Diaphyseal Medullary Stenosis with Malignant Fibrous Histiocytoma

1. Disease Information

1.1 Overview

DMS‑MFH is a rare, autosomal‑dominant syndrome combining a distinctive long‑bone dysplasia with a high lifetime risk of high‑grade bone sarcoma. The skeletal lesion is diaphyseal medullary stenosis — progressive narrowing/obliteration of the medullary cavity of the long‑bone shafts by overlying endosteal cortical thickening — accompanied by scattered bone‑marrow infarctions, metaphyseal striations, pathologic fractures that heal poorly, progressive bowing of the lower limbs, and painful debilitation. Approximately one third of affected individuals develop a high‑grade bone sarcoma, historically diagnosed as malignant fibrous histiocytoma (MFH) or bone fibrosarcoma and, on modern review of at least one case, as osteosarcoma.

Two of the five known families additionally show a progressive limb‑girdle / facioscapulohumeral‑like myopathy, and features reported in individual pedigrees include presenile cataracts, thin skin, easy bruising, and premature graying.

Verbatim (Martignetti et al. 1999, PMID:10053015 abstract):

"Diaphyseal medullary stenosis with malignant fibrous histiocytoma (DMS-MFH) is an autosomal dominant bone dysplasia/cancer syndrome of unknown etiology. This rare hereditary cancer syndrome is characterized by bone infarctions, cortical growth abnormalities, pathological fractures, and eventual painful debilitation. Notably, 35% of individuals with DMS develop MFH, a highly malignant bone sarcoma."

Verbatim (Camacho‑Vanegas et al. 2012, PMID:22464254 abstract):

"Diaphyseal medullary stenosis with malignant fibrous histiocytoma (DMS-MFH) is an autosomal-dominant syndrome characterized by bone dysplasia, myopathy, and bone cancer."

Verbatim (Norton et al. 1996, PMID:8781110 abstract):

"Hardcastle syndrome is a rare, autosomally dominant inherited skeletal dysplasia, characterized by diaphyseal sclerosis, medullary stenosis, pathological fractures, bony infarction, and malignant transformation."

1.2 Identifiers (all verified)

Cross-reference block validated against Orphadata: api.orphadata.com ORPHA 85182 (CC‑BY‑4.0).

1.3 Synonyms and alternative names

• Hardcastle syndrome (most common eponym)
• Diaphyseal medullary stenosis–malignant fibrous histiocytoma syndrome
• Bone dysplasia–medullary fibrosarcoma syndrome
• Bone dysplasia with malignant fibrous histiocytoma
• BDMF (OMIM/GTR gene‑test abbreviation)
• DMSMFH / DMS‑MFH
• Hereditary bone dysplasia with malignant change / with sarcomatous degeneration (historical, Arnold 1973; Hardcastle 1986)
• Myopathy, limb‑girdle, with bone fragility (MedGen lists this as a synonym; formerly separate at MIM 609940)

Nomenclature note (important for curation). "Malignant fibrous histiocytoma" was retired as a diagnostic term in the 2013 WHO soft‑tissue classification and replaced by undifferentiated pleomorphic sarcoma (UPS); the 2020 WHO classification confirms UPS as "the correct designation for the storiform and pleomorphic variant of MFH." The disease name retains "MFH" for historical continuity, but the tumor should be curated as undifferentiated pleomorphic sarcoma of bone (UPSB) and/or osteosarcoma, per the 2012 re‑review of a family‑4 tumor.

1.4 Provenance of information

Aggregated, disease‑level, pedigree‑derived. There is no EHR cohort, no registry, and no population dataset for DMS‑MFH. All clinical data derive from five published multigenerational pedigrees (case series with radiographic and, latterly, molecular characterization). No dataset exists in GEO/dbGaP/SRA specific to this disease. This should be reflected in dismech prevalence.measure_type: CASES_IN_LITERATURE.

2. Etiology

2.1 Primary causal factor

Germline heterozygous mutation in a non‑canonical terminal exon (exon 9) of MTAP (9p21.3), acting through dysregulated alternative splicing of MTAP isoforms. Inheritance is Mendelian autosomal dominant; there is no known environmental or infectious contribution.

Verbatim (PMID:22464254):

"We now demonstrate that DMS-MFH results from mutations in the most proximal of three previously uncharacterized terminal exons of the gene encoding methylthioadenosine phosphorylase, MTAP."

2.2 Genetic risk factors

• Causal variants (2 known, see §4.2): c.885A>G (p.(=), synonymous R100R) and c.813-2A>G, both in/adjacent to the retroviral‑derived MTAP exon 9.
• Being a first‑degree relative of an affected individual is the only meaningful "risk factor" — 50% transmission risk per offspring.
• Somatic second hit: loss of the wild‑type allele (LOH at 9p21) is the tumor‑initiating event (§4.2, §6.1).
• No modifier genes identified. The variable myopathy across families (present in families 4 and 5, absent from the three original families) is unexplained; it is compatible with either an unrecognized modifier or with allele‑specific effects (the myopathic families 4 and 5 carry different mutations — family 4 c.885A>G, family 5 c.813-2A>G — so a simple allele–myopathy correlation is not supported).
• Susceptibility loci: No GWAS exists for DMS‑MFH (impossible at n≈5 families). The neighbouring 9p21 CAD/MI GWAS locus is discussed by the authors as a speculative overlap (§2.4).

2.3 Environmental risk and protective factors

None identified. No toxic, occupational, dietary, infectious, or radiation exposure has been associated with DMS‑MFH onset, bone phenotype severity, or sarcoma risk. CTD/TOXNET contain no DMS‑MFH entries. Mechanical loading is a plausible but unstudied modifier of the fracture phenotype.

No protective genetic or environmental factors are described. No protective alleles, no dietary or lifestyle intervention has been shown to reduce sarcoma risk.

Curation note: the environmental: section should be left empty or, if a curator has searched, carry the repo's uncited‑exposure waiver pattern (review_notes beginning "Left deliberately uncited." plus a record of the searches run), rather than manufacturing an exposure.

2.4 Gene–environment interaction

None documented. One speculative gene–environment/comorbidity thread from the primary paper is worth recording as a hypothesis rather than a finding — a possible MTAP contribution to coronary artery disease:

Verbatim (PMID:22464254, Discussion):

"First, in DMS-MFH-affected family 1, two male family members died of heart disease in their early forties without other known risk factors. A third family member has been recently diagnosed with early CAD (J.A.M., unpublished data). As such, CAD might represent a previously unrecognized aspect of the disease phenotype in this syndrome."

This is explicitly flagged by the authors as unpublished/anecdotal and should be curated as a KNOWLEDGE_GAP discussion, not as a phenotype.

3. Phenotypes

3.1 Core skeletal phenotypes

All HPO IDs below were verified against the HPO release in this repository via OAK (sqlite:obo:hp).

The bone‑marrow infarction component ("scattered infarctions within the bone marrow", PMID:22464254) has no precise HPO term; the closest ontology anchor is via UBERON UBERON:0002371 (bone marrow) plus a curated pathophysiology node. This is a genuine ontology gap worth noting.

3.2 Neoplastic phenotypes

Verbatim (PMID:22464254):

"Approximately one-third of affected individuals within our families developed bone sarcomas arising between the second and fifth decades of life. The diagnoses were either MFH or bone fibrosarcoma." "…given the presence of osteoid, the histopathological analysis of a tumor from a DMS-MFH-affected individual (III-3 from family 4; c.885A>G) is consistent with the diagnosis of osteosarcoma. Thus, inherited MTAP alternative-splicing mutations can result in histology-proven osteosarcoma."

Frequency caveat. The "35%" figure derives from three pedigrees in 1999 and was restated as "approximately one-third" in 2012. It is a within‑pedigree proportion, not an age‑adjusted penetrance estimate, and is subject to severe ascertainment bias (families were identified because of sarcoma). Curate as FrequencyEnum band with a note; do not present as a validated lifetime risk.

3.3 Neuromuscular phenotypes (families 4 and 5 only)

Verbatim (PMID:22464254, Introduction):

"We recently expanded the known clinical features of the syndrome by characterizing two new unrelated families affected by a progressive form of muscular disease consistent with facioscapulohumeral muscular dystrophy (FSHD [MIM 158900])."

