Six tiers · never blended · last evidence search 19 September 2026
Ewingella americana evidence ledger: every tier, kept separate
The reason most accounts of this organism mislead is not that they are wrong. It is that they put a mouse experiment, a case report and a farm survey into one paragraph. Here they stay apart.
Each block below is one class of evidence. A finding in one class does not become stronger by sitting next to a finding in another. The strongest thing anyone can say about Ewingella americana and cancer is an animal result; the strongest thing anyone can say about it and human disease is a set of case reports; and there is nothing at all in the interventional-human tier.
1. The tiers on this site
| Tier | What it means here | What exists for this organism |
|---|---|---|
| Human interventional | a trial in which people were given the thing | Nothing. No registered study, no treated patient |
| Case report | what happened to individual patients, written up | 19 patients across 16 studies, plus later single cases |
| Animal | a controlled experiment in a living animal | one study, one tumour model, one laboratory |
| In vitro | cells or spheroids in culture | spheroid destruction within the same 2025 study |
| Genomic | sequence, gene content, enzymology | reference genome, resistome, virulome, Antarctic enzymes |
| Field | surveys of real populations, farms, foods | the largest body of work on this species |
2. Animal evidence
Iwata S, Yamasita N, Asukabe K, Sakari M, Miyako E. Gut Microbes 2025;17(1):2599562, PMID 41376334. A strain from the gut of Dryophytes japonicus, given as a single intravenous dose of 200 µL at 5 × 109 CFU/mL, produced a 100% complete response at day 30 in syngeneic Colon-26 BALB/c mice; ~3000-fold tumour enrichment between 3 and 24 hours; zero recovery from lung, liver, spleen, kidney or heart; 0/10 rechallenged cured mice grew tumours against 10/10 naïve controls; comparators anti-PD-L1 and liposomal doxorubicin underperformed it in the same experiment.
Weight: high internal quality, single source. Not replicated independently. One tumour model. Full breakdown →
3. Case-report evidence — the complete human record
Forty-two years of human medicine with this organism amounts to a few dozen patients. That is the honest denominator.
| Year | Presentation | Identifier |
|---|---|---|
| 1983 | polymicrobial bacteraemia — and the species description itself | PMID 6630449; PMID 6847036 |
| 1985–87 | pseudobacteraemia outbreaks from contaminated blood-collection tubes | PMID 4019740; PMID 3571454; PMID 3571455 |
| 1986 | nosocomial bacteraemia in intensive care; wound colonisation | PMID 3942442; PMID 3958154 |
| 1991–92 | conjunctivitis; septicaemia | PMID 1794367; PMID 1400980 |
| 1999–2001 | CAPD peritonitis; keratoconjunctivitis; catheter-related bacteraemia | PMID 10523587; PMID 11063558; PMID 11298155 |
| 2003–05 | fatal Waterhouse–Friderichsen syndrome; pneumonia in chronic renal failure | PMID 12604997; PMID 15716620 |
| 2007–08 | multidrug-resistant isolates | PMID 17956962; PMID 18626541 |
| 2012–14 | acute conjunctivitis; peritonitis on peritoneal dialysis; sepsis in an infant with congenital nephropathy | PMID 22998132; PMID 24594296; PMID 24444748 |
| 2019–20 | infection in a healthy 4-year-old; first meningitis in a term newborn; PD peritonitis | PMID 31060497; PMID 32596194; PMID 33363216 |
| 2022–24 | PD-related peritonitis; first documented urinary tract infection; Andersson lesion detected by metagenomic sequencing | PMID 35223116; PMID 37009357; PMID 39033154 |
| 2025 | sepsis during transfusion in a chemotherapy patient, Istanbul | PMID 40636648 |
What the review concluded: median patient age 55, 47.4% male; bloodstream, respiratory and peritoneal infections commonest; resistance to cephalosporins, aminoglycosides and trimethoprim–sulfamethoxazole minimal; overall mortality one patient. The authors also record that no included study reported the genetic or molecular basis of this organism’s resistance — a gap they name explicitly.
