague-specific.

Antimicrobial resistance. Resistance is rare but a documented threat: the 1995 Madagascar isolate IP275 carried a self-transmissible IncA/C plasmid (pIP1202) "that conferred resistance to many of the antimicrobials recommended for plague treatment and prophylaxis" (PMID: 17375195); independent streptomycin- and doxycycline-resistance plasmids have also been found (PMID: 29030266). However, a survey found "no resistance in 392 Y. pestis isolates from 17 countries to eight antimicrobials used for treatment or prophylaxis of plague" (PMID: 22024826).


Prevention


Other Species / Natural Disease


Model Organisms


Key Findings (with statistical evidence)

1. Pla protease is essential for primary pneumonic plague

The Y. pestis outer-membrane omptin protease Pla (plasminogen activator, encoded on pPCP1/pPst) is indispensable specifically for the pneumonic form. In mouse intranasal infection, "the plasminogen activator Pla is essential for Y. pestis to cause primary pneumonic plague but is less important for dissemination during pneumonic plague than during bubonic plague" and "Pla allows Y. pestis to replicate rapidly in the airways, causing a lethal fulminant pneumonia; if unexpressed, inflammation is aborted, and lung repair is activated" (PMID: 17255510). This makes Pla the pivotal airway-replication switch and a rational target. Notably, its interaction with the host substrate α2-antiplasmin appears not to be the operative in-vivo mechanism (PMID: 26438794).

2. Disease follows a biphasic anti-inflammatory → pro-inflammatory course

The mouse model reveals "a strikingly biphasic syndrome, in which the infection begins with an antiinflammatory state in the first 24-36 h that rapidly progresses to a highly proinflammatory state by 48 h and death by 3 days," with mice succumbing to "a purulent multifocal severe exudative bronchopneumonia that closely resembles the disease observed in humans" (PMID: 16306265). In vivo the yop-ysc T3SS and the chromosomal pgm locus are upregulated. This kinetic — stealth then storm — is the organizing principle of the pathophysiology.

3. The T3SS suppresses leukotriene B4 to evade early immunity

Leukocytes normally sense the T3SS to trigger LTB4-driven neutrophil recruitment, but Y. pestis actively blocks it: "we demonstrate that leukocytes recognize the T3SS to initiate the rapid synthesis of LTB4," and "exogenous administration of LTB4 prior to infection limited bacterial proliferation, suggesting that the absence of LTB4 synthesis during plague contributes to Y. pestis immune evasion" (PMID: 38271464). This identifies a druggable early-immune-evasion node.

4. LcrV drives TLR2/CD14–IL-10 immunosuppression

Secreted LcrV induces the anti-inflammatory cytokine IL-10 and suppresses TNFα/IFNγ: "recombinant LcrV signals in a CD14- and toll-like receptor 2 (TLR2)-dependent fashion leading to immunosuppression by interleukin 10 induction" (PMID: 12391013). The suppression is IL-10-dependent — "TNF-alpha suppression was absent in LcrV-treated macrophages of IL-10-deficient (IL-10-/-) mice" (PMID: 11801671) — and IL-10−/− mice are highly resistant to Yersinia. A de-immunomodulated LcrV variant (rV10) retains protection with reduced IL-10 induction (PMID: 16041032).

5. F1 capsule + T3SS jointly make Y. pestis antiphagocytic

"F1 is encoded by the caf1 gene located on the large 100-kb pFra plasmid, which is unique to Y. pestis," and "F1 and the virulence plasmid-encoded type III system act in concert to make Y. pestis highly resistant to uptake by phagocytes"; a strain lacking both was phagocytosed ~95% (PMID: 11854232). F1 is both a virulence factor and the key diagnostic/vaccine antigen.

6. F1 antigen rapid diagnostic test enables bedside diagnosis

The lateral-flow RDT "detected concentrations of F1 antigen as low as 0.5 ng/mL in up to 15 min, and had a shelf life of 21 days at 60 degrees C. Its sensitivity and specificity were both 100%," outperforming bacteriology and ELISA on clinical specimens (PMID: 12547544) — transformative for endemic, resource-limited settings.