Note the internal tension in the literature: the 2012 paper describes the myopathy as FSHD‑like in the Introduction and as "features overlapping the symptoms of facioscapulohumeral muscular dystrophy and limb-girdle muscular dystrophy" in the Discussion; Watts/Mehta (2005/2006) describe family 5 as limb‑girdle. Curate as limb‑girdle with an explicit note about the FSHD‑overlap description. FSHD1/2 (D4Z4) and LGMD genes were not, to my knowledge, formally excluded in these families.

3.4 Ocular, integumentary and connective-tissue phenotypes

Verbatim (PMID:22464254, Introduction):

"Affected individuals endure pathologic fractures that subsequently heal poorly, progressive wasting, bowing of the lower extremities, painful debilitation, and the development of presenile cataracts."

Mehta et al. 2006 (PMID:16419137) additionally report "premature graying with thin hair, thin skin, hernias" and clotting abnormalities in family 5 — features that have not been confirmed in the three original DMS‑MFH families and should be curated with a subtype/family qualifier rather than as disease‑wide.

3.5 Inheritance annotation

• Autosomal dominant inheritance — HP:0000006

3.6 Quality-of-life impact

No formal QoL instrument (EQ‑5D, SF‑36, PROMIS, TESS, MSTS) has ever been applied to a DMS‑MFH cohort. Reported impact is qualitative and severe:

• Repeated pathologic fractures with poor healing → prolonged immobilization, chronic pain ("eventual painful debilitation", PMID:10053015).
• Progressive lower‑limb bowing → gait impairment, reduced ambulation.
• Osteomyelitis complicating slow‑healing fractures has led to amputation (HP:0005010).
• Superimposed limb‑girdle myopathy in two families compounds mobility loss.
• Sarcoma diagnosis in the 2nd–5th decade brings the full burden of neoadjuvant/adjuvant chemotherapy plus limb‑salvage surgery or amputation.
• The tumor‑predisposition status imposes lifelong surveillance and reproductive/genetic‑counselling burden.

This is a legitimate KNOWLEDGE_GAP for the dismech entry: clinical_burden# with no quantitative instrument available.

4. Genetic / Molecular Information

4.1 Causal gene

MTAP — S‑methyl‑5′‑thioadenosine phosphorylase (methylthioadenosine phosphorylase).

Verbatim (PMID:22464254):

"MTAP is a ubiquitously expressed homotrimeric-subunit enzyme critical to polyamine metabolism and adenine and methionine salvage pathways and was believed to be encoded as a single transcript from the eight previously described exons."

The key structural discovery is that MTAP is not an eight‑exon gene. Three additional terminal exons exist:

• Exon 9 — a 192 bp ORF (GenBank AF216650), located 65 kb downstream of the previously annotated MTAP termination site; derived from a MER50I retroviral element.
• Exon 10 — derived from a THE1A element.
• Exon 11 — third terminal exon.

Verbatim:

"Intriguingly, two of these MTAP exons arose from early and independent retroviral-integration events in primate genomes at least 40 million years ago, and since then, their genomic integration has gained a functional role."

Exon 9 integrated "into the primate genome at some point in evolution between the divergence of the ring-tailed lemur and the common woolly monkey approximately 40 million years ago" — i.e., after prosimian and New World monkey divergence. The authors describe this as exaptation and state:

"However, we are unaware of any other example wherein the loss of a co-opted gene and/or protein domain results in a disease phenotype."

4.2 Pathogenic variants

Two germline variants, both in/adjacent to exon 9, both splicing‑altering with no predicted amino‑acid change. Numbering is relative to the EST reference clone GenBank AK309365, not the canonical NM_002451 transcript — an important caveat for anyone reconciling these to ClinVar.

Verbatim:

"one mutation was a synonymous change at position c.885A>G (p.(=)), effectively R100R, and was present in affected families 1, 3, and 4. The second mutation, c.813-2A>G, was an intronic change present in affected family 2."

Segregation and population evidence:

"The sequence changes segregated appropriately with the disease phenotype within all respective family members in each family… Neither mutation was identified in 1,000 chromosomes from 500 unaffected control individuals. Similarly, the mutations were not present in dbSNP build 131."

ACMG/AMP classification. Neither variant has, to my knowledge, been formally re‑classified under ACMG/AMP 2015 criteria. Under those criteria the evidence would be roughly: PS3 (well‑established functional studies — minigene splicing assays), PS4_moderate/PM (absent from 1,000 control chromosomes), PP1_strong (co‑segregation across five multigenerational pedigrees), PS3 supporting from the elevated serum MTA — but counterweighted by BS1‑type concerns from the ClinGen dosage curation and the fact that the reference transcript is non‑canonical. Best current summary: "reported pathogenic in the primary literature; not independently classified; ClinGen gene-level evidence rated as limited."

ClinVar status is confusing and must be handled carefully. Searching ClinVar for "Diaphyseal medullary stenosis-bone malignancy syndrome" returns records that are not the two published variants: e.g. NM_002451.4:c.*2837T>A (RCV000310131, VUS, 1‑star, Illumina Clinical Services, submitter note "No publications were found"), c.*2968T>C (RCV000332282), c.566G>T p.Trp189Leu (RCV000317123), c.315C>T p.Gly105= (RCV000330004). These are incidental submissions tagged to the condition, not disease‑causing alleles. The OMIM‑derived allelic variant record for c.885A>G is RCV000022659. Do not cite the 3′‑UTR VUS records as evidence for DMS‑MFH.

Somatic vs germline / two‑hit:

"Moreover, and in agreement with Knudson's two-hit hypothesis for a tumor-suppressing gene, direct sequencing of this patient's osteosarcoma genomic DNA demonstrated homozygosity for the c.885A>G mutation. LOH analysis with microsatellite markers spanning the originally defined 2.9 Mb DMS-MFH critical region revealed complete loss of the WT allele from the unaffected chromosome."

Functional consequence class. This is best modeled in dismech as functional_impact_category: LOSS_OF_FUNCTION for the exon‑9‑containing isoforms (v1, v2, v4, v5), with a simultaneous gain in v3/v6 — an unusual dual effect the authors explicitly flag:

"Given our findings that the DMS-MFH mutations also result in overexpression of two splice variants, MTAP_v3 and _v6, the possibility that at least two of the MTAP isoforms could represent oncogenic variants must also be considered at this time."

4.3 The six retroviral‑derived MTAP isoforms

Identified by 3′ RACE from control and patient fibroblast, lymphoblast and patient‑derived tumor cell lines. None contain wild‑type terminal exon 8; all alter the C‑terminus.

(9S = short 103 nt form of exon 9; 9L = long 192 nt form.) v4–v6 had "appreciably shorter half-lives"; activity remained undetectable even under proteasome inhibition with MG132.

All six isoforms "can physically interact with archetype MTAP" (co‑immunoprecipitation), supporting a heterotrimer / dominant‑negative‑like model at the subunit interface — the structural basis for dominant inheritance despite MTAP being a classic recessive‑style metabolic enzyme.

4.4 Linkage history

Candidate genes excluded by direct sequencing before exon 9 was found: CDKN2A (p16), p14‑ARF, CDKN2B (p15), the IFN gene cluster, and the eight canonical exons of MTAP.

4.5 Modifier genes, epigenetics, chromosomal abnormalities

• Modifier genes: none identified.
• Epigenetics: no DNA‑methylation, histone‑modification, or chromatin study has been performed in DMS‑MFH tissue. However, the mechanism is intrinsically epigenetic downstream (see §6.1): MTA accumulation inhibits PRMT5, reducing symmetric dimethylarginine marks including H4R3me2s and H3R8me2s. No ENCODE/Roadmap/DiseaseMeth dataset for this disease.
• Chromosomal abnormalities: none constitutional. Somatically, 9p21 deletion/LOH is the second hit; 9p21 is "one of the most frequently deleted and/or translocated chromosomal regions in human cancer" (gliomas, melanoma, NSCLC, acute leukemias, osteosarcomas). The 2012 authors raise a methodological warning worth curating:

"The facts that MTAP is more complex than previously recognized and that its terminal coding exon lies within 25 kb of the p15/p16 locus has immediate significance to LOH mapping and copy number variation (CNV) studies in human cancer. Deletions including the p15/p16 locus will more than likely also include the 3′ region of MTAP and therefore might affect MTAP biochemical activity. Thus, the interpretation of many of these studies with regard to the genes being affected should be reevaluated."

