Diets-Jongmans Syndrome (DIJOS) — Comprehensive Research Report

1. Disease Information

Overview. Diets-Jongmans syndrome (DIJOS) is a rare, autosomal dominant neurodevelopmental/malformation syndrome caused by heterozygous pathogenic variants in KDM3B (lysine demethylase 3B, chromosome 5q31.2). It is characterized by mild-to-moderate intellectual disability/developmental delay, short stature, neonatal feeding difficulties, and a recognizable facial gestalt (long ears, broad/prominent nasal tip with low columella, thin upper lip, wide mouth, pointed chin). It was first delineated as a discrete syndrome by Diets, van der Donk, Baltrunaite et al. in the American Journal of Human Genetics in 2019 (PMID:30929739), using trio exome sequencing, GeneMatcher-facilitated international collaboration, and computational facial-recognition analysis (Face2Gene/Clinical Face Phenotype Space) to establish gestalt validity.

Key identifiers: - OMIM: #618846 — DIETS-JONGMANS SYNDROME; DIJOS (omim.org/entry/618846) - Gene OMIM: KDM3B, #609373 - MONDO: MONDO:0030012 (confirmed via GenCC submission, consistent with the MONDO ID supplied in the research request) - Orphanet: ORPHA:633004 — "KDM3B-related intellectual disability-facial dysmorphism-short stature syndrome" (orpha.net/en/disease/detail/633004) - MedGen: C5394263 (ncbi.nlm.nih.gov/medgen/1714920) - HGNC (gene): HGNC:1337 (KDM3B; previous symbols C5orf7, JMJD1B) - GeneCC/ClinGen: Definitive gene-disease relationship, "Syndromic intellectual disability" (MONDO:0000508), Intellectual Disability and Autism GCEP, curated 2024-02-21 (search.clinicalgenome.org/kb/genes/HGNC:1337); GenCC entry independently lists KDM3B–Diets-Jongmans syndrome (MONDO:0030012) as Definitive, autosomal dominant, submitted by Ambry Genetics, evaluated 2024-04-03. - No GeneReviews chapter currently exists* for this condition (searched; not found as of this report). GTR lists the condition under C5394263.

Synonyms: "KDM3B-related intellectual disability-facial dysmorphism-short stature syndrome" (Orphanet's working name); no other widely used alternate names were found. Caution: an AI-generated web summary encountered during this search expanded "DIJOS" as an acronym for "developmental delay, intellectual disability, joint contractures... immunodeficiency... short stature" — this expansion could not be traced to any primary source (OMIM, the 2019 AJHG paper, or Orphanet) and should be treated as unverified/likely fabricated rather than used in curation.

Evidence base composition: The evidence base is aggregated disease-level literature (case series and case reports pooled across cohorts), not a single large EHR-derived or registry cohort. As of this search, published literature totals roughly 27+ molecularly confirmed individuals across the 2019 founding cohort (17 individuals, PMID:30929739) plus at least 9 subsequently published individuals in independent case reports/series (see §9 for citations), so no large single denominator exists — frequencies below are cohort-level percentages from the cited sources, not population data.

2. Etiology

Disease causal factor: Purely genetic — heterozygous (predominantly de novo) pathogenic/likely pathogenic variants in KDM3B. No infectious, toxic, or purely environmental cause is described.

Genetic risk factors: - Heterozygous missense, nonsense, frameshift, and splice-site variants in KDM3B, clustering in/near the zinc-finger DNA-binding domain and the C-terminal catalytic Jumonji-C (JmjC) domain (PMID:30929739). - KDM3B is highly constrained against loss-of-function variation: original 2019 report cites ExAC pLI = 1.00 and Z-score = 4.99 (missense constraint); ClinGen's current summary similarly reports pLI ≈ 1.0 and a low LOEUF (~0.12), consistent with strong haploinsufficiency intolerance (ClinGen KDM3B page). - No modifier genes have been formally identified; a co-occurring SIN3A variant was reported alongside a KDM3B variant in one atypical pulmonary-hypertension case (PMID:38314229), but this is a single dual-diagnosis observation, not an established modifier relationship.

Environmental risk factors: None established; this is a monogenic disorder.

Protective factors: None reported in the literature to date.

Gene-environment interactions: Not described. The single case with rhGH (recombinant human growth hormone) exposure (PMC13025784) is a treatment-response observation, not a G×E etiologic interaction.

3. Phenotypes

Developmental / Cognitive

• Intellectual disability / developmental delay — present in 16/17 (94.1%) of the founding cohort (PMID:30929739); pooled across later literature (27 patients, PMC13025784) 81.5%. Predominantly mild (47.1%) to mild-moderate (23.5%); severe ID is uncommon. Speech/language delay is typically more pronounced than motor delay.
• HP:0001249 (Intellectual disability); HP:0011342 (Mild intellectual disability); HP:0001263 (Global developmental delay); HP:0000750 (Delayed speech and language development)
• Motor delay — 70.4% (pooled, PMC13025784).
• HP:0001270 (Motor delay)
• Childhood hypotonia — 38.5% (PMID:30929739).
• HP:0001252 (Hypotonia)
• Behavioral problems — 53.3%; ADHD 18.5–26.7%; autism spectrum disorder 14.8–20%. A 2025 report of two Chinese patients specifically highlights autistic features and global developmental delay as presenting complaints (PMID:41408900).
• HP:0000708 (Behavioral abnormality); HP:0007018 (Attention deficit hyperactivity disorder); HP:0000717 (Autism)
• Epilepsy — 20% (two transient cases) (PMID:30929739).
• HP:0001250 (Seizure)
• Variable neuropsychological profile within families — a 2026 case series of four family members carrying the same variant reported a "variable neuropsychological profile," i.e., marked intrafamilial variability in cognitive presentation despite an identical KDM3B variant (Archives of Clinical Neuropsychology, PMID:42186234; full abstract text could not be retrieved due to access restrictions — treat as a lead pending direct verification).