7. Antibiotics transform prognosis; ciprofloxacin monotherapy is sufficient

US surveillance: "Mortality differed significantly among those receiving high-efficacy therapy (9%) and only limited-efficacy therapy (51%)," with aminoglycosides and tetracyclines associated with survival (PMID: 32435801). The Madagascar RCT: "Ciprofloxacin monotherapy was noninferior to aminoglycoside-ciprofloxacin therapy" (PMID: 40768716). PEP is doxycycline or ciprofloxacin for 7 days (PMID: 15677847).

8. F1/LcrV immunity protects via antibodies + Th1 cytokines

Protection by the F1-V vaccine requires TNFα and IFNγ (PMID: 20840834); YopE(69-77) is a protective CD8 epitope (PMID: 21653834); next-gen adenoviral, mRNA-LNP, and nanolipoprotein platforms give 90–100% protection against aerosol challenge (PMID: 41736398, PMID: 40279638, PMID: 40642079).

9. Transmission ecology: flea-borne zoonosis + biofilm + early-phase transmission

Y. pestis is "primarily a rodent-associated, flea-borne zoonosis maintained in sylvatic foci" (PMID: 23590319); pneumonic spread is airborne (PMID: 25643450). Flea transmission occurs by hms-biofilm proventricular blockage — "Yersinia pestis biofilm formation causes massive adsorption of haemin or Congo red in vitro as well as colonization and eventual blockage of the flea proventriculus in vivo" (PMID: 17074909) — and by a complementary biofilm-independent early-phase mechanism, where "Biofilm-defective mutants transmitted... as efficiently as the parent strain, whereas the EPT efficiency of fleas fed the biofilm-overproducing strain was significantly less" (PMID: 20395271).

10. Epidemic potential: 2017 Madagascar

78% of the 2414 suspected cases were pneumonic (PMID: 30930106), with >20 independent introductions from rural foci (PMID: 38270131) and atypical presentations complicating diagnosis (PMID: 32274983).

11. MDR is rare but a documented, transmissible threat

The first MDR isolate (IP275, 1995) carried a self-transmissible plasmid conferring resistance to many recommended antimicrobials (PMID: 17375195), yet a 392-isolate, 17-country survey found no resistance to eight anti-plague antimicrobials (PMID: 22024826).


Mechanistic Model / Interpretation

Pneumonic plague is best understood as a race between bacterial immune subversion and the host's ability to mount protective inflammation — a race the pathogen almost always wins unless antibiotics intervene early. The virulence factors are not redundant; they attack complementary arms of innate defense:

Virulence factor Genetic locus Immune arm neutralized Net effect
Pla protease pPCP1 Fibrinolytic control of airway Rapid airway replication
T3SS/Yop pCD1 LTB4 → neutrophil recruitment No early neutrophil influx
LcrV pCD1 TLR2/CD14 → IL-10 → TNFα/IFNγ Global cytokine suppression
F1 capsule pFra (caf1) Phagocytosis Extracellular survival
hms biofilm chromosome (vector stage) Flea transmission

The early anti-inflammatory phase is the therapeutic window and the reason mortality is so exquisitely time-dependent: once bacterial burden crosses the threshold that triggers the late cytokine storm, tissue destruction, sepsis, and DIC become self-sustaining and antibiotics can no longer reverse the trajectory. This model explains three clinical observations at once: (1) near-100% untreated lethality (unopposed early evasion), (2) the dramatic mortality drop with early high-efficacy antibiotics, and (3) why vaccines that restore/require TNFα and IFNγ (or that provide neutralizing anti-F1/anti-LcrV antibody before challenge) are protective — they pre-empt the very axes the pathogen suppresses.