5. Environmental Information

Not applicable. DMS‑MFH is a purely Mendelian disorder with no established environmental, lifestyle, or infectious contribution.

• Environmental factors: none identified. No CTD entries.
• Lifestyle factors: none identified. Weight‑bearing/mechanical loading is a theoretical modifier of fracture frequency but is unstudied.
• Infectious agents: none. The retroviral element in exon 9 is an endogenous retroviral (ERV) remnant fixed in the primate germline 40 Mya, not an active infection. This distinction must be preserved in any curated text — MER50I and THE1A are exapted genomic sequence, not pathogens, and there is no NCBITaxon organism to annotate.

6. Mechanism / Pathophysiology

6.1 The causal chain (upstream → downstream)

Step 1 (MOLECULAR) — Germline exon‑9 splicing mutation. c.885A>G abolishes an ESE; c.813-2A>G destroys the splice acceptor. Both are heterozygous and germline. GO:0000380 (alternative mRNA splicing, via spliceosome)

Step 2 (MOLECULAR) — Dysregulated MTAP isoform stoichiometry. Verbatim: "c.813-2A>G and c.885A>G resulted in markedly decreased [expression]… the c.813-2A>G mutation ablated expression of all isoforms containing exon 9 and significantly increased the [archetype MTAP]… The c.885A>G mutation decreased the expression levels of all exon 9 isoforms by approximately 70% but had no [effect on archetype]." Loss of the enzymatically active exon‑9‑containing isoforms v1/v2 and the inactive v4/v5; overexpression of v3/v6.

Step 3 (MOLECULAR) — Perturbed MTAP holoenzyme assembly. All six isoforms co‑immunoprecipitate with archetype MTAP; molecular modelling on PDB 1CG6/1CB0 shows the splice‑variant insertion points (after K271 for v1‑type, after A230 for v4‑type) sit at the trimer subunit interface, adjacent to the exon‑6/exon‑7‑encoded substrate‑binding site (only L279 comes from exon 8). Verbatim: "The trimeric subunit interface of MTAP does appear to be affected by the alternate splicing events or, possibly, the MTA active site." GO:0017061 (S-methyl-5-thioadenosine phosphorylase activity), modifier DECREASED

Step 4 (ORGANISM/MOLECULAR) — Systemic MTA accumulation. ← the key human in‑vivo evidence MTAP normally phosphorolyses MTA (CHEBI:17509) to adenine (CHEBI:16708) + 5‑methylthioribose‑1‑phosphate, which is recycled to L‑methionine (CHEBI:16643). MTA is the by‑product of polyamine (spermidine CHEBI:16610, spermine) synthesis from SAM (CHEBI:15414). GO:0071267 (L-methionine salvage), GO:0043101 (purine-containing compound salvage), GO:0006595 (polyamine metabolic process)

Verbatim:

"MTA is not normally present in human serum. Cells lacking MTAP activity are unable to metabolize MTA, and functional inhibition or dysregulation of MTAP activity would therefore be expected to result in intracellular MTA accumulation and secretion… All three serum samples from unaffected individuals had no detectable MTA levels. In marked contrast, both affected individuals had accumulations of MTA detectable in their serum."

This is a biomarker‑grade finding (n=2 affected vs 3 unaffected, blinded assay) and the single strongest piece of human in‑vivo mechanistic evidence for the disease. It is also, at n=5, a fragile one.

Step 5 (MOLECULAR, inferred) — PRMT5 hypomethylation. Not demonstrated in DMS‑MFH tissue, but established for MTAP‑null cancer generally: accumulated MTA is a SAM‑competitive inhibitor of PRMT5 with >100‑fold selectivity over other PRMT family members, reducing symmetric arginine dimethylation of histone and non‑histone substrates. GO:0019918 (peptidyl-arginine methylation, to symmetrical-dimethyl arginine), GO:0032259 (methylation) Curate as a MECHANISTIC_HYPOTHESIS / EMERGING hypothesis group with cancer‑biology (not DMS‑MFH) evidence.

Step 6a (CELLULAR/TISSUE) — Bone dysplasia arm. How MTAP dysfunction produces endosteal cortical thickening, medullary obliteration, marrow infarction, and impaired fracture healing is genuinely unknown. The only mechanistic thread the authors offer is polyamine‑dependent angiogenesis:

"Third, defects in polyamine metabolism have been associated with defects in angiogenesis and altered myocyte function, whereas a nearly pathognomonic feature of DMS-MFH bone dysplasia is the presence of scattered infarctions throughout the medullary cavity."

This is the highest‑value knowledge gap in the entry. Cell types implicated by the phenotype (not by direct evidence): osteoblast CL:0000062, osteocyte CL:0000137, osteoclast CL:0000092, bone‑marrow mesenchymal stem cell CL:2000079/CL:0000134, and marrow microvascular endothelium. GO:0001649 (osteoblast differentiation), GO:0001503 (ossification), GO:0030282 (bone mineralization)

Step 6b (TISSUE) — Sarcomagenesis arm. Germline heterozygous MTAP defect → somatic LOH/loss of the WT 9p21 allele (Knudson two‑hit) → biallelic MTAP dysfunction in a mesenchymal progenitor → high‑grade sarcoma (UPS/osteosarcoma) in the 2nd–5th decade. The 2012 authors present this as the first in‑vivo human demonstration of MTAP tumour suppression:

"…the identification of the MTAP splice variants and the fact that their genetic loss results in DMS-MFH provide a possible in vivo demonstration that MTAP can act as a tumor suppressor."

Supporting in vitro precedent: "Reintroduction of MTAP expression into the MCF7 breast adenocarcinoma cell line, which lacks endogenous MTAP gene expression and enzymatic activity, inhibits the cells' ability to grow both in vitro and in vivo."

Step 6c (TISSUE) — Myopathy arm (families 4, 5). Mechanism unknown. Only linkage cited: "defects in polyamine metabolism have been associated with… altered myocyte function." Cell type: skeletal muscle fiber CL:0008002.

6.2 Protein dysfunction

Not misfolding or aggregation. The defect is isoform stoichiometry within an obligate homotrimer: wild‑type‑length archetype MTAP retains catalytic competence, but the normal complement of C‑terminally variant subunits that co‑assemble with it is lost or skewed. Predicted (unvalidated) structural detail: "Secondary-structure and disulfide-bond prediction analyses predict the generation of a disulfide bond between cysteine residues in exons 9 and 10; these prediction analyses require future biochemical analysis for validation."

6.3 Metabolic changes

• Methionine salvage — impaired (MTA→methylthioribose‑1‑P→methionine).
• Adenine salvage — impaired; MTAP phosphorolysis "is the principle source of free adenine in human cells."
• Polyamine metabolism — MTA, the by‑product of spermidine/spermine synthesis, is not cleared.
• Methylation potential — MTA:SAM ratio rises, biasing SAM‑dependent methyltransferases (PRMT5 most sensitively).

6.4 Immune involvement

None in the human syndrome. In the mouse, however, Mtap heterozygosity produces a lymphoid phenotype (§15) — an interesting species divergence and a candidate HUMAN_MODEL_MISMATCH.

6.5 Tissue damage mechanisms

• Ischemia/infarction: medullary bone infarcts are near‑pathognomonic; presumed vascular (polyamine–angiogenesis link, unproven).
• Impaired repair: fractures "heal poorly"; slow union permits osteomyelitis and, in some cases, amputation.
• Mechanical: stenosed medulla + thickened cortex alters bone biomechanics, predisposing to pathologic fracture and progressive bowing.

