Cherubism — Comprehensive Disease Research Report

Quick orientation before the deep dive: Cherubism is a rare, mostly self-limiting childhood bone disease where the jaw quietly eats itself and refills with fluffy, giant-cell-rich fibrous tissue. The wild part is the biology — a single-letter change in an adapter protein makes it un-killable by the cell's normal disposal crew, and the whole thing behaves less like a bone tumor and more like a slow-burning autoinflammatory tantrum aimed at the jaws. Then, in most kids, it just... turns itself off around puberty. Below is everything the literature has to say, section by section.

1. Disease Information

Cherubism is a rare, benign, self-limiting fibro-osseous disorder in which bone of the mandible and maxilla is resorbed and replaced by expansile, soft, fibrous giant-cell-rich tissue, producing bilateral, symmetric jaw swelling. The name comes from the resulting round-cheeked face plus the upturned-eye look ("eyes raised to heaven") when the orbital floor is involved — reminiscent of cherubs in Renaissance art (first named by Jones in 1933).

Key identifiers: - OMIM: #118400 (Cherubism); causal gene SH3BP2 is 602104 - MONDO: MONDO:0007038 (suggested — curator should confirm via OAK) - Orphanet: ORPHA:184 - MeSH: D002636 (Cherubism) - ICD-10: commonly coded K10.8 ("Other specified diseases of jaws"); ICD-11 near LB43/DA0D range (coding is inconsistent across sources — flag for verification)*

Synonyms / alternative names: familial fibrous dysplasia of the jaws, familial multilocular cystic disease of the jaws, familial bilateral giant cell tumor of the jaw, cherubinism, bilateral giant cell tumor. (Note: "familial fibrous dysplasia" is a misnomer — cherubism is molecularly distinct from GNAS-driven fibrous dysplasia.)

Data provenance: The disease-level knowledge is drawn from aggregated resources (OMIM, Orphanet, GeneReviews) and a literature base of ~300–600 published cases/case series worldwide — not from large EHR cohorts. Individual-patient granularity comes from case reports and small family studies.

Sources: OMIM 118400, GeneReviews NBK1137, Orphanet

2. Etiology

Primary cause — genetic: - SH3BP2 gain-of-function (≈80% of cases): heterozygous missense variants clustered in exon 9, within the 6-residue motif RSPPDG (p.Arg415–Gly420). Autosomal dominant. (Ueki et al., Nat Genet 2001, PMID:11381256 — "identified mutations in the SH3BP2 gene… All mutations were in exon 9 and affected 3 amino acids within a 6-amino acid sequence (RSPPDG).") - OGFRL1 biallelic loss-of-function (rare, recessive): two consanguineous families (Syria, India) with homozygous loss-of-function OGFRL1 variants; autosomal recessive. A newly recognized second locus. (JBMR Plus 2024, ziae050, PMC11062026) - ~20% of clinically classic cases have no identified SH3BP2 variant — genetic heterogeneity remains.

Genetic risk factors: having a pathogenic SH3BP2 allele is essentially deterministic (high penetrance); no established polygenic/susceptibility loci. De novo mutations account for a substantial share of simplex cases.

Environmental risk / trigger factors: Cherubism is genetically driven, but disease expression and severity are modulated by inflammatory challenge. In heterozygous cherubism mice, oral microbial burden (periodontal infection) dramatically worsens alveolar bone destruction — "microbe-dependent exacerbated alveolar bone destruction in heterozygous cherubism mice" (PMC7285758). This maps to the human observation that lesions flare with dental eruption, trauma, and infection. There is also anecdotal exacerbation with tooth extraction/surgery.

Protective factors: No genetic protective alleles described. Empirically, the strongest "protective" force is puberty/aging itself — most lesions spontaneously regress after adolescence. Avoiding elective jaw surgery during the active proliferative phase is considered protective against flare.

Gene–environment interaction: The unifying model is that mutant SH3BP2 lowers the threshold for a myeloid inflammatory response, so ordinary physiologic/microbial challenges to the jaw (tooth eruption, oral flora) that a normal jaw shrugs off instead ignite a self-amplifying TNF-α/RANKL loop. This explains both jaw-restriction (the tooth-bearing bones face the most microbial/eruption challenge) and the age-limited course.