Growth

• Short stature — ~50–55.6% of patients (<-2 to -2.5 SD), with height <-1.5 SD in 68.8% (PMID:30929739; PMC13025784).
• HP:0004322 (Short stature)
• Low birth weight / postnatal growth restriction — 59.3% pooled (PMC13025784); one recent 4-patient case series found growth retardation to be the predominant/sole clinical feature with minimal neurodevelopmental involvement, expanding phenotypic heterogeneity (PMC13025784, MDPI Genes 2026, doi:10.3390/genes17030294).
• HP:0001518 (Small for gestational age); HP:0004325 (Decreased body weight)
• Growth hormone deficiency — documented and biochemically confirmed (stimulation testing) in at least two independently reported cases (PMC13025784; Turkish case, PMID:41878155/PMC13008413), described as "rarely reported" previously.
• HP:0000824 (Growth hormone deficiency)
• Microcephaly-range head circumference reduction — reported in the Turkish case (head circumference SDS -3.96) (PMC13008413).
• HP:0000252 (Microcephaly)

Feeding / Gastrointestinal

• Neonatal feeding difficulties — 51.9–60% (PMID:30929739; PMC13025784).
• HP:0011968 (Feeding difficulties)
• Umbilical/inguinal hernia — 12.5% each; diaphragmatic hernia — 12.5% (PMID:30929739).
• HP:0001537 (Umbilical hernia); HP:0000023 (Inguinal hernia); HP:0000776 (Congenital diaphragmatic hernia)
• Duodenal atresia — reported in one individual (PMID:30929739).
• HP:0002247 (Duodenal atresia)

Musculoskeletal

• Joint hypermobility — 31.3% (PMID:30929739).
• HP:0001382 (Joint hypermobility)
• Pectus excavatum — reported in a Chinese case (PMID:41914422).
• HP:0000767 (Pectus excavatum)

Craniofacial (facial gestalt — see also §4 for facial-recognition analysis)

• Long ears — 52.9%; broad/prominent nasal tip — 82.4%; wide mouth — 70.6%; thin upper lip vermilion — 70.6%; pointed/prominent chin — 70.6%; low columella — 35.3% (PMID:30929739).
• HP:0009884 (Long ear); HP:0000455 (Broad nasal tip); HP:0000154 (Wide mouth); HP:0000219 (Thin upper lip vermilion); HP:0000307 (Prominent chin); HP:0011832 (Low columella)
• Non-neutral palpebral fissures — 4/16 upslanting, 5/16 downslanting (PMID:30929739).
• HP:0000582 (Upslanted palpebral fissure); HP:0000494 (Downslanted palpebral fissure)
• Micrognathia (Turkish case) — HP:0000347.

Ophthalmologic / Otologic

• Eye abnormalities (nystagmus, strabismus, refractive error) — 31.3% (PMID:30929739). HP:0000639 (Nystagmus), HP:0000486 (Strabismus).
• Hearing loss — 23.5% (PMID:30929739). HP:0000365 (Hearing impairment).

Genitourinary

• Cryptorchidism, hypospadias — 12.5% each (PMID:30929739). HP:0000028 (Cryptorchidism); HP:0000047 (Hypospadias).
• Oligospermia — documented in one adult male individual with the Kdm3b-knockout mouse subfertility phenotype cited as biological plausibility (PMID:30929739).

Cardiac / Vascular

• Ventricular septal defect, heterotaxy, interrupted inferior vena cava (individual reports, PMID:30929739). HP:0001629 (VSD); HP:0001696 (Heterotaxy).
• Cardiomyopathy and neutropenia — newly reported as expanded phenotypes in a 2023 report of two patients (one with de novo splice variant c.5070+1G>A, first reported KDM3B splice variant; one with c.2828G>A p.R943Q), one presenting with palpitation/dyspnea/edema, the other with recurrent URIs and transient neutropenia (PMID:36757469, Neurogenetics 2023).
• HP:0001638 (Cardiomyopathy); HP:0001875 (Neutropenia)
• Pulmonary arterial hypertension / childhood interstitial lung disease (chILD) — reported in a 10-year-old boy (Sudanese) with a novel KDM3B missense variant (c.4406T>C, p.Ile1469Thr) who also carried a SIN3A variant (c.1381G>A, p.Gly461Arg); described as "the second described case worldwide" of PAH associated with a KDM3B mutation (Benslimane et al., Pan Afr Med J 2023, PMID:38314229). Caution: dual-variant case — causal attribution to KDM3B alone is not established.
• HP:0002092 (Pulmonary hypertension)

Oncologic

• Malignancy — 2/17 (11.8%) in the founding cohort: acute myeloid leukemia and Hodgkin lymphoma (PMID:30929739). The authors explicitly flag ascertainment bias (both patients were recruited through a cancer-predisposition study) and recommend clinical vigilance rather than a formal surveillance protocol. This is biologically plausible given that somatic KDM3B alterations and Kdm3b-knockout-mouse myelodysplastic-syndrome-like hematopoietic defects are independently described (see §6).
• HP:0004377 (AML — HP:0004808 more precisely "Acute myeloid leukemia"); HP:0012156 (Hodgkin lymphoma, if using more specific term) — verify exact HPO IDs against the cache before binding.

Frequency caveat

All percentages above are drawn from small, ascertainment-biased cohorts (primarily clinical-genetics referral for developmental delay); true population frequencies are unknown given the extreme rarity of the condition.