6.6 Molecular profiling — what exists and what does not

7. Anatomical Structures Affected

7.1 Organ level

Primary: the skeleton, specifically the long tubular bones. - Long bone — UBERON:0002495 - Diaphysis — UBERON:0004769 (the defining site) - Metaphysis — UBERON:0001438 (striations) - Femur UBERON:0000981, tibia UBERON:0000979, fibula UBERON:0001446, humerus UBERON:0000976 (diaphysis of humerus UBERON:0004652); the 1996 Norton report's MeSH indexing names femur, fibula and tibia specifically.

Secondary / additional systems: - Musculoskeletal (muscle): skeletal muscle, limb‑girdle distribution (families 4, 5). - Ocular: lens — UBERON:0000965 (presenile cataracts). - Integumentary: skin (thin/soft), hair (premature graying). - Hematologic/vascular: bone‑marrow vasculature (infarction); reported clotting abnormality in family 5. - Cardiovascular: speculative early CAD in family 1 (unpublished, do not curate as phenotype).

Body systems: skeletal (primary), muscular, ocular, integumentary, hematopoietic/marrow.

7.2 Tissue and cell level

Curation warning: every cell‑type assignment above except fibroblast is inferred from tissue phenotype, not demonstrated. There is no cell‑type‑resolved study of DMS‑MFH. Bind these with descriptions that make the inference explicit, or leave them out.

7.3 Subcellular level

MTAP is cytosolic. Relevant GO cellular components: cytosol (GO:0005829), with nuclear relevance downstream via PRMT5 substrate methylation. No organellar pathology (mitochondrial, ER, lysosomal) is described.

7.4 Localization and lateralization

• Bilateral and symmetric. The dysplasia is described as symmetric diaphyseal medullary stenosis of the long bones. Radiographic screening therefore uses bilateral long‑bone films.
• Distribution: long tubular bones of the limbs; the 2012 description emphasizes "diffuse diaphyseal medullary stenosis with overlying endosteal cortical thickening" — i.e. length‑wise diffuse rather than focal.
• Tumors arise focally and are not symmetric; extremity location predominates (consistent with UPSB generally).

8. Temporal Development

8.1 Onset

Onset pattern: insidious and chronic for the dysplasia; acute for each fracture event; subacute for tumor presentation (pain, mass, or pathologic fracture through tumor).

Note the developmental sequencing this implies: the skeletal dysplasia is essentially congenital/constitutional in genotype but adolescent in radiographic expression, and the malignancy is a late, stochastic, second‑hit event.

8.2 Progression

Stages (not formally defined — proposed for curation from the natural history): 1. Latent / radiographic-only — childhood to puberty; asymptomatic, dysplasia visible on plain film. 2. Fracture phase — adolescence through adulthood; recurrent pathologic fractures with poor union; progressive bowing. 3. Debilitation phase — adult; chronic pain, deformity, infarction burden, superimposed myopathy in some families; osteomyelitis/amputation risk. 4. Malignant transformation — 2nd–5th decade; ~1 in 3 affected individuals; converts prognosis from chronic disability to life‑threatening.

Progression rate: slow and steady for the dysplasia (decades); the myopathy is explicitly "progressive"; sarcoma behaves as a high‑grade, rapidly progressive malignancy.

Course pattern: chronic progressive with superimposed episodic events (fractures, infarcts, tumor). Not relapsing‑remitting.

Duration: lifelong. There is no self‑limited form.

8.3 Patterns

• Remission: none for the skeletal dysplasia — it is structural and irreversible. Tumor remission is treatment‑induced only (see §11, §12).
• Critical periods:
• Puberty — the radiographic screening window opens (Norton 1996 recommendation).
• 2nd–5th decades — the sarcoma surveillance window; this is the period where surveillance could plausibly change outcome.
• Peri‑fracture — the intervention window for preventing malunion and osteomyelitis.

9. Inheritance and Population

9.1 Epidemiology

• Prevalence: not documented. Neither Orphanet nor GARD publishes a prevalence estimate. Orphanet describes the disorder only as "very rare."
• Total reported cases: five families worldwide (as of the 2012 gene‑discovery paper and, to my knowledge, still the case in 2026). Family origins: American (Vermont/New York), Australian, English, and two New York families — one of which (family 5) is the 1958 Canadian Henry kindred with AD bone fragility and limb‑girdle myopathy.
• Incidence: not documented. No registry, no national surveillance, no SEER capture (SEER codes the tumor, not the syndrome).

dismech prevalence record shape:

prevalence:
- population: Worldwide
  measure_type: CASES_IN_LITERATURE
  prevalence_class: ULTRA_RARE
  notes: >-
    Five unrelated multigenerational families reported worldwide (Arnold 1973
    American kindred; Hardcastle 1986 English and Australian families; Norton
    1996 New York family; a second New York family and the 1958 Henry Canadian
    kindred added by Camacho-Vanegas 2012). No prevalence estimate is published
    by Orphanet or GARD.

9.2 Genetic parameters

9.3 Population demographics

• Affected populations: all reported families are of European/European-diaspora ancestry (United States, England, Australia, Canada). This almost certainly reflects ascertainment, not biology — the disorder has only ever been studied by centres in the US, UK and Australia. Do not curate as an ancestry association.
• Geographic distribution: North America (US, Canada), United Kingdom, Australia. No endemic focus.
• Variant geography: no meaningful pattern at n=2 variants / 5 families.
• Sex ratio: 1:1 expected and consistent with reports — autosomal dominant, "males and females equally affected with 50% transmission risk per child" (MedGen). No sex bias in bone or tumor phenotype has been reported.
• Age distribution of affected individuals: all ages; clinical burden concentrates in the 2nd–5th decades.

10. Diagnostics

10.1 Imaging — the diagnostic cornerstone

Plain radiography of the long bones is the primary diagnostic modality. Characteristic findings: - Symmetric diaphyseal medullary stenosis with endosteal cortical thickening (HP:0100254, HP:0005045) - Metaphyseal striations (HP:0031367) - Patchy osteosclerosis (HP:0005686) - Scattered medullary infarcts — serpiginous sclerotic rims - Healed/healing pathologic fractures, bowing deformity

MRI characterizes marrow infarction and is essential for tumor staging and local extent (used in the Norton 1996 report).

Thallium‑201 scintigraphy — proposed specifically for this syndrome as a tumor‑sensitive screen, because ordinary bone scintigraphy is confounded by the diffusely abnormal, infarcted, remodelling skeleton:

"Thallium scanning is proposed as a more tumor-sensitive screening agent in affected individuals." (PMID:8781110)

CT for cortical detail; FDG‑PET/CT is now standard for sarcoma staging generally, though it has not been specifically validated in DMS‑MFH and would be expected to have reduced specificity against a background of bone infarction and remodelling — a real and under‑discussed diagnostic problem in this disease.

10.2 Laboratory tests and biomarkers

• Serum MTA (5′‑deoxy‑5′‑methylthioadenosine, CHEBI:17509) — the only disease‑specific biochemical marker. MTA is undetectable in normal serum; affected individuals had 11.5 and 4.3 pmol/100 µL. This is a research assay, not a clinical test — it has been performed in a single study on two affected individuals and three controls, has no validated reference interval, no established cutoff, no LOINC code, and is not offered by any clinical laboratory. Curate as an association/biomarker with validation_status language making this explicit, not as a diagnostic test.
• Routine bone chemistry (calcium, phosphate, ALP, PTH, 25‑OH‑D) is used to exclude metabolic bone disease; no characteristic abnormality is reported.
• CK is appropriate in the myopathic families; specific values are not reported in the accessible literature.

10.3 Biopsy and pathology

Sarcoma diagnosis requires histopathology. In the one modern re‑review reported:

"(A) Histologic analysis revealed that 95% of the studied tumor specimen displayed the typical pattern of malignant spindle (fibroblastic) cells of bone MFH. (B) Malignant cells forming neoplastic bone. (C) Focal sheets of neoplastic bone within the tumor are shown to be entrapping pre-existing bone trabeculae, and the overtly malignant cells are shown to produce bone."

The presence of osteoid reclassified this tumor as osteosarcoma. This matters practically: UPS of bone and osteosarcoma are treated on the same protocol, so the reclassification does not change management, but it does change the phenotype annotation.