3. Phenotypes

Craniofacial / skeletal (core, near-universal): - Bilateral, symmetric mandibular and/or maxillary swelling — clinical hallmark; onset typically age 2–5 yr, progresses to puberty. Frequency ~100%. Suggested HPO: HP:0000303 (Mandibular prognathia), HP:0012802 (Abnormal maxilla morphology), plus round/full cheeks. - Multilocular radiolucent, expansile jaw lesions with cortical thinning, at the mandibular angles/rami; condyles usually spared. - "Eyes to heaven" appearance / exposed inferior sclera / upward globe tilt when infraorbital rim and orbital floor are involved. Suggested HPO: HP:0000520 (Proptosis) and exposure of sclera.

Dental (very frequent): - Displaced, unerupted, ectopic, hypoplastic, or absent teeth; premature exfoliation of primary teeth; malocclusion. Suggested HPO: HP:0000668 (Hypodontia), HP:0000689 (Dental malocclusion), HP:0006480/HP:0006349 (abnormal tooth morphology), premature tooth loss (HP:0006480).

Regional / soft tissue: - Submandibular and cervical lymphadenopathy — common in early active disease, tends to regress. Suggested HPO: HP:0002716 (Lymphadenopathy).

Functional complications (severe cases): - Obstructive sleep apnea / upper-airway obstruction — GeneReviews: "Respiratory manifestations can include obstructive sleep apnea and upper-airway obstruction." Suggested HPO: HP:0002870 (Obstructive sleep apnea). - Visual/ophthalmologic compromise (proptosis, diplopia, rarely optic involvement) with severe maxillary/orbital disease. Suggested HPO: HP:0000505 (Visual impairment), HP:0000651 (Diplopia). - Speech, chewing, and swallowing difficulty; psychosocial impact from facial disfigurement.

Characteristics summary: Onset early childhood; severity highly variable (mild grade-1 to grossly disfiguring grade-3); course progressive then regressive (grows to puberty, stabilizes, involutes in 2nd–3rd decade). Intellect and general development are normal.

Quality-of-life impact: Main burdens are cosmetic disfigurement (psychosocial, especially school-age), functional (mastication, speech, vision, sleep/airway), and — in a subset — surgical morbidity. No cherubism-specific validated QoL instrument; generic pediatric craniofacial QoL tools apply.

Sources: GeneReviews, Cherubism: best clinical practice, Orphanet J Rare Dis 2012;7(Suppl 1):S6

4. Genetic / Molecular Information

Causal gene 1 — SH3BP2 (SH3-domain binding protein 2; a.k.a. 3BP2) - Locus: chromosome 4p16.3; HGNC symbol SH3BP2 (HGNC:10825, lowercase hgnc: per repo convention — verify ID); OMIM 602104. - Mapping: linkage to 4p16.3 established by Mangion et al. (Am J Hum Genet 1999). (ScienceDirect) - Protein: 561-aa adapter with PH domain (membrane lipid binding), a proline-rich (PR) region with SH3-binding motifs, and a C-terminal SH2 domain*.

Pathogenic variants (SH3BP2): - Cluster in exon 9, within the RSPPDG motif (codons 415–420). - Recurrent variants: p.Pro418Arg (c.1253C>G) — most common; p.Pro418Leu, p.Arg415Gln/Pro, p.Gly420Glu/Arg, p.Pro416Arg, p.Asp419 changes. ~13 distinct variants reported (12 missense + 1 single-base deletion), ~80% in exon 9. - Variant type: overwhelmingly missense; germline; gain-of-function (not haploinsufficiency — Wolf-Hirschhorn 4p deletions that delete one SH3BP2 copy do not cause cherubism). - Allele frequency: absent/vanishingly rare in gnomAD (private/de novo pathogenic changes). - ACMG classification: recurrent RSPPDG missense variants are classified Pathogenic/Likely Pathogenic (strong functional + genetic evidence).