4. Genetic / Molecular Information

Causal gene: KDM3B (Lysine Demethylase 3B; previous symbols JMJD1B, C5orf7), HGNC:1337, chromosome 5q31.2, gene OMIM #609373.

Pathogenic variant spectrum (as of the most recent literature review, PMC13025784, MDPI Genes 2026): - Across 27 pooled reported patients: 23 distinct KDM3B variants — 12 missense, 5 nonsense, 4 frameshift, 2 splice-site; 4 variants recurrent. - Original 2019 cohort (17 individuals): 14 distinct variants — 5 truncating (nonsense) and other missense; "All 14 different variants found were absent from the gnomAD and ExAC databases," and "13/14 had high CADD scores (13–40)" (PMID:30929739). - Variants cluster in/near the zinc-finger domain (DNA-binding) and the Jumonji-C (JmjC) catalytic domain responsible for H3K9 demethylation; missense variants affect residues that are "evolutionarily highly conserved across other species," including conservation with paralogs KDM3A and JMJD1C. - MetaDome analysis of missense variant positions found several located in regions "intolerant" or "highly intolerant for missense variation," supporting pathogenicity (PMID:30929739). - Representative individual variants across reports: p.Glu93 (inherited, mother-to-child), p.Asp336Gly (inherited, father-to-child), p.Glu1731Lys (inherited, father-to-child), c.2743_2745del p.(Arg915del) (ClinVar RCV002468724), c.5147T>C p.(Leu1716Pro) (PMID:41914422, de novo), c.5068C>T p.(Gln1690Ter) (Turkish case, PMC13008413, inherited from mildly-affected mother), c.5070+1G>A (first reported splice variant, de novo, PMID:36757469), c.2828G>A p.(Arg943Gln) (PMID:36757469), c.1970C>G p.(Ser657) and c.3973-1G>C (2025 Chinese cohort, PMID:41408900), c.2832-3C>G, c.1188del p.(Glu397Argfs21), c.4580T>C p.(Leu1527Pro), c.3220dup p.(Glu1074Glyfs48) (4-patient growth-predominant series, PMC13025784), c.4406T>C p.(Ile1469Thr) (chILD/PAH case, co-occurring with SIN3A, PMID:38314229).

ACMG/ClinVar classification: Individual variants are typically classified pathogenic/likely pathogenic under ACMG/AMP criteria on a case-by-case basis (e.g., ClinVar RCV002468724 for p.Arg915del); no single hotspot dominates.

Population frequency: All reported disease-causing variants are absent from gnomAD/ExAC, consistent with a severely constrained gene (pLI ≈ 1.0, LOEUF ≈ 0.12 per ClinGen's summary) and de novo dominant disease mechanism.

Somatic vs. germline: DIJOS itself is caused exclusively by germline variants. Note (important for the dismech "keep germline syndromes separate from somatic tumors" convention): KDM3B also carries somatic alterations reported in hematologic malignancies (independent of DIJOS) — the 2019 authors explicitly cite this as biological plausibility for the myeloid malignancy signal in their germline cohort, but somatic KDM3B cancer biology is a distinct body of literature not curated here.

Functional consequence / mechanism: The 2019 authors state: "We identified nonsense variants in five individuals, implying that haploinsufficiency is the most likely mechanism of disease in our individuals as well" (PMID:30929739) — i.e., predominantly loss-of-function via haploinsufficiency, consistent with the gene's extreme LoF intolerance. Direct biochemical confirmation was attempted but inconclusive: "We assessed the mono- and tri-methylation status of H3K9 in HEK293 cells that were overexpressing wild-type and missense mutants of KDM3B, but these experiments were inconclusive due to large variability in expression levels" (PMID:30929739). A later case report (PMID:41914422) proposes that a specific missense variant (p.Leu1716Pro) in the JmjC domain "disrupts hydrogen bonding, reducing protein stability," a structural (not functionally assayed) inference.

Genotype-phenotype correlation: The 2025/2026 literature-review case series (PMC13025784) notes that loss-of-function (truncating) variants "may confer a relatively milder neurodevelopmental phenotype" compared to missense variants — an emerging but not yet firmly established correlation; label as a hypothesis from a small pooled series, not a settled rule.

Modifier genes / epigenetics: No modifier genes established. KDM3B itself is an epigenetic "writer/eraser" — a chromatin-modifying enzyme — making DIJOS itself a disorder of epigenetic regulation (a "chromatinopathy"), rather than a disorder with reported secondary epigenetic dysregulation as a downstream mechanism.

Chromosomal abnormalities: No microdeletion/microduplication or translocation mechanism is described for DIJOS; all reported cases are single-nucleotide/small-indel/splice variants.

5. Environmental Information

Not applicable — DIJOS is a purely monogenic disorder. No environmental, lifestyle, or infectious contributing factors are described in the literature reviewed.

6. Mechanism / Pathophysiology

Ordered causal chain (from initiating lesion to clinical manifestation)