Immunohistochemistry is used mainly by exclusion — UPS is a diagnosis of exclusion after ruling out specific lineages (SMA/desmin for leiomyosarcoma, S100/SOX10 for melanoma and MPNST, keratins/EMA for sarcomatoid carcinoma, CD34, MDM2/CDK4 FISH for dedifferentiated liposarcoma). MTAP immunohistochemistry is now a validated surrogate for 9p21 co‑deletion in other tumor contexts (notably mesothelioma) and is worth considering here — though its interpretation in a germline exon‑9 splicing mutation, where archetype MTAP protein is retained or even increased, is genuinely unclear and would likely be falsely negative. Worth flagging as a knowledge gap.

10.4 Genetic testing

• Recommended approach: targeted sequencing of MTAP exon 9 and its intron–exon boundaries in a proband from a family with the characteristic radiographic dysplasia.
• Critical technical caveat — the single most actionable diagnostic point in this report: exon 9 lies 65 kb downstream of the canonical MTAP termination site, in what standard annotation treats as intergenic/3′ region. Standard exome sequencing (WES) and standard clinical gene panels will not capture or will not report this region. Any DMS‑MFH testing must either be a custom targeted assay or genome sequencing (WGS) with explicit non‑coding analysis. The 2012 authors reached the answer only after "DNA-sequence analysis of known candidate genes within the original critical region had previously failed to identify causative mutations… In particular, this analysis included the known eight exons and corresponding intron-exon boundaries of MTAP."
• WGS: the preferred agnostic approach for a new suspected case; requires targeted interrogation of the exon‑9 region.
• Gene panels: MTAP appears on the Genomics England Childhood solid tumours panel but as Red, meaning it is not reported diagnostically on that panel. Practically, this means a suspected DMS‑MFH case will not be resolved by ordering a routine hereditary‑cancer or skeletal‑dysplasia panel. GTR lists tests under the abbreviation BDMF.
• RNA sequencing / transcript analysis is arguably the most sensitive functional assay, given that both mutations act through splicing — patient fibroblast or lymphoblast RT‑PCR/3′ RACE demonstrating loss of exon‑9‑containing isoforms is the direct functional readout.
• CMA, karyotyping, FISH: not indicated (no constitutional chromosomal abnormality). FISH for 9p21 is relevant only in the tumor, to demonstrate the somatic second hit.
• mtDNA and repeat‑expansion testing: not applicable.

10.5 Omics-based diagnostics

None validated. RNA‑seq is the modality with real potential (§10.4). No proteomic, metabolomic (beyond the research MTA assay), epigenomic, or liquid‑biopsy diagnostic exists.

10.6 Clinical criteria and differential diagnosis

There are no formal, published diagnostic criteria. A practical working definition: symmetric diaphyseal medullary stenosis with endosteal cortical thickening on long‑bone radiographs, in an autosomal dominant pedigree, with or without pathologic fractures, marrow infarction, myopathy, or bone sarcoma.

Differential diagnosis:

10.7 Screening

• Radiographic screening of at‑risk relatives from puberty onward — the one explicit published recommendation (Norton 1996).
• Thallium scintigraphy for tumor surveillance in known‑affected individuals (Norton 1996).
• Cascade genetic testing of at‑risk relatives once a familial variant is identified — feasible in principle, constrained by the assay problem in §10.4.
• Newborn screening: not applicable; no biochemical newborn screen and no benefit from presymptomatic neonatal identification.
• Carrier screening: not applicable (dominant).

Honest assessment: these recommendations date from 1996 and have never been prospectively evaluated. Thallium‑201 is now largely obsolete in clinical practice, displaced by FDG‑PET/CT, and no modern surveillance protocol for DMS‑MFH exists. This is a defensible KNOWLEDGE_GAP.

11. Outcome / Prognosis

11.1 Survival and mortality

No DMS‑MFH‑specific survival data exist. No cohort of sufficient size has ever been assembled. Prognosis must be assembled from two components: the chronic skeletal/myopathic disability, and the sarcoma.

Sarcoma survival (proxy data from UPS of bone cohorts — the best available surrogate):

The gap between these two figures is instructive: the COSS cohort is a specialist, protocol-treated, younger population; the registry cohort is unselected. A DMS‑MFH patient — typically young, extremity primary, treated at a sarcoma centre — is closer to the COSS profile, but carries the added liability of a diffusely abnormal skeleton complicating limb salvage.

Life expectancy: in the absence of sarcoma, likely near‑normal but with substantial disability. With sarcoma, governed by the figures above. Individuals never developing sarcoma (~2/3) face a chronic musculoskeletal disease rather than a life‑limiting one.

11.2 Morbidity and function

• Chronic bone pain and "eventual painful debilitation" (PMID:10053015).
• Recurrent fractures with poor union; progressive lower‑limb bowing.
• Osteomyelitis leading to amputation due to slow healing fractures (HP:0005010) — an explicitly curated HPO feature of this disease.
• Progressive limb‑girdle weakness in the myopathic families, superimposing neuromuscular disability on structural bone disease.
• Presenile cataracts → visual impairment requiring surgery.
• No ICF‑coded disability data, no EQ‑5D/SF‑36/PROMIS/MSTS/TESS data.

11.3 Complications

Pathologic fracture · fracture nonunion/malunion · osteomyelitis · amputation · bone infarction · progressive deformity · sarcoma · chemotherapy toxicity (anthracycline cardiotoxicity, cisplatin ototoxicity/nephrotoxicity, ifosfamide neuro/nephrotoxicity, secondary malignancy) · possible early CAD (unpublished/speculative).

11.4 Recovery potential

The skeletal dysplasia is irreversible. No treatment restores medullary architecture. Fractures heal — poorly and slowly. Sarcoma is curable in a substantial fraction with multimodal therapy (see §12), and this is the one domain where treatment materially changes outcome.

11.5 Prognostic factors

For the sarcoma (extrapolated from UPSB cohorts, COSS 2026): - Favourable: age <40 years; extremity primary; localized (M0) disease at presentation. - Unfavourable: metastatic disease at presentation (M1 5-yr DSS 16.9% vs M0 56.4%); age ≥40 years (and ≥65 in the registry analysis); non‑extremity site; pathologic fracture at presentation — a factor of specific concern in DMS‑MFH, where pathologic fracture is a baseline feature of the disease and may confound this prognostic variable entirely. - Notably: in the COSS cohort only 38% achieved good histologic response (<10% viable tumor) to preoperative chemotherapy, and — unusually for bone sarcoma — this "correlated poorly with prognosis," meaning the standard osteosarcoma response‑to‑neoadjuvant prognostic marker may not transfer to UPSB.

Disease-specific prognostic factors for DMS‑MFH: none identified. Neither serum MTA level, nor which of the two mutations a family carries, nor presence of myopathy has been shown to predict outcome — the sample size makes such analysis impossible.

Prognostic biomarkers: none validated.

12. Treatment

There is no disease‑modifying or targeted therapy for DMS‑MFH. Management is entirely symptomatic, orthopaedic, and oncologic.

12.1 Skeletal management

Bisphosphonates and other antiresorptives are not indicated and are theoretically hazardous — DMS‑MFH is a sclerosing dysplasia with medullary obliteration and impaired fracture healing, not an osteopenic one. Suppressing remodelling further is unlikely to help and may worsen union. No study addresses this; flag as a knowledge gap rather than a recommendation.

12.2 Sarcoma management

Patients are treated on osteosarcoma protocols. From the COSS UPSB cohort (n=132, 2026):

• Chemotherapy in 100% of patients: doxorubicin 98%, cisplatin 93%, ifosfamide 82%, high‑dose methotrexate 64%.
• Surgery in 96%; among extremity tumors, 81% limb salvage vs amputation.
• Representative dosing reported elsewhere: doxorubicin 60 mg/m², ifosfamide 9 g/m², cisplatin 90 mg/m², with methotrexate 8 g/m² added for poor histologic responders.

dismech treatment record shape:

treatments:
- name: Osteosarcoma-Type Multiagent Chemotherapy
  therapeutic_modality: SMALL_MOLECULE
  treatment_term:
    preferred_term: chemotherapy
    term: {id: NCIT:C15632, label: Chemotherapy}
    therapeutic_agent:
    - preferred_term: doxorubicin
      term: {id: CHEBI:28748, label: doxorubicin}
    - preferred_term: cisplatin
      term: {id: CHEBI:27899, label: cisplatin}
    - preferred_term: ifosfamide
      term: {id: CHEBI:5864, label: ifosfamide}
    - preferred_term: methotrexate
      term: {id: CHEBI:44185, label: methotrexate}

(All four CHEBI IDs verified via OAK. regimen_term should be omitted — MAP/MAPI is not a named NCIT Treatment Regimen concept that reliably resolves; do not invent one.)