Causal gene 2 — OGFRL1 (opioid growth factor receptor-like 1) - Biallelic loss-of-function, autosomal recessive; homozygous frameshift/LoF variants "not reported in any variant databases." Represents a mechanistically distinct route to a cherubism-like phenotype. (JBMR Plus 2024)

Modifier genes: none formally validated; disease severity likely modified by inflammatory-response genetic background (inferred from mouse work) and sex.

Epigenetics / chromosomal abnormalities: No recurrent DNA-methylation/histone signature or large structural rearrangement is implicated. Cherubism is a point-mutation disease, not a copy-number/aneuploidy disorder. (Notably, 4p deletion removing SH3BP2 does not cause disease — reinforcing gain-of-function.)

Suggested ontology: gene GO annotations GO:0017124 (SH3 domain binding), GO:0035591 (signaling adaptor activity).

Sources: Nature Genetics 2001, OMIM 602104, GeneReviews

5. Environmental Information

• Environmental factors: No toxin/radiation/occupational cause. The relevant environmental input is local oral–microbial and mechanical challenge to the jaws. Mouse data show oral microbes drive exacerbated alveolar bone destruction in heterozygous cherubism mice (PMC7285758).
• Lifestyle factors: none established. Good oral hygiene / periodontal-inflammation control is biologically rational given the microbe-dependence data.
• Infectious agents: No single pathogen causes cherubism. Periodontal/oral bacterial burden acts as a modifier/trigger of severity, not an etiologic agent. (Suggested taxon anchor if modeled: oral microbiota, NCBITaxon of specific periopathogens not disease-defining.)

6. Mechanism / Pathophysiology

This is the mechanistically richest part, so let me lay out the causal chain from mutation to melted jawbone.

Upstream trigger — loss of protein disposal (the keystone lesion): Normally, the adapter protein 3BP2/SH3BP2 is kept on a short leash. Tankyrase (TNKS/TNKS2, a PARP-family enzyme) binds SH3BP2 at the RSPPDG (RxxPDG) motif, poly-ADP-ribosylates it, which flags it for the E3 ubiquitin ligase RNF146, which ubiquitylates it for proteasomal degradation. Cherubism mutations (R415G, P418L, P418R, G420R) destroy the tankyrase recognition site → SH3BP2 is no longer ADP-ribosylated, no longer ubiquitylated, and accumulates. So the defect isn't a broken protein — it's a protein the cell can't throw away. (Levaot et al., Cell 2011, PMID:22153076 — "Loss of Tankyrase-Mediated Destruction of 3BP2 Is the Underlying Pathogenic Mechanism of Cherubism"; Guettler et al., Cell 2011, companion paper on tankyrase substrate recognition.) Elegant corollary: tankyrase inhibitors phenocopy cherubism, inducing bone loss by accumulating SH3BP2 (PMC6406327).

Middle — two amplifying myeloid arms (from the Sh3bp2 P416R knock-in mouse; Ueki et al., Cell 2007, PMID:17218256): 1. Osteoclast arm (bone destruction): Stabilized SH3BP2 hyperactivates SYK/SRC/VAV and, downstream of RANKL–RANK, boosts PLCγ2 phosphorylation → IP₃ → Ca²⁺ release → calcineurin → NFATc1 nuclear translocation — NFATc1 being the master transcriptional switch for osteoclastogenesis. Result: more, more-active, bone-resorbing osteoclasts. (Mukai et al., JBMR 2014, doi:10.1002/jbmr.2295 — "SH3BP2 cherubism mutation potentiates TNF-α–induced osteoclastogenesis via NFATc1.") 2. Macrophage/inflammation arm (the engine): Mutant myeloid cells over-respond to M-CSF and RANKL, with elevated ERK1/2 and Syk (pTyr346) signaling via an autocrine feedback loop, driving excess TNF-α production and systemic macrophage inflammation.

