SETD5 Haploinsufficiency Syndrome — Comprehensive Research Report

Prepared: 2026-07-31 · Target entity: Intellectual disability–facial dysmorphism syndrome due to SETD5 haploinsufficiency (MRD23 / IDD23 / "SETD5 Disorder") · Intended use: dismech knowledge-base entry population

⚠️ Verification preamble (read before curating)

Three provenance caveats apply to everything below:

1. Quote status. Quotes marked [CACHE-VERIFIED] were read character-for-character from references_cache/PMID_*.md files already present in this worktree and are safe to use as evidence snippet: values. Quotes marked [UNVERIFIED-QUOTE] were extracted by a summarizing web-fetch layer; they are probably exact but must be re-checked with just fetch-reference PMID:XXXX + just validate-references before being committed as snippets.
2. Ontology IDs. Every HP/GO/CL/UBERON/CHEBI/NCIT term below is a suggestion. Two GO terms curators commonly reach for here are obsolete and must not be used: GO:0010452 ("obsolete histone H3-K36 methylation") and GO:0034968 ("obsolete histone lysine methylation") — both were deprecated as mis-namespaced BP terms. Confirmed-live alternatives are given in §6. Run just validate-terms on everything.
3. Named Entity Confusion (NEC) risk — HIGH for this disease. SETD5 sits in a naming minefield. Specifically:
4. ORPHA:435638 "Proximal 3p25.3 microdeletion syndrome" (MONDO:0018564) is a different disease — an ID/epilepsy/stereotypic-hand-movement entity from a more proximal 3p25.3 interval, not the SETD5 critical region. Do not merge.
5. OMIM:615743 is the gene SETD5; OMIM:615761 is the phenotype MRD23. Easy to swap.
6. KBG syndrome (ANKRD11) and Cornelia de Lange syndrome (NIPBL) are genuine clinical mimics with real published SETD5 cases; literature retrieved under "KBG" or "CdLS" queries may or may not be about SETD5.
7. Much of the PubMed hit list for "SETD5" is oncology (PDAC, NSCLC, breast, glioma, colorectal), where SETD5 is overexpressed/hyperactive — the mechanistic opposite of the germline haploinsufficiency disorder. Do not cross-import mechanism.

1. Disease Information

Overview

SETD5 haploinsufficiency syndrome is an autosomal dominant, essentially always de novo, neurodevelopmental chromatinopathy caused by heterozygous loss-of-function of SETD5 at 3p25.3. It is defined by global developmental delay/intellectual disability, prominent speech and language impairment, hypotonia, feeding difficulties in infancy, a recognizable but non-specific facial gestalt (brachycephaly, high forehead, synophrys/full broad eyebrows, long tubular nose, upslanting palpebral fissures, low-set fleshy ears, thin upper lip, low anterior hairline), and a high burden of behavioral/psychiatric comorbidity (autism, ADHD, obsessive-compulsive features, hand flapping, stereotypies). Variable additional features include congenital heart defects, skeletal/limb anomalies (notably leg-length discrepancy), short stature, epilepsy, and a growing list of newly reported systemic associations.

It is the principal single-gene explanation for the classic 3p25 (3p–) microdeletion syndrome phenotype: SETD5 lies within the 124 kb critical interval and deletion versus intragenic LoF produce a largely overlapping presentation.

"SETD5 lies within the critical interval for 3p25 microdeletion syndrome. The individuals with SETD5 mutations showed phenotypic similarity to those previously reported with a deletion in 3p25, and thus loss of SETD5 might be sufficient to account for many of the clinical features observed in this condition." — Grozeva et al. 2014, Am J Hum Genet (PMID:24680889) [UNVERIFIED-QUOTE]

Key identifiers

Synonyms

• Intellectual developmental disorder, autosomal dominant 23 (IDD23)
• Mental retardation, autosomal dominant 23 (MRD23) — historic/discouraged
• Intellectual disability–facial dysmorphism syndrome due to SETD5 haploinsufficiency
• SETD5-related neurodevelopmental disorder / SETD5-related disorder / SETD5 Disorder (the term used by the National Brain Gene Registry)
• 3p25.3 microdeletion syndrome (SETD5-containing form) / 3p– syndrome critical-region phenotype

Data provenance type — mixed

Both individual-patient and aggregated. Notably: - EHR/billing-code-derived: the National Brain Gene Registry cohort explicitly validated 10 clinical features against ICD-10 billing codes vs. manual chart review (PMID:40265665) — a rare instance of EHR-derived phenotyping in this disorder. - Patient-reported/registry: a 2026 Facebook support-group survey (n=51, 12 countries) (PMID:42468298). - Clinician-ascertained case series: the 28-patient European multicenter cohort (PMID:39603091), Powis et al. diagnostic-exome cohort (PMID:28881385). - Aggregated resources: HPO/OMIM annotations (currently derived from only ~7–9 individuals — see §3), ClinGen, ClinVar, DECIPHER, SFARI Gene.

2. Etiology

Disease causal factors

Purely genetic and monogenic: heterozygous loss of function of SETD5, arising as either (a) an intragenic sequence variant (nonsense, frameshift, canonical splice) or (b) a copy-number deletion encompassing the gene (interstitial 3p25.3 microdeletion or larger 3p terminal deletion). Haploinsufficiency is the established mechanism — nonsense-mediated decay of the mutant transcript has been directly demonstrated:

"CRISPR/Cas9 mutation modelling of the two intragenic variants demonstrated nonsense-mediated decay of the resulting transcripts, pointing to a loss-of-function (LoF) and haploinsufficiency as the common disease-causing mechanism of intragenic SETD5 sequence variants and SETD5-containing microdeletions." — Kuechler et al. 2015, Eur J Hum Genet (PMID:25138099) [UNVERIFIED-QUOTE]

ClinGen dosage sensitivity: Haploinsufficiency score 3 — "Sufficient Evidence for Haploinsufficiency" (curated 2014-11-06); Triplosensitivity score 0 — No Evidence (source: ClinGen Dosage Map, HGNC:25566). ClinGen Gene-Disease Validity: DEFINITIVE, autosomal dominant, Intellectual Disability and Autism GCEP, evaluated 2023-07-27.

Genetic risk factors

• Causal variant class: LoF only. No convincing missense/GoF disease mechanism has been established.
• Constraint: SETD5 is among the most LoF-intolerant genes in the genome. ExAC pLI = 0.9999996 (rank 259 / 18,225 genes); Sanders TADA score 0.0025 (rank 22 / 18,665); EAGLE score 28.05 ("Strong"); SFARI Gene score 1S (High Confidence, Syndromic), 181 rare variants catalogued, 0 common variants (source: SFARI Gene, gene.sfari.org/database/human-gene/SETD5).
• Genomics England PanelApp: GREEN (diagnostic-grade) on Intellectual disability, DDG2P, Fetal anomalies, Skeletal dysplasia, and Early onset or syndromic epilepsy; AMBER on Cerebral vascular malformations and Monogenic short stature. All monoallelic.
• Genetic background / "second hits": rare variants in the genetic background modulate expressivity in this class of disorder (Pizzo et al. 2019, Genet Med 21:816–825). A functional two-hit study showed SETD5 homologs synergistically interact with MOSMO homologs in Drosophila and X. laevis, producing axon-outgrowth defects absent with single knockdown (PMID:33819264) [UNVERIFIED-QUOTE].