• Radiotherapy (NCIT:C15313) has a limited role — for unresectable or positive‑margin disease, as in bone sarcoma generally.
• Guideline reference: NCI PDQ Osteosarcoma and Undifferentiated Pleomorphic Sarcoma of Bone Treatment (NBK65736) is the appropriate clinical guideline anchor — note that PDQ explicitly co‑names UPS of bone with osteosarcoma, confirming the shared‑protocol approach.

12.3 Genetic counselling

NCIT:C15240 Genetic Counseling. Autosomal dominant, 50% offspring risk; discussion of variable expressivity (myopathy present in only some families), the ~1‑in‑3 sarcoma risk, and the technical limitation that standard panels/exomes will not detect the causal variant.

12.4 The MTAP/PRMT5 therapeutic axis — the major recent development

This is where 2023–2026 literature is genuinely active, and it is the most interesting forward‑looking content for this entry. It is relevant to DMS‑MFH by mechanism, not by evidence — no DMS‑MFH patient has ever received these agents, and the direction of effect may not even be favourable (see the caveat below).

The synthetic lethality. MTAP loss → MTA accumulation → MTA is a SAM‑competitive inhibitor of PRMT5 with >100‑fold selectivity over other PRMTs → PRMT5 enters a "frail," partially inhibited state → the cell becomes selectively hypersensitive to further PRMT5 inhibition. This is the basis of an entire drug class: MTA‑cooperative PRMT5 inhibitors, which preferentially bind PRMT5 in the MTA‑bound state and therefore spare MTAP‑intact normal tissue — solving the hematologic toxicity that sank first‑generation PRMT5 inhibitors.

Epidemiologic scale: homozygous MTAP deletion (almost always co‑deleted with CDKN2A) occurs in ~10–15% of all solid tumors; by type (C‑CAT data, per the 2025 Int J Mol Sci review, PMC12733126): mesothelioma 33.1%, urothelial carcinoma 23.8%, CNS tumors 19.0%, pancreatic 18.4%, cholangiocarcinoma 15.6%, NSCLC 14.3%; lowest in prostate (1.2%), colon (1.7%), cervix (3.6%).

Clinical-stage agents:

Biomarker: "The only factor qualifying for therapy is the presence of a homozygous deletion of the MTAP gene, detected by NGS or FISH."

The critical caveat for DMS‑MFH — do not curate this as a treatment. DMS‑MFH is not an MTAP‑homozygous‑deletion state. The germline lesion is a heterozygous splicing mutation in a non‑canonical exon that preserves and in one allele class increases archetype MTAP expression. Whether a DMS‑MFH tumor (which does undergo somatic LOH at 9p21) reaches the MTA‑high, PRMT5‑frail state required for MTA‑cooperative PRMT5 inhibitor sensitivity is unknown and untested. The elevated serum MTA is suggestive but not equivalent to intratumoral MTA saturation. Additionally, MTAP IHC — the cheap surrogate biomarker — would likely be falsely negative here because archetype protein is retained.

Curate this as a mechanistic_hypotheses entry with status: EMERGING and a discussions entry of kind: KNOWLEDGE_GAP asking whether DMS‑MFH tumors phenocopy MTAP‑deleted tumors for PRMT5 dependency — a well‑posed, tractable, and genuinely open question that this KB entry can usefully record.

12.5 Other therapeutic modalities

• Gene therapy, gene editing, ASOs, siRNA, cell therapy, immunotherapy: none developed, none in trial. Splice‑switching ASOs are a theoretically attractive modality given that both mutations act on splicing — but no such program exists, and the direction of correction differs between the two alleles.
• Pharmacogenomics: no DMS‑MFH‑specific PGx. Standard sarcoma‑chemotherapy PGx applies (e.g. TPMT/NUDT15 not relevant here; UGT1A1 not relevant; consider standard anthracycline cardiotoxicity risk factors and cisplatin ototoxicity pharmacogenetics — TPMT/COMT variants have CPIC‑adjacent literature).
• Clinical trials: the only DMS‑MFH‑specific registered study is NCT00007046, "Genetic Study of Patients and Families With Diaphyseal Medullary Stenosis With Malignant Fibrous Histiocytoma of the Bone" (Mount Sinai; a natural‑history/gene‑discovery study, not interventional). I was unable to retrieve its current status; verify with just fetch-reference NCT00007046 before citing it as an evidence item.

12.6 Treatment strategy summary

There is no DMS‑MFH treatment algorithm. The practical pathway is: (1) confirm the dysplasia radiographically and, where possible, molecularly; (2) orthopaedic management of fractures and deformity with awareness that medullary stenosis constrains fixation options; (3) lifelong surveillance for sarcoma; (4) on sarcoma diagnosis, refer to a sarcoma centre and treat on an osteosarcoma protocol; (5) genetic counselling and cascade evaluation of relatives.

13. Prevention

No primary prevention is possible. The disorder is a germline Mendelian condition; nothing prevents its occurrence in a person who inherits the allele, and no intervention is known to reduce the ~35% sarcoma risk.

• Immunization: not applicable — no infectious component. (Standard immunosuppression‑related vaccination guidance applies during chemotherapy.)
• Genetic screening: carrier screening not applicable (dominant); PGT‑M and prenatal testing are technically feasible where the familial exon‑9 variant is known — one of the few concretely actionable options.
• Risk stratification: no validated model. Family history plus radiographic phenotype is the entirety of current stratification.
• Behavioural interventions / public health / environmental interventions: not applicable.
• Prophylaxis: no prophylactic medication. Prophylactic surgery has no established role — unlike, say, risk‑reducing mastectomy in BRCA1/2, there is no organ to remove, since the at‑risk tissue is the entire skeleton.

14. Other Species / Natural Disease

14.1 Taxonomy and orthology

• Human — NCBITaxon:9606 — the only species with the natural disease.
• Mouse — NCBITaxon:10090 — Mtap, MGI:1914152, NCBI Gene 66902.
• MTAP is broadly conserved across eukaryotes and archaea (the Sulfolobus solfataricus MTAP‑II structure is solved), consistent with a core metabolic salvage function.

14.2 The evolutionary caveat that dominates this section

The disease mechanism cannot be modelled in any non‑primate species. Exon 9 (MER50I‑derived) and exon 10 (THE1A‑derived) entered the genome by retroviral integration after prosimian/New World monkey divergence, ~40 Mya. Verbatim:

"First, given the evolutionary species restriction of these HERVs, the existence of MTAP variants and possible biochemical regulation resulting from their expression must be unique to primates."

Mice, zebrafish, Drosophila, and C. elegans have Mtap orthologs but do not have exon 9. There is therefore no mouse, rat, fish, or fly in which the DMS‑MFH mutation can be recapitulated. This is the defining constraint on the entire experimental biology of this disease and belongs in the dismech entry as an explicit HUMAN_MODEL_MISMATCH discussion.

14.3 Natural disease in other species

None reported. OMIA contains no DMS‑MFH entry. No naturally occurring MTAP bone dysplasia/sarcoma syndrome is described in dogs, cats, horses, or any other companion or wildlife species. No breed association (no VBO term applies).

14.4 Comparative pathology

• Divergent phenotype in mouse. Homozygous Mtap loss is embryonic lethal in mouse; heterozygotes develop T‑cell lymphoma, not bone dysplasia or sarcoma (§15). The human heterozygote develops a skeletal dysplasia with mesenchymal sarcoma. This is a striking species divergence, plausibly attributable to the different lesion class (whole‑gene knockout vs primate‑specific isoform dysregulation) and worth curating as a FAILS_TO_RECAPITULATE/PARTIALLY_RECAPITULATES link.
• Conservation of the core enzyme (methionine/adenine salvage, polyamine metabolism) is deep; conservation of the regulatory isoform layer is primate‑restricted.