Convergence & the self-amplifying loop: Hyperactive macrophages pump out TNF-α, which drives systemic inflammation, stimulates stromal cells to secrete RANKL and M-CSF, and feeds back to generate still more hyperactive osteoclasts → jaw bone resorption replaced by fibrous, giant-cell-rich tissue. Osteoblasts are also perturbed (excess immature osteoblasts, ~20% fewer mature ones, reduced osteoprotegerin/OPG → higher RANKL:OPG ratio further favoring resorption; PMID:20691350).

Genetic dissection of the causal chain (mouse epistasis): - Cross onto TNF-α–null → infiltrative lesions disappear, bone phenotype partially rescued → TNF-α is necessary for the inflammatory/infiltrative disease. - Cross onto M-CSF–deficient (op/op) → bone loss and infiltrates essentially gone (TNF-α still high) → M-CSF needed for the osteolytic output. - Cross onto NFATc1 conditional KO → skeletal phenotype fully rescued despite persistent high TNF-α → NFATc1 is the essential bone-resorption node, while TNF-α inflammation runs through a parallel, NFATc1-independent path.

Framing: The Reichenberger/Ueki review concludes cherubism is best understood as "a systemic autoinflammatory response to physiologic challenges despite the localized appearance of bone resorption" — a myeloid-cell disorder that happens to manifest in the jaws. (Orphanet J Rare Dis 2012;7(Suppl 1):S5, PMC3359958)

Why the jaws, and why self-limiting? Best current explanation: the tooth-bearing jaws face the greatest eruption/microbial/mechanical challenge in childhood; once dentition is complete and the pubertal hormonal/immune milieu shifts, the driving stimulus wanes and lesions ossify and regress.

Suggested ontology terms: - Biological processes (GO): GO:0030316 (osteoclast differentiation), GO:0045672 (positive regulation of osteoclast differentiation), GO:0045453 (bone resorption), GO:0032760 (positive regulation of TNF production), GO:0042116 (macrophage activation), GO:0033173 (calcineurin-NFAT signaling cascade), GO:0070371 (ERK1/ERK2 cascade), GO:0006471 (protein ADP-ribosylation), GO:0043161 (proteasome-mediated ubiquitin-dependent protein catabolic process), GO:0038095/RANK signaling. - Cell types (CL): CL:0000092 (osteoclast), CL:0000235 (macrophage), CL:0000576 (monocyte), CL:0000062 (osteoblast), multinucleated giant cell. - Molecular players/chemicals: TNF-α, M-CSF (CSF1), RANKL (TNFSF11), NFATc1, SYK, poly-ADP-ribose.

Sources: Levaot 2011 PMID:22153076, Ueki 2007 PMID:17218256, Mukai 2014 JBMR, pathophysiology review PMC3359958

7. Anatomical Structures Affected

Organ / structure level (primary): - Mandible — UBERON:0001684; especially the angle and ramus; symphysis and body variably; condyles typically spared. - Maxilla — UBERON:0002397; involvement drives orbital-floor/infraorbital-rim disease. - Jaw region overall — UBERON:0003278 (jaw region) / UBERON:0001710 (lower jaw region).

Secondary involvement: - Bony orbit / orbital floor (UBERON:0006800) → globe displacement, proptosis, "eyes to heaven." - Teeth / dentition (UBERON:0001091) → displacement, agenesis, malocclusion. - Cervical & submandibular lymph nodes (UBERON:0002429) → reactive lymphadenopathy. - Upper airway → obstruction/OSA in severe maxillary/mandibular expansion. - Rare extragnathic reports exist (e.g., ribs) but classic cherubism is jaw-restricted — a key diagnostic feature.

Body systems: skeletal (craniofacial), plus innate immune/myeloid system as the mechanistic driver; secondarily ophthalmic, respiratory, and dental systems.

Tissue / cell level: normal jaw bone tissue (UBERON:0002481) is replaced by fibrous connective tissue stroma rich in multinucleated giant cells (osteoclast-like) and spindle-shaped mesenchymal stromal cells. Target/effector cells: osteoclasts (CL:0000092) and macrophages (CL:0000235).

Subcellular / molecular compartment: the pathology localizes to cytoplasmic protein-degradation machinery — SH3BP2 accumulates in the cytoplasm because it escapes the proteasome (GO:0000502) after failing tankyrase-directed ADP-ribosylation/RNF146 ubiquitylation. Downstream signaling touches the plasma membrane (RANK/M-CSFR) and nucleus (NFATc1 translocation).