Environmental risk factors

None established. Paternal age effects for de novo variants are plausible in principle (as for all de novo dominant disorders) but have not been demonstrated specifically for SETD5. No toxin, infectious, dietary, or occupational exposure has been implicated. Not applicable / no data.

Protective factors

No protective genetic or environmental factors identified. No data.

However, the existence of transmitting parents with mild or no phenotype (see §9 Penetrance) implies unmeasured genetic or stochastic buffering. Powis et al. describe "an apparently unaffected carrier mother of an affected individual and a carrier mother with normal intelligence and affected twin sons" [CACHE-VERIFIED, PMID:28881385]. Mechanistic basis unknown — this is a genuine knowledge gap.

Gene–environment interactions

No documented GxE for SETD5. One review positions SETD5 among NSPC-proliferation regulators that converge with teratogenic insults (Zika, valproate, metabolic stress) on shared cell-cycle control (PMID:41283518) — a hypothesis-generating convergence claim, not a demonstrated interaction. Curate as such if at all.

3. Phenotypes

3.1 Authoritative HPO annotation set (OMIM:615761, via ontology.jax.org)

⚠️ Critical caveat for KB frequency curation: the current HPO annotation frequencies for this disease derive from a very small seed cohort (denominators of 7, 9, or 2). Do not convert these fractions to HPO FrequencyEnum bands without acknowledging the tiny n. Prefer the larger cohorts in §3.2 for frequency claims.

Inheritance: HP:0000006 (Autosomal dominant inheritance)

3.2 Frequencies from larger, better-powered cohorts (preferred for curation)

European multicenter neurological/psychiatric cohort, n = 28 (De Falco et al. 2025, PMID:39603091) — 26 SNV, 2 CNV:

"In our cohort neurological symptoms include hypotonia (39.2 %), hyperkinetic movement disorders including stereotypies and chorea (21.4 %) and gait abnormalities ranging from tip-toe or unsteady walking and alterations of fine motor skills (35.7 %). Epilepsy was present in about 14 % of patients, including different types of seizures as epileptic spasms, focal motor, and non-motor seizures. Concerning the cognitive phenotype, intellectual disability or global developmental delay depending on age, ranging from mild to severe, was present in 75 % of cohort, 21.4 % exhibit borderline intellectual functioning while an individual has a normal intelligence quotient. Other psychiatric comorbidities include autism, ADHD, psychotic disorder and other internalizing and externalizing symptoms." [CACHE-VERIFIED]

Facebook support-group survey, n = 51 respondents from 12 countries (Talaba et al. 2026, Pediatr Neurol, PMID:42468298) — patient/caregiver-reported; 80% of affected individuals <18 years; mean age at diagnosis 9.2 years:

"Common features included developmental delay (96%), hypotonia (78%), intellectual disability (75%), gait abnormality (59%), vision problems (51%), constipation (47%), and anxiety (47%). Potential novel findings included high pain tolerance (43%), persistent leg pain (31%), and joint pain (27%)." [CACHE-VERIFIED]

Note the striking discordance between HPO's hypotonia annotation (absent) and both cohorts (39–78%) — hypotonia is under-annotated in HPO for this disease and should be curated as FREQUENT.

National Brain Gene Registry, n = 13, ages 2–37 y (Callahan et al. 2025, PMID:40265665):

"Participants in our cohort had features not previously reported, including brain and musculoskeletal abnormalities. One participant had cerebral palsy." [CACHE-VERIFIED]

3.3 Phenotypes by category

Behavioral / psychiatric (search: HPO, DSM-5) - Autistic behavior / ASD (HP:0000729) — the most consistently reported comorbidity; SFARI 1S gene - Obsessive-compulsive behavior with hand flapping and ritualized behavior (HP:0000722, HP:0100023 stereotypical hand wringing — verify) — Grozeva et al. 2014 called these "prominent features" - ADHD (HP:0007018) - Anxiety (HP:0000739) — 47% - Psychotic disorder (HP:0000709) — reported in the De Falco cohort; notable and under-recognized - Internalizing/externalizing symptoms - SETD5 also emerged in a whole-genome de-novo-mutation study of obsessive-compulsive disorder (Lin et al. 2022, Sci Adv 8:eabi6180)

Neurological - Hypotonia; motor delay; delayed ability to walk (HP:0031936) - Gait abnormality including toe-walking (HP:0040083 — verify) - Hyperkinetic movement disorder: stereotypies, chorea - Epilepsy (~14%): epileptic spasms, focal motor and focal non-motor seizures. A dedicated case report documents evolution to focal and generalized seizures in a 6-year-old (PMID:40462669) — "epilepsy may arise after SETD5 variants, with subtle clinical manifestations" [UNVERIFIED-QUOTE] - Severe cerebral cortical dysplasia (HP:0002539) in one nonsense case with congenital diaphragmatic hernia (PMID:28263952) - Cerebral palsy in 1/13 BGR participants - Moyamoya angiopathy (HP:0011834 — verify) — a de novo SETD5 variant identified among 39 MMA trios (PMID:31474762); adult-onset cerebrovascular pleiotropy. Interpretive caution: the paper's replication support was for CHD4/CNOT3, not SETD5 — curate as PARTIAL/emerging, not established. - Drooling (HP:0002307)

Speech / language - Delayed speech and language development (HP:0000750) — 8/9; among the most consistent features - Receptive–expressive language disorder, speech disorder

Craniofacial (the gestalt) — from Grozeva 2014, verbatim description: brachycephaly; prominent high forehead with synophrys or "striking full and broad eyebrows"; long, thin, tubular nose; long, narrow upslanting palpebral fissures; large, fleshy low-set ears. Plus: low anterior hairline, thin upper lip vermilion, long/smooth philtrum, downturned corners of mouth, micrognathia, depressed/wide nasal bridge, bulbous nose, anteverted nares, dental crowding.

Skeletal / musculoskeletal - Leg-length discrepancy / lower limb asymmetry (HP:0100559) — distinctive and repeatedly reported ("significant leg-length discrepancy... a frequent finding") - Scoliosis (HP:0002650), kyphosis (HP:0002808), hyperlordosis (HP:0003307) - Variable hand and skeletal abnormalities (Powis 2018) - Broad distal phalanges, sandal gap, postaxial polydactyly (rare) - Bone fragility — novel association (Anderson et al. 2021, Clin Genet 100:352–354, PMID:34169511); HP:0002659 (Increased susceptibility to fractures — verify) - Vertebral fusion in the Setd5 het mouse (IMPC) — worth a targeted look in humans

Growth - Growth retardation / short stature (HP:0004322); a severe case at −5.22 SDS height with delayed bone age (PMID:40869907) - PanelApp lists SETD5 (Amber) on the Monogenic short stature panel - Feeding difficulties in infancy (HP:0011968) — 5/7

Cardiac - Congenital heart defects: ASD, VSD (including ostium primum ASD detected prenatally, PMID:41368699), conotruncal/outflow-tract defects. Mouse Setd5 haploinsufficiency produces double outlet right ventricle and perimembranous VSD (PMID:34050709), giving strong mechanistic corroboration.