1. Heterozygous de novo (occasionally inherited) loss-of-function or missense variant in KDM3B (chr5q31.2) → disrupts or destabilizes the KDM3B protein's zinc-finger DNA-binding domain or its C-terminal JmjC catalytic domain (demonstrated by variant clustering and conservation analysis; PMID:30929739). Inferred structural consequence for individual missense variants (e.g., p.Leu1716Pro disrupting hydrogen bonding), not directly assayed in most cases (PMID:41914422).
2. Truncating variants (≈5/14–5/23 of reported variants) most plausibly cause haploinsufficiency — reduced dosage of functional KDM3B protein — stated explicitly by the discovery cohort as "the most likely mechanism of disease" (PMID:30929739). This step is a genetic-mechanism inference from variant type and gene constraint (pLI≈1.0), not from a direct dosage/protein-level assay in patient tissue.
3. Reduced KDM3B H3-lysine-9 (H3K9) demethylase activity → failure to remove repressive H3K9me1/me2 marks at KDM3B target loci. This step is inferred by homology/mechanism (KDM3B is a well-characterized JmjC-domain H3K9 demethylase; UniProt DI-05814) rather than directly measured in DIJOS patient cells — the authors' own attempt to assay H3K9 methylation status in an overexpression cell system was "inconclusive due to large variability in expression levels" (PMID:30929739). This is the single largest evidentiary gap in the human mechanistic chain and should be flagged as a knowledge gap / HUMAN_MODEL_MISMATCH candidate when curated, since the mechanistic step downstream of variant identification rests on mouse-model and biochemical-homology inference rather than confirmed patient-tissue chromatin data.
4. Branch A — Transcriptional dysregulation: Because H3K9 methylation is a repressive chromatin mark, loss of KDM3B demethylase activity is predicted by the authors to result in failure to activate transcription at target loci — stated as: "methylation of H3K9 is generally considered to be a repressive mark, demethylation of H3K9 by KDM3B is predicted to result in the activation of transcription and thus might explain subsequent growth retardation" (PMID:30929739). This transcriptional consequence is itself inferred from the general biology of H3K9 methylation, not measured via patient-derived transcriptomics.
5. Branch A, continued — Neurodevelopmental consequence: transcriptional dysregulation during neurodevelopment (via disrupted chromatin regulation of neuronal gene programs) → intellectual disability, developmental delay, hypotonia, behavioral/autistic features. A separate line of mouse evidence supports a specific downstream neuro-mechanism: cerebellar studies indicate "diminished H3K9 demethylase activity impairs cerebellar memory consolidation" and that KDM3B "plays an essential role in motor learning" (cited in PMC13025784) — this is model-organism (mouse) evidence whose translational fidelity to human DIJOS cognitive/motor phenotypes has not been directly tested (a HUMAN_MODEL_MISMATCH-type gap).
6. Branch B — Growth/endocrine consequence: Mouse Kdm3b-knockout studies (Kim et al., PMID:25892958, and related work) show that Kdm3b ablation decreases hepatic/renal IGFBP-3 expression → accelerated IGF-1 degradation → ~36% reduction in circulating IGF-1 → postnatal somatic growth restriction. The 2019 human cohort explicitly extrapolates this mouse mechanism to explain patient short stature: "kdm3b-knockout mice present with a somatic growth restriction, possibly explained by decreased concentrations of insulin-like growth factor-1 (IGF1)" (PMID:30929739) → clinically, this is corroborated in humans by two independently reported cases with biochemically confirmed growth hormone deficiency on stimulation testing (PMC13025784; PMC13008413) → short stature, and rhGH-responsive growth improvement documented in one patient (94.3 cm → 135 cm over 6 years) (PMC13025784).
7. Branch C — Hematopoietic/reproductive consequence: Kdm3b-knockout mice independently show (i) hematopoietic-system defects "consistent with myelodysplastic syndrome" (PMID:30929739, citing prior mouse literature) and (ii) male subfertility with reduced sperm reserves and motility (PMID:26681924) and female reproductive defects (irregular estrous cycles, reduced ovulation/fertilization rates, decreased circulating estradiol) (PMID:25892958). In humans, this maps to: (a) two malignancies (AML, Hodgkin lymphoma) among 17 founding-cohort patients — flagged by the authors as possibly ascertainment-biased but biologically plausible given known somatic KDM3B alterations in hematologic malignancy, and (b) one adult male patient with documented oligospermia, cited by the authors as "in accordance" with the mouse phenotype (PMID:30929739).
8. Branch D (peripheral, weakly evidenced) — Isolated reports link KDM3B variants to cardiomyopathy and transient neutropenia (PMID:36757469) and to pulmonary arterial hypertension/childhood interstitial lung disease (PMID:38314229, confounded by a co-occurring SIN3A variant). No mechanistic chain has been proposed connecting KDM3B loss-of-function to cardiac or pulmonary vascular pathophysiology; these should be treated as emerging, single/dual-case-level phenotype associations rather than established downstream nodes of the pathway above.

Category checklist detail

• Molecular pathways: Chromatin/epigenetic regulation — H3K9 demethylation (JmjC/Jumonji-domain demethylase family); downstream transcriptional activation of KDM3B target genes (specific target gene identities in human neurodevelopment not established in the cited literature).
• GO:0032454 (histone H3-K9 demethylase activity, note-check exact ID against GO); GO:0006338 (chromatin remodeling, verify).
• Cellular processes: Chromatin remodeling / epigenetic gene regulation; (mouse) hematopoietic differentiation; (mouse) germ cell/oocyte and spermatocyte maturation.
• Protein dysfunction: Loss-of-function via haploinsufficiency (predominant model per truncating variant burden); for missense variants, structural destabilization within the JmjC catalytic domain or zinc-finger domain is proposed but not biochemically confirmed for most reported variants.
• Protein domains: N-terminal zinc-finger DNA-binding domain, LXXLL nuclear-receptor-interaction motif, C-terminal JmjC catalytic domain (per PMC13025784 summary; verify against UniProt directly before binding InterPro/Pfam IDs).
• Metabolic changes: IGF-1/IGFBP-3 axis dysregulation (mouse-demonstrated; human growth hormone deficiency documented but the human GH-axis lesion has not been mechanistically linked back to KDM3B at the molecular level — it is a phenotypic correlation).
• Immune system involvement: No autoimmune or immunodeficiency mechanism is established in the primary literature (see caution above about the unverified "immunodeficiency" acronym expansion). Neutropenia is reported in a single case (transient) (PMID:36757469).
• Tissue damage mechanisms: Not specifically characterized; no oxidative stress/fibrosis/necrosis mechanism has been reported for DIJOS itself (the pulmonary fibrosis findings in the chILD case report are radiologic/histologic descriptions, not mechanistic data, and are confounded by co-occurring SIN3A variant).
• Biochemical abnormalities: Reduced/absent H3K9 demethylase enzymatic activity (inferred from gene function and variant type, not directly measured in patient tissue in the cited studies); biochemically confirmed GH deficiency in ≥2 patients on stimulation testing.
• Epigenetic changes: DIJOS is itself a primary disorder of an epigenetic "eraser" enzyme (a chromatinopathy) rather than a disorder with reported secondary/consequential DNA methylation changes; no patient-derived methylome (e.g., EpiSign) data specific to DIJOS were identified in this search — this would be a reasonable knowledge-gap flag (an "episignature" study has not been reported for KDM3B, unlike many other chromatinopathies).
• Molecular profiling (transcriptomics/proteomics/single-cell): No patient-derived RNA-seq, proteomics, or single-cell data specific to DIJOS were found in this search.
• Advanced technologies: The founding study used computational facial-recognition/facial-gestalt analysis (CFPS + OpenFace hybrid model) rather than an -omics technology, to statistically validate the facial phenotype: "Clustering improvement factors (CIFs)...demonstrated that there is a significant similarity between the faces of individuals with a germline variant in KDM3B when compared to the general ID population (CIF = 2.5962; p = 0.0054)" (PMID:30929739).