Localization / lateralization: bilateral and symmetric — this bilaterality is the classic feature distinguishing cherubism from the usually unilateral/solitary central giant cell granuloma.

8. Temporal Development

• Onset: early childhood, typically age 2–5 yr (some series note emergence in the 2nd year); insidious, painless progressive swelling.
• Progression / stages:
• Active/proliferative phase (early childhood → puberty): lesions grow, bone resorbs, swelling increases.
• Stabilization (adolescence): growth arrests.
• Involution/regression (2nd–3rd decade): lesions ossify, remodel, facial contour normalizes. GeneReviews: "By age 30 years, the facial abnormalities… are usually less obvious than during childhood."
• Course pattern: characteristically progressive-then-spontaneously-regressive — one of the few bone diseases that reliably self-corrects. Severity is graded (e.g., Raposo-Amaral/Motohashi/Seward grading systems, I–III, by anatomic extent).
• Duration: effectively a time-limited disease of childhood/adolescence, though residual bone deformity or dental sequelae can persist into adulthood.
• Remission: predominantly spontaneous with skeletal maturity; treatment-induced stabilization reported for aggressive cases.
• Critical window for intervention: the active phase — the rationale for deferring elective reconstructive surgery until after regression (to avoid provoking regrowth), while reserving early intervention for airway/vision-threatening disease.

9. Inheritance and Population

Epidemiology: - Prevalence: unknown/very rare; Orphanet class <1 in 1,000,000. ~300–600 cases reported worldwide across ethnic groups; no strong ethnic predilection. - Incidence: not established (too rare for reliable incidence figures).

Inheritance (genetic): - Autosomal dominant (SH3BP2, ~80% of molecularly solved cases). Many simplex cases are de novo. - Autosomal recessive (OGFRL1) in rare consanguineous families. - Penetrance: high; classically described as near-complete. Historically stated as "complete by age 5," though GeneReviews cautions it "has not been systematically studied"; some non-penetrant/very-mild carriers reported. One clinical–genetic series "found no evidence of non-penetrance." - Expressivity: markedly variable, even within a family. - Sex effect: older literature reported a male excess (~2:1), but this is now attributed to ascertainment; recent series report females on average more severely affected than males — an intriguing and clinically relevant reversal. - Genetic anticipation: not a feature (not a repeat-expansion disorder). - Germline/gonadal mosaicism: possible — reported basis for recurrence in apparently unaffected parents; relevant to counseling. - Founder effects / consanguinity: no SH3BP2 founder mutations; consanguinity is central to the recessive OGFRL1 families. - Carrier frequency: not defined (private mutations).

Population demographics: worldwide, pan-ethnic; onset in early childhood; no endemic geographic clustering. Recessive OGFRL1 form reported from consanguineous Middle Eastern/South Asian pedigrees.

Sources: GeneReviews, Orphanet best clinical practice, clinical & genetic analysis series (ResearchGate)

10. Diagnostics

Diagnostic approach: clinical + radiographic + (as needed) histologic pattern, confirmed by molecular testing.

Clinical / imaging: - Radiographs / CT (imaging is central): bilateral, symmetric, multilocular ("soap-bubble") radiolucencies, expansile remodeling, cortical thinning, at mandibular angles/rami; often displaced/unerupted "floating" teeth. Condyles spared. Suggested modality terms (RadLex/DICOM): panoramic radiograph, CT maxillofacial. - CBCT/MRI: delineate extent, airway, orbital involvement; MRI for soft-tissue characterization and follow-up.

Laboratory / biomarkers: - Serum calcium, phosphate, PTH are typically normal (helps exclude hyperparathyroidism/brown tumors). Alkaline phosphatase may be elevated during active resorption. No validated circulating diagnostic biomarker; TNF-α elevation is mechanistically expected but not a clinical test.