Evidence Base

PMID Contribution
17255510 Pla essential for primary pneumonic plague (mouse)
16306265 Biphasic immunopathology; validated mouse model
38271464 T3SS suppression of LTB4 immune evasion
12391013 LcrV → TLR2/CD14 → IL-10 axis
11801671 IL-10-dependence of V-antigen TNFα suppression
16041032 rV10 de-immunomodulated vaccine antigen
11854232 F1 + T3SS antiphagocytic synergy
12547544 F1 rapid diagnostic test performance
32435801 US mortality 9% vs 51% by antimicrobial efficacy
40768716 Ciprofloxacin monotherapy noninferiority RCT
15677847 PEP with doxycycline/ciprofloxacin 7 days
25643450 Untreated CFR ~100%; airborne transmission
30930106 2017 Madagascar epidemiology; global distribution
38270131 Multiple introductions in 2017 epidemic
32274983 Atypical presentations; observed vs expected mortality
20840834 TNFα/IFNγ required for F1-V protection
21653834 YopE(69-77) protective CD8 epitope
41736398 / 40279638 / 40642079 Next-gen vaccine platforms
17375195 / 22024826 / 29030266 MDR plasmids vs. low resistance prevalence
23590319 / 17074909 / 20395271 Flea vector ecology & transmission mechanisms
21628541 / 21486959 Intracellular & pharmacodynamic antibiotic efficacy
39808829 US endemicity & weather-linked variation

Limitations and Knowledge Gaps

  1. Host genetics essentially unknown. No validated human susceptibility/protective loci exist; the TYK2 P1104A link to pneumonic plague is explicitly speculative (PMID: 42382739). Host GWAS in endemic populations are lacking.
  2. Model-organism reliance. Much mechanistic detail (biphasic kinetics, Pla essentiality, LTB4 suppression) derives from mouse and in-vitro systems; direct human tissue confirmation is limited by the disease's rarity and lethality. Analgesia confounds animal readouts (PMID: 42212153).
  3. Epidemiologic uncertainty. Outbreak counts include clinically suspected cases with overdiagnosis; observed mortality figures (e.g., 2017 Madagascar) are confounded by widespread community antibiotic use (PMID: 32274983).
  4. No licensed vaccine. Despite strong animal efficacy, no F1/LcrV vaccine is licensed; correlates of protection in humans are undefined.
  5. Resistance surveillance. MDR is rare but plasmid-mediated resistance is proven and transmissible; ongoing genomic surveillance is essential (PMID: 17375195).
  6. Long-term outcomes/QoL in survivors are essentially uncharacterized.

Proposed Follow-up Experiments / Actions

  1. Human genetic susceptibility study — GWAS/immunogenetic study in Madagascar endemic populations to test host-modifier hypotheses (including TYK2, IL-10 pathway variants).
  2. Host-directed adjunctive therapy trials — test whether IL-10 blockade, or early exogenous LTB4/TNFα/IFNγ restoration, augments antibiotics in animal models, exploiting the identified evasion nodes (PMID: 38271464, PMID: 12391013).
  3. Advance a de-immunomodulated LcrV (rV10)-based mRNA/adenoviral vaccine toward human Phase I, defining human correlates of protection (PMID: 16041032, PMID: 40279638).
  4. Deploy and evaluate point-of-care CRISPR/isothermal NAATs alongside F1 RDTs for earlier confirmation in endemic settings (PMID: 42413880).
  5. Strengthen genomic AMR surveillance in Madagascar and other foci to detect emerging transmissible resistance plasmids (PMID: 22024826, PMID: 29030266).
  6. Operational research on time-to-antibiotic — given the 9% vs 51% mortality gradient, quantify and shorten the symptom-onset-to-treatment interval in endemic health systems (PMID: 32435801).

Report compiled from 14 confirmed findings and 53 reviewed papers. Evidence types span human clinical/epidemiologic studies, mouse and non-human primate models, in-vitro cellular assays, and computational/genomic analyses, as annotated per claim.