14.5 Transmission

Not applicable. No zoonotic potential, no cross‑species susceptibility. The endogenous retroviral elements are fixed germline sequence, not transmissible agents.

15. Model Organisms

15.1 Existing genetic models

Mouse — Mtap whole‑gene disruption (Kadariya et al., Cancer Res 2009;69(14):5961–9, PMID:19567676).

• Genotype: Mtap^lacZ knockout allele; heterozygous Mtap^+/lacZ.
• Homozygotes: early embryonic lethal.
• Heterozygotes: born at Mendelian ratios, indistinguishable from wild-type through the first year, then die prematurely with median survival 585 days of T‑cell lymphoma. Necropsy shows "greatly enlarged spleens, altered thymic histology, and lymphocytic infiltration of their livers." Older surviving heterozygotes show mild, non‑clonal T‑cell lymphoproliferative disease, "suggesting that loss of Mtap might stimulate T-cell proliferation."
• Interpretation: supports MTAP as a bona fide tumor suppressor — the strongest independent in‑vivo evidence for tumor suppression, though in a different lineage.

Mouse — Mtap^+/− × Myc (PLoS One 2013, "Germline mutations in Mtap cooperate with Myc to accelerate tumorigenesis in mice"). Demonstrates cooperativity between Mtap loss and MYC in tumor acceleration.

Model organism databases: MGI (MGI:1914152); IMPC/KOMP/IMSR for allele availability; Alliance of Genome Resources for orthology.

15.2 dismech animal_models shape

animal_models:
- name: Mtap heterozygous knockout mouse (Mtap+/lacZ)
  species: Mouse
  genotype: Mtap<lacZ> heterozygous null
  publication: PMID:19567676
  modeled_mechanisms:
  - target: MTAP Loss of Function
    relationship: PARTIALLY_RECAPITULATES
    fidelity: LOW
    description: >-
      Heterozygous Mtap disruption in mouse establishes MTAP as a tumor
      suppressor in vivo, supporting the tumor-predisposition arm of DMS-MFH.
    limitations: >-
      The mouse develops T-cell lymphoma, not bone dysplasia or bone sarcoma,
      and dies at a median of 585 days. Critically, the mouse Mtap ortholog
      lacks exons 9-11 entirely: those exons derive from primate-restricted
      retroviral integrations approximately 40 million years old, so the
      DMS-MFH splicing lesion cannot be modeled in any non-primate species.
      A whole-gene knockout is therefore not the same lesion class as the
      human exon-9 splicing mutation.

A companion discussions entry of kind: HUMAN_MODEL_MISMATCH should attach to the MTAP pathophysiology node, with the prompt framed as: does whole-gene Mtap knockout in a species lacking the retroviral-derived exon-9 isoform layer inform the human exon-9 splicing disease at all?

15.3 Cellular and in vitro models

Available and used in the primary paper: - Patient‑derived fibroblast lines — 3′ RACE isoform discovery, splicing analysis. - Patient‑derived lymphoblastoid lines — same. - A patient‑derived tumor cell line established by the authors — the only DMS‑MFH tumor line in existence, as far as I can determine. Its availability (ATCC/Cellosaurus deposition) is not stated; this is worth chasing for anyone planning work. - MTAP minigene constructs — three constructs (WT, c.813-2A>G, c.885A>G) spanning ~11.5 kb of genomic sequence; the definitive functional assay for the splicing mechanism, and the model system a new lab would most plausibly rebuild. - Heterologous isoform expression + MTAP enzymatic assay — established that v1, v2, v3 have activity and v4, v5, v6 do not, including under MG132 proteasome inhibition. - Co‑immunoprecipitation — established physical interaction of all six isoforms with archetype MTAP. - MCF7 — MTAP‑null breast adenocarcinoma line used historically for MTAP re‑expression growth‑suppression experiments.

Not available: no iPSC line, no organoid, no osteoblast‑differentiation model, no NAM/organ‑chip system, no isogenic CRISPR knock‑in of the exon‑9 variants in a human osteoblastic or mesenchymal background. The last of these is the single most obvious missing experiment — a CRISPR knock‑in of c.885A>G or c.813-2A>G into human iPSCs differentiated toward osteoblast lineage would be the first model capable of addressing the bone‑dysplasia arm at all.

15.4 Computational models

Molecular modelling of the MTAP trimer on PDB 1CG6/1CB0, examining splice‑variant insertion points (after K271 for v1‑type, after A230 for v4‑type) relative to the trimer interface and MTA active site. Rendered in O v.13 and PyMOL v1.3. Secondary‑structure and disulfide‑bond prediction suggested an exon‑9/exon‑10 cysteine disulfide, explicitly flagged by the authors as requiring biochemical validation. No kinetic, systems‑biology, or whole‑cell metabolic model of the MTAP salvage pathway in this disease exists.

15.5 Research applications and limitations

Can be studied: MTAP isoform biology, splicing regulation, exaptation of retroviral sequence, MTAP tumor‑suppressor function, MTA as a circulating biomarker.

Cannot currently be studied in any model: the bone dysplasia itself, the marrow infarction, the impaired fracture healing, the myopathy, and the tissue‑specificity of the disease (why bone and muscle, when MTAP is ubiquitously expressed?). That last question is arguably the central unanswered question of the disease and has no experimental system attached to it.

Appendix A — Consolidated ontology term suggestions

All IDs below were verified against this repository's ontology adapters (OAK sqlite:obo:hp / :go / :cl / :uberon / :chebi) on 2026-08-29, except the NCIT terms, which are taken from the verified table in CLAUDE.md.

Disease: MONDO:0007205 diaphyseal medullary stenosis-bone malignancy syndrome

Gene: hgnc:7413 MTAP

Phenotypes (HP): HP:0100254 · HP:0005045 · HP:0031367 · HP:0005686 · HP:0002756 · HP:0003084 · HP:0002979 · HP:0000938 · HP:0005010 · HP:0002754 · HP:0002653 · HP:0012315 · HP:0002669 · HP:0100244 · HP:0003198 · HP:0003325 · HP:0003701 · HP:0003202 · HP:0007819 · HP:0000963 · HP:0000977 · HP:0000978 · HP:0002216 · HP:0000006 (inheritance)

Biological processes (GO): GO:0071267 L-methionine salvage · GO:0043101 purine-containing compound salvage · GO:0006595 polyamine metabolic process · GO:0000380 alternative mRNA splicing, via spliceosome · GO:0019918 peptidyl-arginine methylation, to symmetrical-dimethyl arginine · GO:0032259 methylation · GO:0001649 osteoblast differentiation · GO:0001503 ossification · GO:0030282 bone mineralization

Molecular function (GO): GO:0017061 S-methyl-5-thioadenosine phosphorylase activity

Cell types (CL): CL:0000062 osteoblast · CL:0000137 osteocyte · CL:0000092 osteoclast · CL:0000134 mesenchymal stem cell (or CL:2000079 mesenchymal stem cell of femoral bone marrow) · CL:0000057 fibroblast · CL:0008002 skeletal muscle fiber

Anatomy (UBERON): UBERON:0004769 diaphysis · UBERON:0001438 metaphysis · UBERON:0002495 long bone · UBERON:0001439 compact bone tissue · UBERON:0009859 endosteum · UBERON:0002371 bone marrow · UBERON:0002484 bone marrow cavity · UBERON:0000981 femur · UBERON:0000979 tibia · UBERON:0001446 fibula · UBERON:0000976 humerus · UBERON:0000965 lens of camera-type eye

Chemical entities (CHEBI): CHEBI:17509 5'-S-methyl-5'-thioadenosine · CHEBI:15414 S-adenosyl-L-methionine · CHEBI:16708 adenine · CHEBI:16643 L-methionine · CHEBI:16610 spermidine · CHEBI:17148 putrescine · CHEBI:28748 doxorubicin · CHEBI:27899 cisplatin · CHEBI:5864 ifosfamide · CHEBI:44185 methotrexate

Treatments (NCIT): NCIT:C15632 Chemotherapy · NCIT:C15329 Surgical Procedure · NCIT:C16186 Orthopedic Surgical Procedure · NCIT:C15313 Radiation Therapy · NCIT:C15302 Physical Therapy · NCIT:C15315 Rehabilitation · NCIT:C15747 Supportive Care · NCIT:C15240 Genetic Counseling

Appendix B — Reference list with quotable snippets

References marked ✅ have been fetched into this worktree's references_cache/ and their snippets verified as exact substrings.