Biopsy / histopathology: - Fibrous stroma with numerous multinucleated osteoclast-like giant cells, hemorrhage, hemosiderin; microscopically indistinguishable from central giant cell granuloma (CGCG) — so histology alone cannot make the call; bilaterality + genetics distinguish it. (PMID:6937832) Perivascular eosinophilic cuffing around vessels is a classically cited (if inconsistent) clue.

Genetic testing: - First-line: targeted SH3BP2 exon 9 sequencing (single-gene), given the tight RSPPDG mutational hotspot. High yield for classic cases. - If negative: broader SH3BP2 sequencing, then exome/genome (to catch OGFRL1 recessive form or novel loci) — especially with consanguinity or atypical/recessive pedigrees. - Gene panels (bone dysplasia/giant-cell-lesion panels) where available; CMA/karyotype/FISH not indicated (no CNV mechanism); mtDNA/repeat testing N/A.

Clinical criteria / differential diagnosis — key mimics to exclude: - Central giant cell granuloma (usually unilateral/solitary). - Fibrous dysplasia / McCune-Albright (GNAS; ground-glass, often unilateral, extragnathic). - Hyperparathyroidism-jaw tumor syndrome (CDC73/HRPT2; abnormal calcium/PTH), brown tumors of hyperparathyroidism. - Aneurysmal bone cyst, giant cell tumor, Noonan/RASopathy-associated giant-cell lesions, and multiple giant cell lesion syndromes.

Screening: no population screening (too rare). Cascade genetic testing of at-risk relatives once a familial variant is known; prenatal/PGT available when the variant is identified.

Sources: GeneReviews, JOMR clinicoradiographic review

11. Outcome / Prognosis

• Survival / mortality: Excellent — cherubism is benign and non-lethal. Normal life expectancy. Death is essentially never disease-attributable except vanishingly rare severe-airway scenarios.
• Natural history: the defining prognostic fact is spontaneous regression after puberty; most patients reach adulthood with substantial or complete cosmetic recovery.
• Morbidity / disability: driven by the active phase — facial disfigurement (psychosocial), dental/orthodontic problems, malocclusion, and in severe cases vision compromise, airway obstruction/OSA, feeding/speech difficulty. Long-term functional disability is uncommon with modern multidisciplinary care.
• Complications: OSA, orbital/visual compromise, tooth loss/agenesis, malocclusion, surgical morbidity or lesion regrowth if operated during the active phase, and rare rebound hypercalcemia after denosumab in children.
• Recovery potential: high — natural involution plus staged reconstruction after stabilization yields good outcomes in most.
• Prognostic factors: age (regression expected with maturity), anatomic grade/extent, orbital/airway involvement, and possibly sex (females more severely affected in recent series). No molecular prognostic biomarker in clinical use, though specific SH3BP2 genotype–severity correlations are debated and imperfect.

12. Treatment

Overarching strategy: Most cases need only observation ("watchful waiting") through the active phase, with staged reconstruction after regression. Aggressive/function-threatening disease (airway, vision, rapid growth) warrants active intervention. Multidisciplinary craniofacial-clinic management. Suggested MAXO anchors: MAXO:0000950 (supportive care), MAXO:0000004 (surgical procedure), MAXO:0000011 (physical/rehab therapy — speech), MAXO:0000079 (genetic counseling); observation/active-surveillance.

Pharmacotherapy (all off-label, evidence = small case reports/series; a 2023 systematic review pooled ~18 patients across 14 mostly-single-case studies — PMC10044089): - Denosumab (anti-RANKL monoclonal antibody; blocks osteoclastogenesis) — increasingly reported, including successful adult cherubism control with a 60 mg every-6-months regimen (JBMR Plus 2024/25, ziae164, PMC11742083). Caution: rebound hypercalcemia reported in a child after denosumab — a real pediatric safety concern. Suggested therapeutic_agent: denosumab (NCIT drug-class monoclonal antibody). - Tacrolimus (calcineurin inhibitor) — hits the calcineurin–NFATc1 node directly; a 4-yr-old with aggressive disease improved over 1 yr, with reduced TRAP+ osteoclasts and reduced NFATc1 nuclear staining on biopsy (PMID:25491283). CHEBI: tacrolimus. - Imatinib (tyrosine kinase inhibitor; targets SYK/downstream signaling) — "paradigm shift" preliminary reports and pediatric case(s) with marked lesion reduction, well tolerated (JOMS 2019). CHEBI:45783 (imatinib). - TNF-α blockers (adalimumab, etanercept) — mechanistically apt (TNF-α is the disease engine); anecdotal benefit. - Calcitonin — trialed in ≥5 pediatric cases (6–30 mo), mixed results (some regression, some none). - Bisphosphonates (e.g., pamidronate) — anti-resorptive; limited/variable evidence.