Ophthalmologic - Myopia, astigmatism, strabismus, ptosis; "vision problems" 51% in survey - Ptosis was the presenting feature in a 10.1-Mb 3p25 terminal deletion (PMID:28951171)

Gastrointestinal / genitourinary / other - Constipation (47%) - Inguinal hernia; hypospadias (HP:0000047) - Congenital diaphragmatic hernia (HP:0000776) — rare, one case - CAKUT — machine-learning analysis of 515 clinical-exome cases "implicated ADNP and SETD5 genes as associated with increased CAKUT risk" (PMID:40913078) [UNVERIFIED-QUOTE]; emerging, low confidence - Congenital hypopituitarism — variants in SETD5 found in a CH cohort after mouse pituitary-malformation screening (PMID:38822427); emerging - Aberrant blind-ending bronchus (single case, PMID:28905509) - Neuroblastoma — one report expanding SETD5 haploinsufficiency into neuroblastoma (PMID:32748512). Curate cautiously: no cohort-level cancer-risk estimate exists; there is currently no evidence base for tumor surveillance.

3.4 Phenotype characteristics

• Onset: congenital to infantile (HP:0003593 Infantile onset annotated 2/2). Facial dysmorphism and CHD are congenital; hypotonia/feeding difficulty neonatal–infantile; DD apparent in the first 1–2 years; behavioral/psychiatric features school-age; epilepsy variable.
• Severity: highly variable — from normal IQ (1 individual in the 28-patient cohort) and borderline functioning (21.4%) through mild, moderate, to severe ID. Original series emphasized "moderate to severe ID."
• Progression: the core cognitive phenotype is static/non-progressive (a developmental, not neurodegenerative, disorder). Musculoskeletal features (scoliosis, leg-length discrepancy) and epilepsy may be progressive/emergent. Late-onset cerebrovascular (moyamoya) pleiotropy has been proposed, prompting the recommendation to "assess clinical complications into adulthood" (PMID:31474762).
• Quality-of-life impact: no EQ-5D/SF-36/PROMIS study exists for this disorder — data gap. The Facebook survey is the closest proxy: it documents caregiver-reported burden domains (feeding, gait, vision, constipation, anxiety, pain) and found that group participation "empowered families through shared experiences and information exchange" [CACHE-VERIFIED, PMID:42468298]. The mean 9.2-year diagnostic delay is itself a QoL-relevant finding.

4. Genetic / Molecular Information

Causal gene

Source for structure/expression: Li et al. 2023 review (PMID:36875494), read from cached full text: "The human SETD5 gene (OMIM 615743), also known as MRD23, SETD5A, 2900045N06Rik or mKIAA1757, is located on the chromosome 3p25.3 and encodes the SETD5 protein composed of 1442 amino acids... The SETD5 gene consists of 31 exons and is ubiquitously expressed in human tissues such as the brain, thyroid, skin, ovary, lung and endometrium." [CACHE-VERIFIED]

Pathogenic variants

Variant classes observed (all LoF): nonsense > frameshift > canonical splice-site > whole-gene/partial-gene deletion. No pathogenic missense mechanism established.

Landmark allelic series — Grozeva et al. 2014 (PMID:24680889), 7 de novo LoF variants:

Additional published alleles: - c.3848_3849insC (Chr3:9,517,294 A>AC) — mild ID in a 36-year-old male (PMID:28549204) - c.890_891delTT — severe short stature + overlap syndrome, rhGH-treated (PMID:40869907) - NM_001080517.3:c.3601_3605del, p.Trp1201GlufsTer2 — de novo, prenatally ascertained via ostium primum ASD (PMID:41368699) - Frameshift in exon 12 — ID + aberrant blind-ending bronchus (PMID:28905509) - 81 bp deletion spanning a splice-donor site + a nonsense variant, both NMD-confirmed (PMID:25138099) - SETD5^S1257 — a functional truncation used experimentally: it abolishes HDAC3/PAF1 interaction and separates SETD5's proliferation function from its anti-apoptotic function (PMID:36349512*)

Variant classification distribution (National Brain Gene Registry, PMID:40265665): of 11 unique P/LP variants in 13 individuals from 11 families — 6 nonsense, 4 frameshift, 1 splice site [CACHE-VERIFIED].

Allele frequency: pathogenic SETD5 LoF alleles are absent/singleton in population databases; the gene is extremely LoF-depleted (ExAC pLI 0.9999996). No recurrent/founder allele has been described.

Germline vs somatic: disease alleles are germline, overwhelmingly de novo. SETD5 is separately somatically altered/overexpressed in multiple cancers (PDAC, NSCLC, breast, ESCC, bladder amplification ~10%, high-grade glioma, ALL, prostate, colorectal, neuroblastoma) — do not conflate with the germline haploinsufficiency disorder.

Functional consequence: loss of function via NMD → 50% dosage → haploinsufficiency. No dominant-negative or GoF mechanism demonstrated for germline disease.

Modifier genes

• No validated Mendelian modifier. Genetic-background rare-variant burden is the leading model (Pizzo 2019).
• MOSMO is a demonstrated genetic interactor in invertebrate/amphibian two-hit assays (PMID:33819264) — model-organism evidence only.
• ANKRD11 is an upstream regulator of SETD5 (see §6), which reframes the KBG/SETD5 overlap as a pathway relationship rather than coincidental phenocopy.

Epigenetic information

• A validated DNA-methylation episignature exists for MRD23/SETD5. SETD5 is in the EpiSign validated-disorder panel (EpiSign V2/V3 classifiers; ~25 reference samples reported for MRD23). Foundational methodology: Aref-Eshghi et al. 2020, Am J Hum Genet (PMID:32109418), "Evaluation of DNA Methylation Episignatures for Diagnosis and Phenotype Correlations in 42 Mendelian Neurodevelopmental Disorders" — "This study more than doubles the number of published syndromes with DNA methylation episignatures." [UNVERIFIED-QUOTE]; chromatinopathy application: Levy et al. 2022, Genet Med (PMID:34906459).
• A 2026 study of 400 NDD individuals confirmed episignature concordance for SETD5 among chromatinopathy genes and used EpiSign to reclassify VUS (PMID:41957673): "26 individuals (43%) exhibited disorder-specific episignatures consistent with the associated clinical diagnosis" [UNVERIFIED-QUOTE].
• The primary molecular lesion is itself epigenetic — see §6.

Chromosomal abnormalities

• 3p25.3 interstitial microdeletions encompassing SETD5: e.g., a 684 kb deletion refining a 124 kb critical region containing only THUMPD3, SETD5, and LOC440944 (Kellogg et al. 2013, PMID:23613140); a 116 kb deletion partially involving SETD5 in a KBG-suspected patient (PMID:32793091); four independent de novo non-recurrent microdeletions (PMID:25138099).
• 3p terminal deletions (3p– syndrome), e.g., a de novo 10.1 Mb 3p25 terminal deletion (PMID:28951171). Larger deletions add features not attributable to SETD5 alone (microcephaly, seizures, more severe cardiac disease) — the 684 kb case notably lacked cardiac defects, seizures, and microcephaly, supporting a contiguous-gene contribution in larger deletions.
• Detection: chromosomal microarray (aCGH/SNP array); DECIPHER and dbVar hold the CNV records.