Cache commands to run before curating evidence items from the unfetched rows:

just fetch-reference PMID:19567676
just fetch-reference PMID:16244874
just fetch-reference PMID:16419137
just fetch-reference PMID:3745248
just fetch-reference PMID:39293516
just fetch-reference PMID:35184191
just fetch-reference ORPHA:85182   # note: not present in the current Orphanet cache

Appendix C — Explicit gaps and cautions for the curator

1. The gene–disease relationship is contested. ClinGen HI score 1, PanelApp Red. Curate the mechanism, but record the dissent — ideally as a discussions entry with kind: KNOWLEDGE_GAP attached to the MTAP pathophysiology node.
2. No independent replication since 2012. No new family, no independent variant, no confirmatory functional study in 14 years.
3. The variants are not in canonical transcript coordinates. c.885A>G/c.813-2A>G are numbered against EST clone AK309365, not NM_002451.4. The MTAP ClinVar records returned by a condition search are unrelated VUS submissions — do not cite them.
4. Do not curate the elevated serum MTA as a diagnostic test. n=2 affected, research assay, no reference interval, no LOINC code.
5. Do not curate PRMT5 inhibitors as a treatment. The mechanism is suggestive; the evidence in this disease is zero, and the germline lesion is not an MTAP deletion. mechanistic_hypotheses with status: EMERGING is the correct home.
6. Do not curate early CAD as a phenotype. It is unpublished anecdote explicitly flagged as such by the authors.
7. Family-specific features need subtype qualifiers. Myopathy (families 4, 5), thin skin / premature graying / hernias / clotting abnormality (family 5 only) should not be presented as disease-wide.
8. The bone dysplasia mechanism is unknown. This is the highest-value knowledge gap in the entry — the causal chain from MTAP isoform imbalance to endosteal cortical thickening and marrow infarction is entirely unspecified.
9. No non-primate model can carry the lesion. Record as HUMAN_MODEL_MISMATCH.
10. The research/ directory rule applies. This report is analysis, not deep-research-provider output — if it is committed, it belongs under docs/curation-notes/ or docs/reports/, not research/.

Sources

• OMIM #112250 — Diaphyseal medullary stenosis with malignant fibrous histiocytoma
• OMIM *156540 — Methylthioadenosine phosphorylase; MTAP
• Orphanet — Diaphyseal medullary stenosis-bone malignancy syndrome (ORPHA:85182)
• Orphadata cross-referencing API — ORPHA 85182
• MONDO:0007205 — Monarch Initiative
• HPO / ontology.jax.org annotation API — OMIM:112250
• MedGen C1862177 — Diaphyseal medullary stenosis-bone malignancy syndrome
• GARD 10072 — Diaphyseal medullary stenosis-bone malignancy syndrome
• Disease Ontology DOID:0080664
• Camacho-Vanegas et al. 2012, Am J Hum Genet — PMID:22464254
• Martignetti et al. 1999, Am J Hum Genet — PMID:10053015
• Norton et al. 1996, Pediatr Radiol — PMID:8781110
• Watts et al. 2005, Hum Genet — PMID:16244874
• Mehta et al. 2006, Am J Med Genet A — PMID:16419137
• Hardcastle et al. 1986 — PMID:3745248
• Arnold 1973, Ann Intern Med
• Henry et al. 1958, Can Med Assoc J — PMID:13511301
• Kadariya et al. 2009, Cancer Res — Mtap heterozygous mice
• Germline mutations in Mtap cooperate with Myc — PLOS One
• MGI:1914152 — mouse Mtap
• ClinGen Dosage Sensitivity — MTAP (HGNC:7413)
• Genomics England PanelApp — MTAP, Childhood solid tumours
• ClinVar RCV000310131 — MTAP c.*2837T>A
• ClinVar RCV000317123 — MTAP c.566G>T
• UniProt Q13126 — MTAP
• Structure of human MTAP at 1.7 Å — PMID:10404592
• AMG 193 phase I first-in-human, Ann Oncol 2024 — PMID:39293516
• AMG 193, Cancer Discovery 2025
• MTAP deletion as a therapeutic vulnerability in cancer, Int J Mol Sci 2025
• A review of the known MTA-cooperative PRMT5 inhibitors, RSC Adv 2024
• UPSB in the COSS Group, J Cancer Res Clin Oncol 2026
• UPS of bone population-based cohort — PMID:35184191
• NCI PDQ — Osteosarcoma and Undifferentiated Pleomorphic Sarcoma of Bone Treatment
• Undifferentiated Pleomorphic Sarcoma — StatPearls
• NCT00007046 — Genetic Study of Patients and Families With DMS-MFH
• Atlas of Genetics and Cytogenetics in Oncology and Haematology — DMS-MFH
• BoneTumor.org — Diaphyseal Medullary Stenosis with Malignant Fibrous Histiocytoma

Next step: the draft kb/disorders/Diaphyseal_Medullary_Stenosis_With_Malignant_Fibrous_Histiocytoma.yaml in this worktree is currently a placeholder skeleton. Say the word and I'll populate it from this report — starting with the pathophysiology chain in §6.1 and the evidence items in Appendix B, then running just validate, just count-verified-snippets, and just validate-terms on it.

Reference Validation

Checked with linkml-reference-validator 0.2.1.

All extracted references resolved successfully.

Term Validation

Checked with linkml-term-validator 0.4.5, through the ols: adapter.

Terms the report names something else

These identifiers resolve, so nothing about them looks wrong, and the ontology calls them something unrelated to what the report calls them. That usually means the identifier is not the one the sentence needs:

• MONDO:0007205 (6 mentions) - the report calls it "MONDO", "Monarch Initiative"; MONDO calls it diaphyseal medullary stenosis-bone malignancy syndrome
• DOID:0080664 (4 mentions) - the report calls it "Disease Ontology"; DOID calls it diaphyseal medullary stenosis with malignant fibrous histiocytoma
• UBERON:0009859 (2 mentions) - the report calls it "Endosteum — site of pathological apposition"; UBERON calls it endosteum
• NCIT:C15329 (4 mentions) - the report calls it "For osteomyelitis complicating nonunion, or for unresectable tumor", "Presenile cataracts"; NCIT calls it Surgical Procedure

Terms whose name is worth a second look

The report's name for these is recognisably related to the term's own name without being one of them. A loose paraphrase reads the same way as a citation of the wrong sibling term - and so does a related synonym, which the ontology records precisely because it names something adjacent rather than the same thing - so these are listed rather than judged:

• UBERON:0002371 (3 mentions) - the report calls it "Bone marrow — site of infarction"; UBERON calls it bone marrow
• HP:0012315 (2 mentions) - the report calls it "Histiocytoma (MFH → UPS)"; HP calls it Histiocytoma
• HP:0100244 (2 mentions) - the report calls it "Fibrosarcoma (of bone)"; HP calls it Fibrosarcoma
• UBERON:0001439 (2 mentions) - the report calls it "Compact (cortical) bone tissue — thickened endosteally"; UBERON calls it compact bone tissue, and lists "cortical bone tissue" among its other names
• UBERON:0002484 (2 mentions) - the report calls it "Bone marrow cavity — stenosed/obliterated"; UBERON calls it bone marrow cavity

Terms named inconsistently

The report gives these identifiers more than one name of its own:

• MONDO:0007205 - called "MONDO", "Monarch Initiative"
• NCIT:C15329 - called "For osteomyelitis complicating nonunion, or for unresectable tumor", "Presenile cataracts"

Prefixes with no resolver

Terms carrying these prefixes were not checked either way, because no configured ontology covers them. An unrecognised prefix may name an ontology this run could not reach as easily as one that does not exist, so nothing here is evidence of fabrication: ORPHA, MGI, OMIM.