Suggested cell types (CL) / anatomical (UBERON) / GO terms for pathograph nodes (to be independently verified via OAK before binding, per dismech-terms rules — none of the CURIEs below have been looked up live in this session and must be treated as leads): - GO: histone H3-K9 demethylation, chromatin organization, regulation of transcription by RNA polymerase II. - CL: neuron (generic), hematopoietic stem cell, chondrocyte/growth-plate chondrocyte (growth phenotype), spermatid/spermatocyte (mouse fertility phenotype). - UBERON: cerebellum, growth plate/epiphyseal cartilage, bone marrow, pituitary/hypothalamic-pituitary-somatotropic axis, testis.

7. Anatomical Structures Affected

• Primary organs/systems: CNS (cognition, motor learning, cerebellum in mouse studies), skeletal growth system (short stature), craniofacial skeleton/soft tissue (facial gestalt), endocrine (GH axis).
• Secondary/complication-level involvement: Hematopoietic system (neutropenia, malignancy signal), cardiovascular system (cardiomyopathy, VSD, heterotaxy, IVC anomaly), pulmonary vasculature (PAH/chILD — confounded case), gastrointestinal (duodenal atresia, hernias, diaphragmatic hernia), genitourinary (cryptorchidism, hypospadias, oligospermia), ophthalmologic (strabismus, nystagmus, refractive error), otologic (hearing loss).
• Tissue/cell level: Neurons and glia (neurodevelopment); growth-plate chondrocytes and hepatic/renal tissue producing IGFBP-3 (growth axis, mouse-demonstrated); hematopoietic stem/progenitor cells (mouse MDS-like phenotype); germ cells (mouse fertility phenotypes; human oligospermia).
• Subcellular level: Nucleus — chromatin/histone H3 tail is the direct site of KDM3B enzymatic action (GO Cellular Component: nucleus/nucleoplasm/chromatin).
• Localization/laterality: No lateralization pattern reported; facial features are symmetric/bilateral dysmorphism.

8. Temporal Development

• Onset: Congenital/early infancy for feeding difficulties and dysmorphism; developmental delay typically recognized in infancy/early childhood; short stature may be present from birth (low birth weight) or emerge postnatally.
• Progression: Generally a stable, non-progressive developmental/malformation syndrome rather than a degenerative one — no reports of neurological regression were identified. Growth restriction can be progressive relative to peers if untreated, but is described as responsive to rhGH in the one treated case (PMC13025784).
• Disease course: Chronic, lifelong (affected parents in the founding cohort demonstrate the phenotype persists into adulthood, ages up to 42 years reported) (PMID:30929739).
• Critical periods: Neonatal period is a recognized vulnerability window (feeding difficulties); the GH-deficiency subgroup represents a window for endocrine intervention (rhGH therapy) that produced height improvement in one documented case.