5. Environmental Information

• Environmental factors: none identified. Not applicable.
• Lifestyle factors: none identified. Not applicable.
• Infectious agents: none. Not applicable.

The only environmental-adjacent literature is a general review noting that teratogens (Zika, valproate) can phenocopy NSPC-proliferation defects also produced by SETD5 loss (PMID:41283518) — a mechanistic convergence claim, not an etiologic factor for this disease.

6. Mechanism / Pathophysiology

6.1 The central unresolved question — is SETD5 a methyltransferase?

This is the single most important mechanistic controversy for this entry and should be curated explicitly as competing mechanistic_hypotheses rather than silently resolved.

Hypothesis A — SETD5 is a genuine H3K36 methyltransferase (canonical/"catalytic" model). Sessa et al. 2019, Neuron (PMID:31515109):

"Mutations in one SETD5 allele are genetic causes of intellectual disability and autistic spectrum disorders. However, the mechanisms by which SETD5 regulates brain development and function remain largely elusive. Herein, we found that Setd5 haploinsufficiency impairs the proliferative dynamics of neural progenitors and synaptic wiring of neurons, ultimately resulting in behavioral deficits in mice. Mechanistically, Setd5 inactivation in neural stem cells, zebrafish, and mice equally affects genome-wide levels of H3K36me3 on active gene bodies. Notably, we demonstrated that SETD5 directly deposits H3K36me3, which is essential to allow on-time RNA elongation dynamics. Hence, Setd5 gene loss leads to abnormal transcription, with impaired RNA maturation causing detrimental effects on gene integrity and splicing." [CACHE-VERIFIED]

Hypothesis B — SETD5 is catalytically dead and acts as a co-repressor scaffold ("scaffold" model). Wang et al. 2020, Cancer Cell (PMID:32442403):

"SETD5 lacks histone methyltransferase activity but scaffolds a co-repressor complex, including HDAC3 and G9a." [UNVERIFIED-QUOTE]

The 2023 review (PMID:36875494) states both positions and adds the enzymological detail [CACHE-VERIFIED from full text]:

"SETD5 contains a SET (Su(var)3-9, enhancer-of-zeste, trithorax) domain and is thus annotated as a candidate protein of lysine methyltransferase, which methylates H3K36 up to the tri-methyl form (H3K36me3)... However, there is evidence that SETD5 lacks the methyltransferase activity but scaffolds a co-repressor complex, including HDAC3, NCoR, G9a, and PAF1, which couples selective deacetylation of H3K9ac with methylation of this residue."

And on subfamily context: "The yeast SET3 and SET4, Drosophila UpSET, and human MLL5 are homologous to SETD5 over their SET domains and, except for SETD5, contain a PHD finger." [CACHE-VERIFIED]

Curation guidance: model as two hypothesis_group_ids (e.g. setd5_h3k36me3_catalytic [EMERGING/CONTESTED] and setd5_corepressor_scaffold [EMERGING/CONTESTED]), with the downstream node "impaired RNA Pol II elongation and transcriptional fidelity" as a convergent hub that both hypotheses feed. Add a discussions entry with kind: KNOWLEDGE_GAP attached to the catalysis node.

6.2 Substrate/complex map (from PMID:36875494 Table 1, CACHE-VERIFIED)

Additional partners: PAF1 complex (Ctr9), NCoR/HDAC3, G9a/EHMT2, HCF-1, TBL1XR1, BRD2, OGT (in cancer), RNA Pol II.

6.3 Causal chain — germline haploinsufficiency (proposed pathograph)

Trigger (MOLECULAR): Heterozygous SETD5 LoF variant / 3p25.3 deletion → NMD → ~50% SETD5 protein ↓ MOLECULAR: Loss of SETD5-dependent chromatin state — reduced genome-wide H3K36me3 on active gene bodies (Hypothesis A) and/or failure of SETD5–NCoR/HDAC3/G9a/PAF1 co-repressor scaffolding with enhancer/promoter hyperacetylation (Hypothesis B). Osipovich et al. showed Setd5-deficient cells have increased histone acetylation at transcription start sites and downstream regions (PMID:27864380). ↓ MOLECULAR: Dysregulated RNA Pol II dynamics — altered promoter-proximal pausing/release (with HCF-1 and PAF1; demonstrated on E2F target genes in HSCs, PMID:34853439), impaired elongation kinetics, and defective RNA maturation, splicing, and gene integrity (Sessa 2019). ↓ (parallel branch) MOLECULAR: Reduced rDNA transcription — SETD5 normally recruits HDAC3 to the rDNA promoter, removing H4K16ac and its reader TIP5 (a repressor of rDNA); SETD5 loss ⇒ rDNA repression ⇒ ↓ global translation ⇒ specifically ↓ cyclin D1 translation (PMID:32299058). ↓ CELLULAR: Neural stem/progenitor cell proliferation defect and altered cell-cycle dynamics; premature/altered differentiation; deficit of deep-layer cortical neurons (PMID:30655503). ↓ (parallel branch) CELLULAR: Mitochondrial compartment failure — "Low levels of SETD5 resulted in fragmented mitochondria, reduced mitochondrial membrane potential" and reduced ATP production, with mitochondria depleted from neurites and synapses; "Mitochondrial impairment is facilitated by transcriptional aberrations originated by SETD5" (PMID:37264456) [UNVERIFIED-QUOTE]. ↓ (parallel branch, non-cell-autonomous) CELLULAR: Astrocyte dysfunction — SETD5-deficient hiPSC-derived astrocytes show increased extracellular ROS, glutamate, IL-6 and IL-8; "Elevated astrocytic IL-6 exerts a non-cell autonomous harmful effect on healthy neurons", driven by JAK/STAT (PMID:41993368, bioRxiv preprint 2026 — preprint, not peer-reviewed; curate as EMERGING) [UNVERIFIED-QUOTE]. ↓ CELLULAR/TISSUE: Reduced synaptic density and neuritic outgrowth; decreased network activity and synchrony on MEA; abnormal postsynaptic density protein expression; enhanced long-term potentiation (a paradoxical/dissociated LTP finding worth flagging) (PMID:30655503, PMID:30455454). ↓ (developmental branch) TISSUE: Neural crest–derived and cardiopharyngeal-mesoderm developmental defects → craniofacial dysmorphism and outflow-tract cardiac malformation. "Setd5 was required in cardiopharyngeal mesoderm for progression of the heart tube" through the ballooning stage (PMID:34050709) [UNVERIFIED-QUOTE]; Deliu et al. report "neural crest defect-associated phenotypes" [CACHE-VERIFIED]. ↓ ORGANISM: Global developmental delay, intellectual disability, autism/ADHD/OCD, hypotonia, dysmorphism, CHD, skeletal anomalies.