Advanced therapeutics: No gene/cell/RNA therapy in clinical use. Tankyrase-pathway biology suggests future rational targets, but tankyrase inhibitors would worsen disease (they stabilize SH3BP2) — so the therapeutic logic runs toward TNF-α/RANKL/NFATc1/SYK blockade, not tankyrase inhibition.

Surgical / interventional: - Curettage ± bone grafting, contouring/recontouring osteotomies, orthognathic reconstruction — best timed after lesion regression to limit recurrence/regrowth. Emergency surgery for airway or orbital decompression when function is threatened. Suggested: NCIT:C15329 (Surgical Procedure) / curettage.

Supportive / rehabilitative: orthodontics for malocclusion/dental management; speech-language therapy; ENT/sleep management for OSA; ophthalmology for orbital disease; psychosocial support.

Treatment outcomes: No RCT-level efficacy data; response is variable and agent-dependent. Given reliable spontaneous regression, the bar for systemic therapy is aggressive, function-threatening disease.

Sources: Pharmacological management systematic review (PMC10044089), Denosumab adult case (PMC11742083), Tacrolimus (PMID:25491283), Imatinib (JOMS)

13. Prevention

• Primary prevention: none possible (monogenic). The actionable lever is avoiding disease-provoking insults during the active phase — meticulous oral hygiene/periodontal-inflammation control (biologically supported by the microbe-dependence mouse data) and avoiding elective jaw surgery/extractions during active proliferation.
• Secondary prevention (early detection): cascade genetic testing of at-risk relatives once a familial SH3BP2 variant is known; early clinical/radiographic monitoring of known carriers to catch airway/orbital compromise early.
• Tertiary prevention (limit complications): structured surveillance — GeneReviews: clinical/radiographic assessment annually during active growth, then every 2–3 years after growth stops; dental review every 6 months; respiratory/ophthalmologic evaluation as needed. Manage OSA, protect vision, orthodontic maintenance.
• Reproductive prevention / counseling: genetic counseling for AD 50% transmission risk (and AR 25% for OGFRL1 families); prenatal diagnosis / preimplantation genetic testing available once the variant is known; discuss gonadal mosaicism and de novo possibilities.
• Immunization / public health / environmental: not applicable (non-infectious, non-environmental etiology).

Suggested MAXO: MAXO:0000079 (genetic counseling), surveillance/active monitoring, MAXO:0000950 (supportive care).

14. Other Species / Natural Disease

• Taxonomy affected: Human disease (NCBITaxon:9606, Homo sapiens). No well-characterized naturally occurring cherubism in companion animals or wildlife is documented in the veterinary literature (OMIA has no established spontaneous cherubism entry as of this review — flag as "not available").
• Orthologous genes: Sh3bp2 is conserved in mouse (Sh3bp2, human p.Pro418 ↔ mouse Pro416), rat, and other mammals; the tankyrase–RNF146 degradation axis is evolutionarily conserved. Ogfrl1 orthologs exist across vertebrates.
• Comparative biology: the P416R knock-in mouse is the workhorse model and recapitulates the core myeloid/inflammatory bone-loss biology (see §15). Interesting cross-species caveat: OGFRL1 knockout / frameshift mice did NOT reproduce human cherubism, implying the OGFRL1 loss-of-function effect diverges between human and mouse — a genuine human–model mismatch worth flagging for any KB entry.
• Zoonotic / cross-species transmission: not applicable (genetic, non-transmissible).