9. Inheritance and Population

• Inheritance pattern: Autosomal dominant (AD), per OMIM #618846, ClinGen (Definitive, AD), and GenCC (Definitive, AD).
• De novo vs. inherited: In the founding cohort of 17 individuals: 10 confirmed de novo variants and 3 parent-to-child transmissions (affected parent, affected child) — p.Glu93* (mother→child), p.Asp336Gly (father→child), p.Glu1731Lys (father→child) (PMID:30929739). Subsequent reports continue this mixed pattern — e.g., the Turkish case's variant was inherited from a mother with "short stature without neurocognitive impairment" (PMC13008413).
• Penetrance: Appears complete but with markedly variable expressivity — "No evidence of skipped generations" was noted in the founding families, but severity ranges widely, including a mother who is a carrier with an essentially isolated short-stature phenotype and no cognitive impairment (PMC13008413), and a 4-family-member case series explicitly reporting a "variable neuropsychological profile" among relatives sharing one variant (PMID:42186234, abstract not independently verified in this session — treat as a lead).
• Expressivity: Variable, as above — ranges from growth-restriction-predominant presentations with minimal neurodevelopmental involvement (PMC13025784) to classic ID + short stature + dysmorphism + multisystem involvement.
• Genetic anticipation: Not reported/not applicable (not a repeat-expansion disorder).
• Germline mosaicism: Not specifically documented in the literature reviewed, though biologically plausible for any AD de novo-predominant disorder; not confirmed here.
• Founder effects: None reported; cases span multiple ethnicities/geographies (Netherlands/international GeneMatcher cohort, China ×3 reports, Albania, Turkey, Sudan).
• Consanguinity: Not specifically implicated (AD dominant mechanism, not recessive).
• Carrier frequency: Not applicable/not meaningful for an AD ultra-rare disorder with essentially no population carrier-frequency data (all reported variants absent from gnomAD).
• Epidemiology: Prevalence reported as <1/1,000,000 (extremely rare); total published, molecularly confirmed cases are estimated at roughly 25–30+ individuals across all literature identified in this search (17 in the founding 2019 cohort + at least 4 (growth-predominant series) + 2 (Neurogenetics 2023) + 1 (Albania, 2022) + 1 (Turkey, 2026) + 2 (China, 2025 autism report) + 1 (China, 2026 case report) + 1 (chILD/PAH, confounded case) — note some of these patient counts may overlap between the "27 pooled" figure in PMC13025784 and reports published after that review, so this total should be treated as an estimate to be refined during formal curation, not an exact count).
• Geographic distribution: Cases reported from the Netherlands/international collaborative cohort, China (×3 independent reports), Albania, Turkey, and Sudan — consistent with a pan-ethnic, non-geographically-restricted disorder ascertained wherever exome/genome sequencing access exists.
• Sex ratio: Founding cohort was roughly balanced (9 males, 8 females) (PMID:30929739); no strong sex skew reported.
• Age distribution: Founding cohort spanned 9 months to 42 years, reflecting both pediatric probands and their affected adult parents (PMID:30929739).

10. Diagnostics

• Genetic testing approach: Diagnosis is made by identifying a heterozygous pathogenic/likely pathogenic KDM3B variant, typically via exome sequencing (trio-based preferred) or a multigene neurodevelopmental-disorder/intellectual-disability panel that includes KDM3B; whole-genome sequencing would also detect coding variants. No single-gene-first testing strategy is standard given the absence of a specific pathognomonic clinical sign; NDD gene panels and exome sequencing are the practical entry points (consistent with how essentially all reported cases, including the founding cohort, were ascertained — via trio-ES with a chromatinopathy/NDD gene list).
• Chromosomal microarray / karyotype: Not the primary diagnostic modality (no described recurrent CNV mechanism), though often performed as part of standard NDD/short-stature diagnostic workup before/alongside sequencing.
• Clinical/biochemical workup: Growth-hormone stimulation testing is clinically relevant given the documented (though not universal) GH-deficiency subgroup (PMC13025784; PMC13008413). Standard developmental/cognitive assessment (e.g., Denver II, Wechsler scales) is used to characterize the ID/DD phenotype (as in PMC13008413 and PMID:41914422).
• Imaging: No syndrome-specific neuroimaging pattern has been established; HRCT was used in the chILD/PAH case to characterize ground-glass opacities, mosaic attenuation, and fibrosis, but this is not a diagnostic feature of DIJOS itself (confounded by co-occurring SIN3A variant) (PMID:38314229).
• Facial-gestalt/computational phenotyping: The founding study's CFPS/OpenFace-based facial-recognition analysis (statistically significant clustering, CIF = 2.5962, p = 0.0054) suggests that Face2Gene-style deep-phenotyping tools may aid diagnostic suspicion/prioritization pre- or post-sequencing (PMID:30929739), though this is a research tool rather than a validated clinical diagnostic criterion.
• Differential diagnosis: No formal differential-diagnosis list specific to DIJOS was located in the primary literature searched. Given the overlapping presentation (ID + short stature + facial dysmorphism + feeding difficulties + hypermobility), DIJOS falls within the broader chromatinopathy differential (Cornelia de Lange syndrome and related cohesinopathies/chromatin-regulator disorders, Coffin-Siris syndrome, KBG syndrome) on clinical grounds by phenotypic analogy, but no primary source explicitly names these as the DIJOS differential — this should be flagged as an inference rather than sourced fact if curated, and verified against the original AJHG paper's discussion section directly (not fully retrievable via WebFetch in this session due to access restrictions on cell.com/ScienceDirect).
• Standardized diagnostic criteria: No DSM/ICD-specific diagnostic criteria exist for DIJOS as a discrete entity; diagnosis is molecular (variant identification) combined with compatible phenotype.

11. Outcome / Prognosis

• Survival/mortality: No mortality data specific to DIJOS were identified; the disorder is not described as life-limiting in the reviewed literature, though the malignancy signal (AML, Hodgkin lymphoma in 2/17 founding-cohort patients) and the single severe PAH/chILD case represent potentially serious, though uncommon, complications.
• Morbidity/function: Long-term functional outcome is dominated by the degree of intellectual disability (mild-to-moderate in the great majority) and short stature; affected adults in the founding cohort (up to age 42) demonstrate the condition is compatible with adult survival and, in at least some carriers (e.g., the mildly affected mothers in inherited cases), with minimal functional impairment.
• Complications: Malignancy (AML, Hodgkin lymphoma — small numbers, ascertainment-biased), cardiomyopathy, neutropenia, pulmonary arterial hypertension (single confounded case), epilepsy (transient in some), hearing loss, ophthalmologic complications.
• Recovery/treatment response: Growth response to rhGH therapy was substantial in one documented case (94.3 cm → 135 cm over 6 years, moving from below -3 SD to -1 to -2 SD) (PMC13025784) — this is the only quantified treatment-outcome data point identified.
• Prognostic factors: Preliminary/unconfirmed genotype-phenotype signal that truncating (LoF) variants may associate with a milder neurodevelopmental phenotype than missense variants (PMC13025784) — treat as an emerging hypothesis, not an established prognostic rule.

12. Treatment

Management is supportive and multidisciplinary; there is no disease-modifying or gene-targeted therapy for DIJOS.