6.4 Molecular pathways (KEGG/Reactome/GO framing)

• Chromatin/transcription: H3K36 methylation; H3K9 methylation (via G9a); H3K27/H3K9/H4K16 deacetylation (via NCoR-HDAC3); RNA Pol II pausing and elongation (PAF1, HCF-1); enhancer priming→activation transition
• Ribosome biogenesis / translation: rDNA transcription (RNA Pol I), TIP5/NoRC, cyclin D1 translational control
• Cell cycle: E2F target genes, cyclin D1, G1/S
• PI3K–AKT(–mTOR): implicated chiefly in the cancer literature (PMID:37963940, PMID:35063407); its relevance to the germline NDD is unproven — flag as a knowledge gap
• JAK/STAT–IL-6: astrocytic, preprint-stage
• Semaphorin/axon guidance: SetD5–BRD2 co-occupancy at the Sema3a promoter/TSS regulates Sema3A in retinal ganglion cells, independent of the SET domain (PMID:29180574) — a clean, published SET-domain-independent function
• Protein turnover: APC/C-mediated ubiquitin–proteasome degradation of SETD5 acts as a molecular switch for enhancer activation (PMID:34857762)

6.5 Suggested GO / CL / UBERON / CHEBI terms

Molecular function (OLS-verified live): - GO:0046975 — histone H3K36 methyltransferase activity ✅ verified - GO:0140955 — histone H3K36 trimethyltransferase activity ✅ verified - GO:0140003 — histone H3K36me3 reader activity ✅ verified

❌ DO NOT USE (confirmed obsolete): GO:0010452, GO:0034968.

Biological process (suggested — all require OAK verification): - GO:0006357 regulation of transcription by RNA polymerase II - GO:0032968 positive regulation of transcription elongation by RNA polymerase II - GO:0006338 chromatin remodeling - GO:0006360 transcription by RNA polymerase I (rDNA branch) - GO:0042254 ribosome biogenesis - GO:0006412 translation - GO:0000381 regulation of alternative mRNA splicing, via spliceosome - GO:0008283 cell population proliferation - GO:0021895 cerebral cortex neuron differentiation - GO:0007416 synapse assembly - GO:0000266 mitochondrial fission - GO:0014032 neural crest cell development - GO:0016575 histone deacetylation (verify — this namespace was affected by the same GO cleanup)

Cell types (CL — verify each): - CL:0000047 neuronal stem cell / CL:0011020 neural progenitor cell — primary proliferative target - CL:0000679 glutamatergic neuron; deep-layer cortical projection neurons - CL:0000127 astrocyte — non-cell-autonomous IL-6 arm - CL:0000333 migratory cranial neural crest cell — craniofacial/cardiac - CL:0000604 retinal rod cell; CL:0000636 Mueller cell — retinal survival/proliferation (PMID:36349512) - CL:0000037 hematopoietic stem cell — quiescence (PMID:34853439) - CL:0000115 endothelial cell — miR-126-5p/SetD5/Sema3A axis - CL:0000136 adipocyte — adipogenic enhancer switch

Anatomy (UBERON — verify each): see §7.

Chemicals (CHEBI — verify each): - CHEBI:15414 S-adenosyl-L-methionine (methyl donor cofactor) - CHEBI:8871 risperidone (zebrafish rescue) - CHEBI:15361 pyruvate / glycolytic intermediates (cancer arm only)

6.6 Molecular profiling

• Transcriptomics: RNA-seq of fetal Setd5^+/− cortical neurons showed a specific subpopulation with altered neurodevelopmental gene expression (PMID:30655503); Setd5-null mESCs show "substantially altered gene expression" (PMID:27864380); adult zebrafish setd5 brain shows "downregulation of genes encoding proteins involved in the synaptic structure and function" (PMID:36613611). Datasets in GEO under these accessions (specific GSE IDs not retrieved — look up before citing).
• Epigenomics: genome-wide H3K36me3 ChIP-seq (Sessa 2019); H3/H4 acetylation ChIP (Osipovich 2016); Pol II ChIP/pausing-index (Deliu 2018, Li 2022); blood DNA-methylation EPIC arrays for episignature.
• Proteomics: SETD5 interactome (PAF1/Ctr9, NCoR/HDAC3, TBL1XR1, G9a, HCF-1, BRD2) — mass-spec IP studies in Osipovich 2016, Deliu 2018, Wang 2020, Li 2022.
• Metabolomics / lipidomics: none for the germline disorder. Bioenergetic (ATP, ΔΨm) readouts only (PMID:37264456). Data gap.
• Single-cell / spatial: no published scRNA-seq or spatial transcriptomics of patient tissue. Data gap.
• Functional genomics screens: a scalable high-throughput neural-development platform assayed shared ASD-gene impact on cell fate/differentiation including SETD5 (cached as PMID_35197626). DepMap holds SETD5 dependency data for cancer lines.

7. Anatomical Structures Affected

Organ level — primary - Brain / central nervous system — UBERON:0000955 brain; UBERON:0000956 cerebral cortex (deep layers particularly); UBERON:0002336 corpus callosum (MRI abnormalities reported); UBERON:0002037 cerebellum (ataxia/gait). Adult Setd5^+/− mice show MRI-detectable anatomical differences and abnormal brain-to-body weight ratio. - Craniofacial skeleton — UBERON:0001474 bone element / UBERON:0007811 craniocervical region; neural-crest-derived facial structures.

Organ level — secondary / systemic - Heart — UBERON:0000948; specifically outflow tract UBERON:0004145, interventricular septum UBERON:0002094, interatrial septum UBERON:0002085 - Skeleton / limbs — UBERON:0002101 limb; vertebral column UBERON:0000955→ use UBERON:0002240 spinal column (verify); asymmetric long-bone growth - Eye / retina — UBERON:0000966 retina - Pituitary gland — UBERON:0000007 (congenital hypopituitarism, emerging) - Kidney / urinary tract — UBERON:0002113 kidney (CAKUT, emerging) - GI tract — UBERON:0001155 colon (constipation) - Cerebral vasculature — UBERON:0001621/UBERON:0001627 cerebral artery (moyamoya, emerging) - Diaphragm — UBERON:0001103 (CDH, rare) - Lung/airway — UBERON:0002185 bronchus (single aberrant-bronchus case)

Body systems: nervous (dominant), cardiovascular, musculoskeletal, visual, endocrine, digestive, genitourinary.

Tissue and cell level: neuroepithelium/ventricular zone (proliferating NSPCs), cortical plate neurons, astroglia, cranial neural crest, cardiopharyngeal mesoderm, retinal photoreceptors and Müller glia, bone (fragility), hematopoietic compartment (mouse only).

Subcellular level (GO CC — verify): - GO:0005634 nucleus — principal localization; NLS-dependent - GO:0000785 chromatin - GO:0005730 nucleolus / rDNA promoter (SETD5 acts on rDNA but is not itself nucleolar-resident in all reports) - GO:0005739 mitochondrion — secondary, disease-relevant compartment (fragmentation, ΔΨm loss) - GO:0045202 synapse — reduced mitochondrial occupancy in mutant neurites/synapses - GO:0000502 proteasome complex — APC/C-mediated SETD5 turnover

Localization/lateralization: bilateral and symmetric. Lower-limb asymmetry (leg-length discrepancy) is the notable exception and is an asymmetric skeletal feature.