Sources: OGFRL1 study (PMC11062026)

15. Model Organisms

Flagship model — Sh3bp2 P416R knock-in mouse (Ueki et al., Cell 2007, PMID:17218256): - Type: mammalian germline knock-in (point mutation orthologous to human p.Pro418Arg). Heterozygous and homozygous lines. - Phenotype recapitulation: homozygous KI/KI mice show trabecular bone loss/osteoporosis, increased osteoclast numbers, TNF-α–dependent systemic macrophage inflammation, and cortical erosion; mutant myeloid cells hyper-respond to M-CSF/RANKL with elevated ERK/Syk and high TNF-α. Captures the autoinflammatory myeloid mechanism beautifully. - Genetic-dissection value (epistatic crosses): onto TNF-α–null (lesions resolve), op/op M-CSF-deficient (bone loss abrogated), and NFATc1 conditional KO (skeletal phenotype fully rescued) — these crosses established the causal hierarchy (TNF-α drives inflammation; NFATc1 is the essential bone-resorption effector; M-CSF is required for osteolysis). - Limitations: mice develop systemic/generalized inflammatory bone disease rather than the jaw-restricted human lesions — the striking human jaw-specificity and spontaneous pubertal regression are not faithfully reproduced, a key translational gap. Heterozygous mice are relatively mild unless challenged (e.g., oral microbial/periodontal challenge unmasks alveolar bone destruction — PMC7285758).

Related models: - Sh3bp2 knockout mice — used to show SH3BP2 is needed for optimal bone formation/osteoblast differentiation (PMID:20691350); loss-of-function ≠ cherubism (consistent with gain-of-function human mechanism). - Tankyrase-inhibitor / pathway models — pharmacologic tankyrase inhibition induces SH3BP2 accumulation and bone loss, phenocopying the mechanism (PMC6406327). - Ogfrl1 KO and Syrian-frameshift knock-in mice — generated for the recessive form but did not recapitulate human cherubism (human–model mismatch; PMC11062026). - In vitro / cellular: patient-derived and mutant myeloid cultures, osteoclast differentiation assays (M-CSF/RANKL), macrophage TNF-α assays; human lesion tissue immunohistochemistry (TRAP, NFATc1).

Applications: dissecting the tankyrase→SH3BP2→SYK/ERK→TNF-α/NFATc1→osteoclast axis; testing anti-TNF, anti-RANKL (denosumab), calcineurin inhibition, and SYK/TKI strategies.

Resources: MGI (mouse Sh3bp2), IMPC/KOMP for allele availability.

Consolidated Ontology Term Suggestions (for KB population)

Evidence Anchor Summary (verify PMIDs/quotes with just fetch-reference before curating)

Bottom line, Harry: cherubism is a gorgeous little natural experiment in protein garbage collection gone wrong — knock out the tankyrase tag on SH3BP2 and the cell can't take out the trash, so an adapter protein piles up, cranks the myeloid TNF-α/RANKL machinery, and the jaws pay the price until puberty flips the switch back off. When you build the KB entry, the two things I'd flag hardest: (1) it's gain-of-function via escape-from-degradation, not a broken enzyme, and (2) the mouse models capture the mechanism but not the jaw-restriction or the self-regression, and the OGFRL1 mice fail outright — real human-model-mismatch material worth a HUMAN_MODEL_MISMATCH discussion node. Every PMID above should still get run through just fetch-reference before it lands in evidence, since I pulled some from search metadata rather than reading each abstract line-by-line.

Sources: OMIM 118400 · OMIM 602104 (SH3BP2) · GeneReviews NBK1137 · Orphanet · Pathophysiology review (PMC3359958) · Ueki 2007 Cell (PMID:17218256) · Levaot 2011 tankyrase (PMID:22153076) · Mukai 2014 JBMR · OGFRL1 recessive (PMC11062026) · Best clinical practice (OJRD S6) · Pharmacological review (PMC10044089) · Denosumab adult (PMC11742083) · Tacrolimus (PMID:25491283) · Imatinib (JOMS) · Microbe-dependent bone destruction (PMC7285758) · Nature Genetics 2001