• Pharmacotherapy:
• Recombinant human growth hormone (rhGH) for biochemically confirmed GH deficiency — documented efficacy in one case (PMC13025784); the authors of that series explicitly flag a theoretical concern requiring individualized risk-benefit discussion: "given that KDM3B...has been linked to myelodysplastic syndrome and acute myeloid leukemia, the potential oncogenic risk of rhGH must be carefully weighed" — i.e., rhGH use in DIJOS should be considered in the context of the gene's hematologic-malignancy association, though no formal contraindication or surveillance protocol has been published.
  • NCIT:C1691 (Growth Hormone, generic) — treatment_term candidate; verify exact NCIT binding before curation.
• ADHD pharmacotherapy as clinically indicated for the behavioral/ADHD subgroup (general guideline-level statement, not DIJOS-specific evidence).
• Rehabilitative/supportive care: Early referral to developmental/rehabilitation services (physical therapy, occupational therapy, speech-language therapy) for the ID/DD and motor-delay phenotype; individualized educational programs.
• NCIT:C15302 (Physical Therapy); NCIT:C159273 (Speech Therapy, verify); NCIT:C121351 (Occupational Therapy, verify).
• Genetic counseling: Recommended given AD inheritance, de novo predominance but documented parent-to-child transmission with variable expressivity — family variant testing (cascade testing of parents) is standard once a proband variant is identified, as demonstrated in essentially every cited case report.
• NCIT:C15240 (Genetic Counseling).
• Malignancy vigilance: No formal, published surveillance protocol exists; the founding-cohort authors recommend clinical vigilance rather than routine screening, citing ascertainment bias in their 2/17 malignancy signal (PMID:30929739).
• Experimental/investigational therapies: None identified — no gene therapy, targeted molecular therapy, or clinical trials specific to DIJOS were found in this search (searched ClinicalTrials.gov indirectly via general web search; no DIJOS/KDM3B-specific trial surfaced).
• Cardiac/pulmonary management: Standard cardiology/pulmonology management as clinically indicated for the individually reported cardiomyopathy and PAH/chILD cases; no DIJOS-specific cardiac or pulmonary management guideline exists.

13. Prevention

• Primary prevention: Not applicable — this is a predominantly de novo genetic disorder; no modifiable primary-prevention strategy exists.
• Secondary prevention / screening: No population or newborn screening program exists (ultra-rare, molecularly diagnosed condition). Prenatal/preimplantation genetic testing is relevant once a familial pathogenic variant is identified (standard genetic-counseling practice for known AD variants), though this was not explicitly documented as performed in any cited case.
• Genetic counseling: Central preventive/family-planning tool once a proband's variant is identified — enables informed reproductive decision-making and identification of mildly affected/subclinically affected parents (as seen in the inherited cases above, where a parent's own diagnosis was made retrospectively via cascade testing).
• Behavioral/public health interventions: Not applicable.

14. Other Species / Natural Disease

• Taxonomy: No naturally occurring veterinary/companion-animal disease attributable to spontaneous Kdm3b variants was identified in this search (searched OMIA indirectly via general search; no hits surfaced). This differs from the engineered mouse models below.
• Orthologous gene: Mouse Kdm3b (MGI, chromosome 18) is the principal ortholog studied.
• Comparative biology: KDM3B is conserved with paralogs KDM3A and JMJD1C within the JmjC-domain histone demethylase family (PMID:30929739 notes cross-paralog conservation at missense-affected residues).

15. Model Organisms

• Species/model type: Mouse (Mus musculus), germline Kdm3b knockout — the principal and, in the literature reviewed, essentially the only animal model directly informing DIJOS mechanism.
• Key model studies:
• Growth/endocrine: Kdm3b-knockout mice show postnatal somatic growth restriction; mechanistically, Kdm3b ablation reduces renal/hepatic IGFBP-3 expression (~53% reduction reported), lowering circulating IGF-1 (~36% reduction), driving growth restriction (PMID:25892958, "The histone H3K9 demethylase Kdm3b is required for somatic growth and female reproductive function").
• Female reproduction: The same knockout model shows irregular estrous cycles, ~45% reduced ovulation capacity, ~47% reduced fertilization rate, ~44% reduced uterine decidual response, and >50% reduced circulating 17β-estradiol (PMID:25892958).
• Male reproduction: A separate Kdm3b-knockout study shows decreased sperm number/motility and impaired consummatory and motivational sexual behavior (PMID:26681924, "Knockout of the Histone Demethylase Kdm3b Decreases Spermatogenesis and Impairs Male Sexual Behaviors").
• Hematopoiesis: Kdm3b-knockout mice show hematopoietic defects described by the DIJOS founding-cohort authors as "consistent with myelodysplastic syndrome" (cited secondhand in PMID:30929739; primary knockout-hematopoiesis paper not independently retrieved in this session and should be verified directly before citing in curation).
• Neurocognitive (cerebellum): Mouse studies indicate diminished H3K9 demethylase activity "impairs cerebellar memory consolidation" and that Kdm3b "plays an essential role in motor learning" (cited in PMC13025784; primary source not independently retrieved in this session).
• Phenotype recapitulation: The mouse model recapitulates the growth-restriction and (male/female) reproductive phenotypes observed in a subset of human patients reasonably well (the founding-cohort authors explicitly draw this parallel, including citing one patient's oligospermia as "in accordance" with the mouse phenotype). The neurodevelopmental/cognitive phenotype in humans (the dominant clinical feature) is far less directly validated in the mouse model — the cerebellar/motor-learning mouse data are suggestive but have not been shown to reproduce the human ID/developmental-delay phenotype specifically, representing a genuine human-model translational gap.
• Model limitations: Complete germline knockout (not a patient-variant-specific knock-in) — does not model the dominant, haploinsufficient, single-copy-loss human genetics precisely (most mouse studies use homozygous knockouts), and does not model the missense/potentially dominant-negative variant subclass at all. No conditional, humanized, or patient-variant-specific (knock-in) mouse model was identified in this search.
• Other model systems: No zebrafish, Drosophila, C. elegans, iPSC-derived, or organoid models of KDM3B/DIJOS were identified in this search.
• Resources: MGI (Mouse Genome Informatics) is the relevant repository for the Kdm3b knockout allele(s); specific allele IDs were not retrieved in this session and should be looked up directly (e.g., via MGI) before curation.