8. Temporal Development

Onset - Congenital: facial dysmorphism, CHD (detectable on fetal cardiac ultrasound — PMID:41368699), CDH, structural brain anomalies - Neonatal–infantile: hypotonia, feeding difficulties, drooling - Infancy/toddlerhood: global developmental delay, delayed walking, delayed speech (HP:0003593 Infantile onset) - Childhood: ID becomes quantifiable; autism/ADHD/OCD/stereotypies; epilepsy when present - Adolescence/adulthood: scoliosis, short stature, psychiatric decompensation; rare late cerebrovascular events (moyamoya) - Onset pattern: chronic, insidious, developmental

Progression - Core cognitive phenotype: static/non-progressive. There is no evidence of neurodegeneration. - Progressive elements: scoliosis/kyphosis, leg-length discrepancy, possibly bone fragility. - Episodic elements: epilepsy (~14%); psychiatric exacerbations. - Duration: chronic, lifelong. Oldest reported individuals: 37 years (BGR cohort), 36 years (case report), plus mildly affected transmitting parents into middle age. - Progression rate: not formally quantified — no natural-history study exists. This is the single largest clinical-data gap.

Patterns - Remission: none. - Critical periods: (a) embryonic — neural tube closure, somitogenesis, cardiac ballooning, neural crest migration (mouse null lethal at E10.5–11.5); (b) fetal/early postnatal cortical neurogenesis (deep-layer neuron generation); (c) early childhood — the window for developmental/behavioral intervention; (d) growth — the case report of rhGH initiated at age 12 with +3.16 SDS gain in year 1 argues the pubertal window is actionable for the growth phenotype. - Diagnostic delay: mean age at diagnosis 9.2 years (survey, PMID:42468298) — an actionable health-services finding.

9. Inheritance and Population

Epidemiology

• No population prevalence or incidence estimate exists. Orphanet's entry (ORPHA:404440) is obsolete and carries no prevalence class. For a dismech Prevalence record, the defensible values are measure_type: CASES_IN_LITERATURE and prevalence_class: ULTRA_RARE / NOT_YET_DOCUMENTED, with a notes field explaining why.
• Cumulative reported cases: "The disorder was first described in 2014 with fewer than 75 reported cases in the literature to date" (Talaba et al. 2026, PMID:42468298) [CACHE-VERIFIED]. That count predates the 28-patient De Falco cohort and the 13-person BGR cohort being fully absorbed into the literature tally, so ~120–150 published individuals is a reasonable 2026 estimate.
• Diagnostic yield within ID cohorts — the best available "frequency" anchor: ~0.7%. Two independent cohorts converge:
• Grozeva 2014: 7 LoF variants in 996 individuals screened → "rare de novo LoF mutations in SETD5 are a relatively frequent (0.7%) cause of ID" (PMID:24680889) [UNVERIFIED-QUOTE]
• Kuechler 2015: "The present report of two SETD5 LoF variants in 301 patients demonstrates a prevalence of 0.7% and thus SETD5 variants as a relatively frequent cause of ID" (PMID:25138099) [UNVERIFIED-QUOTE]

⚠️ Curate this carefully: 0.7% is a diagnostic yield within an ascertained moderate-to-severe ID cohort, not a population prevalence. Do not enter it as rate_per_100000.

Genetic epidemiology

• Inheritance pattern: autosomal dominant (HP:0000006). Overwhelmingly de novo.
• Penetrance: incomplete/reduced. This is well documented and clinically important for counseling:
• Powis et al. 2018: "We also present an apparently unaffected carrier mother of an affected individual and a carrier mother with normal intelligence and affected twin sons." [CACHE-VERIFIED, PMID:28881385]
• De Falco et al. 2025 explicitly describe the disorder as having "incomplete penetrance" [CACHE-VERIFIED, PMID:39603091]
• Szczałuba et al. 2016 reported the first familial case: two siblings and their father, with "the father demonstrated only mild intellectual impairment" (PMID:27375234) [UNVERIFIED-QUOTE]
• ⚠️ Note tension: Genomics England PanelApp records "complete penetrance" on the Intellectual disability panel. The published literature contradicts this. Curate incomplete penetrance and flag the discrepancy.
• Expressivity: highly variable, both between and within families — normal IQ through severe ID in the same series.
• Genetic anticipation: none (not a repeat-expansion disorder). Not applicable.
• Germline mosaicism: not specifically reported for SETD5, but recurrence in siblings of unaffected parents is theoretically possible; standard counseling caveat applies. No data.
• Founder effects: none identified. No data.
• Consanguinity: not relevant (dominant LoF). Not applicable.
• Carrier frequency: not applicable in the recessive sense; population LoF frequency is essentially zero (pLI ≈ 1).

Population demographics

• Affected populations: no ancestry-specific enrichment reported. Cases published from Europe, North America, South America (Brazil), Asia (Japan, China, India), the Middle East, and 12 countries in the support-group survey. The India NDD cohort study (PMID:38114583) confirms presence in LMIC populations.
• Geographic distribution: worldwide; no endemic pattern; no variant-specific geography.
• Sex ratio: approximately 1:1; no sex bias reported. Autosomal.
• Age distribution of ascertained individuals: heavily pediatric — 80% under 18 years in the survey cohort; BGR range 2–37 years. This reflects ascertainment bias (exome sequencing of children with DD), not true age distribution. Adults are systematically under-ascertained — a recognized gap.

10. Diagnostics

Primary diagnostic modality: genomic sequencing

There is no biochemical or imaging biomarker; diagnosis is molecular.

Recommended approach (in practical order): 1. Trio exome sequencing (WES) or genome sequencing (WGS) — the highest-yield first-line test for unexplained GDD/ID with dysmorphism. Essentially every published SETD5 case was ascertained this way. Trio design is essential given the de novo mechanism. 2. Chromosomal microarray (CMA) — mandatory in parallel or first, since a meaningful minority of cases are 3p25.3 microdeletions or 3p terminal deletions that WES may miss (2/28 in the De Falco cohort were CNVs). Resolution matters: the smallest reported deletion is 116 kb. 3. Targeted NGS ID/NDD panels — SETD5 is on all major panels; Genomics England PanelApp lists it GREEN on Intellectual disability, DDG2P, Fetal anomalies, Skeletal dysplasia, and Early onset or syndromic epilepsy. 4. Single-gene SETD5 sequencing — GTR test 582578 (sequence analysis, all coding exons, postnatal) exists but is rarely the right first test. 5. DNA methylation episignature (EpiSign) — a genuinely useful adjunct for this gene. SETD5/MRD23 has a validated episignature in the EpiSign classifier. Use cases: (a) reclassifying a SETD5 VUS; (b) resolving the KBG/CdLS/SETD5 differential; (c) NGS-negative cases with a compelling phenotype. Supporting evidence: PMID:32109418, PMID:34906459, PMID:41957673. 6. Prenatal: fetal cardiac ultrasound detecting ASD/VSD prompted prenatal WES yielding a de novo SETD5 frameshift (PMID:41368699) — "Fetal cardiac ultrasound represents a valuable tool for early screening" [UNVERIFIED-QUOTE].