4. Genomic and biochemical evidence
The reference genome
Strain B6-1 from Flammulina filiformis: 4.67 Mb, 4,301 predicted genes, 53.80% GC. Common gene sets for secretion systems, virulence effectors, CAZymes and toxins across every host examined; antibiotic resistance and pigment genes for evading host defence; mobile genetic elements as the main evolutionary driver. Strain origin did not significantly affect phylogeny. Liu Z et al., Pathogens 2020;9(5):330, PMID 32354059.
The Antarctic isolate
Isolate L47 (NCBI SAMN54554459) carries one L-arabinose isomerase (araA) and three β-galactosidases (bgaA, bglY, lacZ) — an uncommon combination in a single bacterium. Recombinant AraA is Mn2+-dependent and reached ~18% tagatose from 100 mM galactose in 48 h; BgaA was the most tractable lactose-hydrolysing candidate; AraA is the bottleneck. Rivero K et al., Int J Mol Sci 2026;27(2):1128, PMID 41596769.
Taxonomic position
Family placement refined over time: originally Enterobacteriaceae, now Yersiniaceae within Enterobacterales. Phylogenetic context in Spröer C et al., Int J Syst Bacteriol 1999, PMID 10555323; a related enterobacterial β-lactamase characterised in Lartigue MF et al., J Antimicrob Chemother 2013, PMID 23580557.
Natural susceptibility
Stock I et al., J Chemother 2003, PMID 14598935 — the natural antibiotic susceptibility pattern of Ewingella americana strains, which is the baseline against which every later resistance report should be read.
5. Field evidence — the largest body of work
This is the lane that popular coverage of the cancer result omits entirely, and it is where most of the published effort on this organism has gone.
| Where | Finding | Identifier |
|---|---|---|
| United Kingdom, 1996 | first isolation from cultivated Agaricus bisporus | PMID 8875916 |
| Spain, 2004 | prevalence in retail A. bisporus, Lentinula edodes, Pleurotus ostreatus, Zaragoza | PMID 19712317 |
| Korea, 2009 | internal stipe necrosis of cultivated button mushrooms | PMID 23983509 |
| China, 2019–21 | microflora of fresh button mushrooms in cold storage; type III secretion system re-evaluated in soft rot of Pleurotus eryngii | PMID 30883770; PMID 33123750 |
| Iran, 2023 | 38 farms, 15 provinces, 40 strains; six BOX-PCR clusters; pathogenic strains clustering with non-pathogenic environmental strains | PMID 36131391 |
| China, 2024–25 | bacterial communities of needle mushroom production; volatile organic compounds toxic to Pleurotus pulmonarius | PMID 39741595; PMID 41003513 |
| Various | insect gut (tomato growth promotion), Brazil pine roots, molluscs, Anopheles sporogony inhibition, blown-pack meat spoilage, herbal products | PMID 36362334; PMID 21596540; PMID 7580799; PMID 8375488; PMID 17870199; PMID 18680952 |
6. Absences, recorded after searching
An absence is only worth writing down if somebody went and looked. These were looked for on 19 September 2026.
No human trial
No registered interventional study of Ewingella americana as a therapeutic in any indication.
No Chinese-language antitumour work
CNKI, Wanfang and CQVIP return edible-mushroom bacterial-disease literature. The substantial Chinese contribution to this organism — Shenyang Agricultural University, Jilin Agricultural University — is published in English and is cited above.
No Russian-language work naming it
CyberLeninka and eLIBRARY carry general Russian-language literature on bacterial and microbiota-based tumour therapy, faecal microbiota transplantation and oncolytic virology, but returned nothing naming Ewingella.
No molecular resistance mechanism
Named as a gap by the 2024 narrative review itself: no included study reported the genetic or molecular basis of this organism’s antimicrobial resistance.