Summary of Sourcing Caveats for Curation

1. Full text of the founding paper (PMID:30929739, AJHG 2019) was accessed via a PMC mirror (PMC6451728) through an AI-summarization WebFetch pass; the quotes above are reported as extracted but should be re-verified against the primary PMC/journal HTML directly (exact-substring snippet verification, per dismech's evidence workflow) before being committed as snippet: values in KB YAML — this report is a research lead, not pre-verified curation content.
2. Several PubMed abstract pages (PMID:36757469, PMID:42186234) could not be fetched directly (cookie-consent wall); abstract content for these was reconstructed from WebSearch snippets/secondary summaries only — treat as leads requiring direct primary-source verification, not as verified quotes.
3. The AI-generated acronym expansion of "DIJOS" encountered in one search result (implying "immunodeficiency" and "joint contractures" are part of the name) could not be traced to OMIM, Orphanet, or the founding paper and appears to be a fabrication/hallucination by a web-summarization layer — do not use it.
4. The chILD/PAH case (PMID:38314229) carries a co-occurring SIN3A variant; causal attribution to KDM3B alone is confounded and should be curated as such (e.g., lower-confidence evidence, explicit dual-diagnosis note).
5. All GO/CL/UBERON/NCIT/HP CURIEs suggested above are research leads only and were not independently verified via OAK/runoak lookup in this session, per the "every CURIE is read from a source in the same step it is written" rule — they must be looked up fresh at curation time, not copied from this report.
6. Total published-case counts are estimated by summing across overlapping reviews (PMC13025784's "27 patients" pooled figure predates several 2025–2026 case reports) and should be reconciled carefully during curation rather than taken as an exact denominator.

Reference Validation

Checked with linkml-reference-validator 0.2.1.

Quotes not found in the cited source

Searched the abstract, any retrieved full text, and the title. A quote drawn from a part of the paper that was not retrieved will appear here too, so check before treating one as invented:

• PMID:30929739: "13/14 had high CADD scores (13–40)"
• closest text in source: "All 14 different variants found were absent from the gnomAD and ExAC databases, and 13/14 had high Combined Annotation Dependent Depletion scores (CADD; scores between 13 and 40) and were predicted to be pathogenic by both the PolyPhen-2 and SIFT prediction programs (Table 1).12, 13, 14, 15 All individuals or their legal guardians gave written informed consent, and the study was given IRB approval"

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:0030012 (3 mentions) - the report calls it "if available", "confirmed via GenCC submission, consistent with the MONDO ID supplied in the research request"; MONDO calls it Diets-Jongmans syndrome
• HP:0011342 (1 mention) - the report calls it "Mild intellectual disability"; HP calls it Mild global developmental delay
• HP:0000824 (1 mention) - the report calls it "Growth hormone deficiency"; HP calls it Decreased response to growth hormone stimulation test
• HP:0009884 (1 mention) - the report calls it "Long ear"; HP calls it Tapered distal phalanges of finger
• HP:0011832 (1 mention) - the report calls it "Low columella"; HP calls it Narrow nasal tip
• HP:0001696 (1 mention) - the report calls it "Heterotaxy"; HP calls it Situs inversus totalis
• HP:0012156 (1 mention) - the report calls it "Hodgkin lymphoma, if using more specific term"; HP calls it Hemophagocytosis
• NCIT:C1691 (1 mention) - the report calls it "Growth Hormone, generic"; NCIT calls it Trabectedin

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:

• MONDO:0000508 (1 mention) - the report calls it "Syndromic intellectual disability", "GeneCC/ClinGen: Definitive gene-disease relationship, "Syndromic intellectual disability"; MONDO calls it syndromic intellectual disability**
• HP:0000307 (1 mention) - the report calls it "Prominent chin"; HP calls it Pointed chin, and lists "Pointy chin" among its other names
• HP:0001875 (1 mention) - the report calls it "Neutropenia"; HP calls it Decreased total neutrophil count, and lists "Neutropenia" among its other names
• HP:0002092 (1 mention) - the report calls it "Pulmonary hypertension"; HP calls it Pulmonary arterial hypertension, and lists "Pulmonary artery hypertension" among its other names
• GO:0032454 (1 mention) - the report calls it "histone H3-K9 demethylase activity, note-check exact ID against GO"; GO calls it histone H3K9 demethylase activity, and lists "histone H3-K49 demethylase activity" among its other names
• GO:0006338 (1 mention) - the report calls it "chromatin remodeling, verify"; GO calls it chromatin remodeling
• NCIT:C159273 (1 mention) - the report calls it "Speech Therapy, verify"; NCIT calls it Speech Language Therapy
• NCIT:C121351 (1 mention) - the report calls it "Occupational Therapy, verify"; NCIT calls it Occupational Therapy

Terms named inconsistently

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

• MONDO:0030012 - called "if available", "confirmed via GenCC submission, consistent with the MONDO ID supplied in the research request"
• MONDO:0000508 - called "Syndromic intellectual disability", "GeneCC/ClinGen:** Definitive gene-disease relationship, "Syndromic intellectual disability"

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.