Not indicated / not applicable: karyotyping (resolution far too low), FISH (unless confirming a known deletion), mtDNA testing, repeat-expansion testing.

Supportive clinical evaluations (not diagnostic, but standard-of-care workup)

Biomarkers: none validated. Data gap. No FDA/BEST-listed biomarker.

Biopsy/histopathology: not indicated. The only pathology literature is the single case of severe cerebral cortical dysplasia (PMID:28263952).

Clinical criteria and differential diagnosis

No consensus clinical diagnostic criteria exist (unlike CdLS or KBG). Diagnosis = pathogenic/likely pathogenic SETD5 LoF variant or deletion + compatible phenotype.

Differential diagnosis (all are documented real-world misdiagnoses of SETD5 cases):

Mechanistic basis of the KBG overlap (important, recently resolved): ANKRD11 (KBG) sits upstream of SETD5.

"ANKRD11-deficient neural cells exhibit reduced ribosomal RNA (rRNA) and translation"; "it indirectly promotes rRNA expression by upregulating SETD5"; "ANKRD11 interacts with the Setd5 promoter and recruits WDR5"; "Overexpression of ANKRD11 or SETD5 restores rRNA levels and translational activity." — Ito et al. 2025, iScience (PMID:40520101) [UNVERIFIED-QUOTE]

This converts KBG↔SETD5 phenocopy from a clinical curiosity into a shared ANKRD11→SETD5→rRNA/translation axis and is high-value content for a dismech mechanism module.

Screening

• Newborn screening: not applicable — no biochemical marker, no NBS program.
• Carrier screening: not applicable (de novo dominant).
• Cascade screening: appropriate. Because of documented reduced penetrance and mildly affected transmitting parents, parental testing is mandatory when a SETD5 variant is found — do not assume de novo.
• Prenatal/PGT: available for families with a known variant; PGT-M feasible.

11. Outcome / Prognosis

⚠️ Prognosis for this disorder is essentially uncharacterized. No natural-history study, no survival analysis, no registry-based outcome data. Everything below is inference from case series.

Survival and mortality - Life expectancy: no data. No published deaths attributable to the syndrome; individuals reported into the fourth decade (36 y, 37 y) and mildly affected transmitting parents into middle age. - Mortality risk, where present, would be driven by comorbidities: congenital heart disease, epilepsy (SUDEP risk applies generically), aspiration in severe hypotonia/feeding difficulty, and — in the rare moyamoya association — stroke. - Homozygous/biallelic loss is presumably not viable in humans: the mouse null is embryonic lethal at E10.5–11.5 (PMID:27864380) and homozygous IMPC animals show "preweaning lethality, complete penetrance." - Disease-specific mortality: no data.

Morbidity and function - Lifelong ID in ~75% (mild→severe) plus 21% borderline; a minority with normal IQ. - Communication impairment is near-universal and functionally dominant. - Motor: 78% hypotonia, 59% gait abnormality; most walk, often late. - Independence: not systematically studied. Most published adults required support. Data gap. - Quality-of-life instruments: none applied. No EQ-5D, SF-36, PROMIS, or disease-specific PROM exists. Priority gap.

Complications Epilepsy; scoliosis/kyphosis requiring orthopedic management; leg-length discrepancy; bone fragility/fractures; constipation; feeding difficulty and growth failure; refractive error and strabismus; psychiatric decompensation (including psychotic disorder); rare cerebrovascular events; single reports of neuroblastoma and hypopituitarism.

Recovery potential: none — the developmental lesion is fixed. Interventions are supportive and habilitative; developmental gains occur with therapy but the underlying dosage defect is not correctable with any current treatment.

Prognostic factors - No validated prognostic model. Candidate factors, all unvalidated: - Variant class/position — De Falco et al. 2025 "conduct a comprehensive review of the available literature, suggesting a possible genotype-phenotype correlation" [CACHE-VERIFIED]. One case-report author speculated a 3'-terminal frameshift might retain partial activity, explaining a mild phenotype (PMID:28549204) — plausible but unproven. - Deletion size — larger 3p terminal deletions → more severe, additional features (microcephaly, seizures, cardiac). - Genetic background / second hits (Pizzo 2019). - Presence of epilepsy, CHD, or severe hypotonia at baseline. - Prognostic biomarkers: none. Data gap.

12. Treatment

There is no disease-modifying or targeted therapy. Management is entirely symptomatic, multidisciplinary, and habilitative.

Standard of care (all suggested NCIT terms require OAK verification)

Disease-specific pharmacological reports

Recombinant human growth hormone (rhGH) — the only pharmacologic intervention with a published SETD5-specific outcome:

"This is the first case of a patient with overlap syndrome due to SETD5 mutation treated with rhGH"; "After both one year (+3.16 SDS) and two years (+2.9 SDS), the growth rate significantly increased" (PMID:40869907, Genes 2025) [UNVERIFIED-QUOTE]

n = 1. Presented at −5.22 SDS height, bone age 3 years delayed, variant c.890_891delTT. Suggested annotation: treatment_term NCIT:C15986 Pharmacotherapy, therapeutic_agent somatropin (NCIT:C821 — verify), therapeutic_modality: PROTEIN_REPLACEMENT or PEPTIDE, evidence HUMAN_CLINICAL, and mark clearly as a single case report.

Preclinical / experimental leads (none in human trials)

1. Risperidone (CHEBI:8871) — rescued the social-interest deficit in setd5 heterozygous zebrafish: "Impairment in social interest is rescued by risperidone, an antipsychotic drug used to treat behavioral traits in ASD" (PMID:36613611) [UNVERIFIED-QUOTE]. evidence_source: MODEL_ORGANISM. Note this is a repurposed symptomatic agent, not disease-modifying.
2. JAK/STAT inhibition — in SETD5-deficient hiPSC astrocytes, "Pharmacological JAK-STAT inhibition restored extracellular IL-6 to basal levels and partially rescued astrocyte morphology and neuronal deficits" (PMID:41993368) [UNVERIFIED-QUOTE]. ⚠️ bioRxiv preprint (2026), not peer-reviewed. evidence_source: IN_VITRO, status EMERGING.
3. Mitochondrial targeting — the authors of PMID:37264456 propose "mitochondrial activity and dynamics may represent new therapeutic targets," explicitly contingent on confirmation in patient-derived systems [UNVERIFIED-QUOTE]. evidence_source: IN_VITRO/MODEL_ORGANISM; hypothesis-stage.
4. HDAC3 / epigenetic modulation — mechanistically motivated by the SETD5–HDAC3–NCoR axis, but no in vivo NDD rescue has been shown. Directionality is non-obvious (SETD5 loss already de-represses acetylation). Curate only as a hypothesis with the direction problem flagged.
5. ANKRD11/SETD5–rRNA axis — overexpression of either restored rRNA/translation in KBG models (PMID:40520101); a conceptual, not clinical, lead.

Clinical trials

No interventional clinical trial specific to SETD5 haploinsufficiency was identified on ClinicalTrials.gov. The relevant registries are observational: - National Brain Gene Registry (BGR) — includes SETD5 Disorder participants; multi-site US (PMID:38632549, PMID:40265665). Look up its NCT/registry identifier before citing as a clinical_trials entry. - Patient community: a SETD5-related disorder Facebook support group serving as an informal registry (PMID:42468298).

Pharmacogenomics

No SETD5-specific pharmacogenomic guidance. Standard CPIC guidance applies to any psychotropic/antiseizure drugs used (e.g., CYP2D6 for risperidone/aripiprazole; HLA-B15:02 for carbamazepine). Not disease-specific.*

Gene/RNA/cell therapy

None in development. Conceptually, a haploinsufficiency disorder with a 1,442-aa nuclear scaffold protein and a largely prenatal developmental critical window presents severe barriers to gene replacement or ASO-mediated upregulation. No data.

13. Prevention

Primary prevention: not possible. The disorder arises from de novo mutation; no modifiable exposure is known.

Secondary prevention (early detection): - Early genomic diagnosis is the actionable target. Mean age at diagnosis is 9.2 years — a substantial, reducible delay. First-tier trio WES/WGS + CMA in unexplained GDD/ID would compress this. - Prenatal detection via fetal cardiac ultrasound → prenatal WES has been demonstrated (PMID:41368699). - No population screening program is applicable.

Tertiary prevention (complication prevention in diagnosed individuals): — this is where prevention effort actually lives. Note that no formal surveillance guideline exists for SETD5; the list below is a reasoned synthesis of reported complications, not a published protocol, and should be curated as such. - Baseline echocardiogram at diagnosis - EEG if paroxysmal events; low threshold given ~14% epilepsy with subtle presentations - Serial scoliosis and leg-length assessment through growth - Bone health assessment given the reported fragility association - Growth monitoring; GH-axis evaluation for significant short stature - Ophthalmology and audiology baseline plus periodic re-evaluation - Nutrition/GI review (feeding, constipation) - Proactive psychiatric screening (anxiety 47%; psychotic disorder reported) - Renal ultrasound is reasonable but unproven given the emerging CAKUT signal - No cancer surveillance is recommended — the neuroblastoma link rests on a single report with no risk estimate

Immunization: routine schedule; no contraindication or special schedule. Not applicable as disease-specific prevention.

Genetic screening / counseling: - Parental testing is required in every case — reduced penetrance means an apparently unaffected parent may carry the variant (PMID:28881385). - Recurrence risk: ~1% (gonadal mosaicism) if both parents test negative; 50% if a parent is a carrier, with the crucial caveat that severity is unpredictable. - Prenatal diagnosis and PGT-M available for known familial variants. - Suggested term: NCIT:C15240 Genetic Counseling.

Public health / environmental interventions: not applicable.

14. Other Species / Natural Disease

• Naturally occurring disease in other species: none reported. No OMIA entry for a natural SETD5-related disorder in any domestic or wild species. All animal disease is experimentally induced. Not applicable.
• Zoonotic potential / cross-species transmission: not applicable (non-infectious genetic disorder).
• Veterinary relevance: none.

Orthologs and evolutionary conservation

Comparative pathology: the Setd5^+/− mouse recapitulates the human disorder's neural-crest/craniofacial, cardiac, cognitive, and behavioral axes remarkably well (see §15). Conservation extends to the biochemical mechanism: "SETD5 functions in a manner similar to yeast Set3p and Drosophila UpSET" (PMID:27864380) [UNVERIFIED-QUOTE] — i.e., the HDAC-recruiting, acetylation-restricting function is conserved from yeast to human, which is strong support for Hypothesis B (§6.1).

15. Model Organisms

15.1 Mouse — the primary model

Alleles/resources: - Setd5<tm1a(EUCOMM)Wtsi> — knockout-first conditional-ready; IMPC/KOMP/EUCOMM. MGI:1920145. Available via IMSR/EMMA/MMRRC. - Conditional (floxed) alleles used for cardiopharyngeal mesoderm–specific deletion (PMID:34050709) - Independent lab-generated Setd5^+/− lines: Osipovich/Magnuson (Vanderbilt), Deliu/Novarino (IST Austria), Moore/Muotri-adjacent (PMID:30655503), Sessa/Broccoli (San Raffaele), Nakagawa (Tohoku)

Homozygous null: embryonic lethal E10.5–11.5, with "severe defects in neural tube formation, somitogenesis and cardiac development" and aberrant vasculogenesis in embryo, yolk sac, and placenta (PMID:27864380). IMPC: "Preweaning lethality, complete penetrance."

Heterozygous — IMPC systematic phenotyping (Setd5<tm1a(EUCOMM)Wtsi>, het):

Note the strong craniofacial/dental signal — an excellent cross-species match to the human facial gestalt and dental crowding, and a hint that hearing (ABR) and dentition deserve more systematic human assessment.

Heterozygous — hypothesis-driven studies:

Deliu et al. 2018, Nat Neurosci (PMID:30455454) [CACHE-VERIFIED]:

"Setd5-haploinsufficient mice present developmental defects such as abnormal brain-to-body weight ratios and neural crest defect-associated phenotypes. Furthermore, Setd5-mutant mice show impairments in cognitive tasks, enhanced long-term potentiation, delayed ontogenetic profile of ultrasonic vocalization, and behavioral inflexibility. Behavioral issues are accompanied by abnormal expression of postsynaptic density proteins previously associated with cognition. Our data additionally indicate that Setd5 regulates RNA polymerase II dynamics and gene transcription via its interaction with the Hdac3 and Paf1 complexes."

Moore et al. 2019, Transl Psychiatry (PMID:30655503): reduced synaptic density and neuritic outgrowth in cultured cortical neurons; reduced MEA network activity and synchrony; altered gene expression in a fetal cortical neuron subpopulation; hyperactivity, cognitive deficit, altered social interaction; MRI-detectable adult brain differences; deficit of deep-layer cortical neurons in the developing brain; described as "consistent with a highly penetrant risk factor."

Sessa et al. 2019, Neuron (PMID:31515109): impaired NPC proliferative dynamics and synaptic wiring; genome-wide H3K36me3 loss; behavioral deficits.

Cheung et al. 2021, Genesis (PMID:34050709): double outlet right ventricle + perimembranous VSD; conditional deletion localizes requirement to cardiopharyngeal mesoderm; no genetic interaction with Tbx1.

Nakagawa et al. 2020, iScience (PMID:32299058): Setd5^+/− mice show autism-related behaviors with disturbed ribosomal protein gene and rDNA expression in brain.

Li et al. 2022, Leukemia (PMID:34853439): hematopoietic-specific deletion → increased immunophenotypic HSCs, impaired long-term self-renewal, loss of LT-HSC quiescence via HCF-1/PAF1-dependent Pol II pause release on E2F targets.

Matsumura et al. 2021, Nat Commun (PMID:34857762): SETD5–NCoR-HDAC3 gates Cebpa/Pparg enhancers; APC/C-mediated SETD5 degradation is the adipogenic switch.

Phenotype recapitulation — strong. The mouse het reproduces: craniofacial/neural-crest dysmorphology, outflow-tract cardiac defects, cognitive deficits, social/communication deficits (USV), behavioral inflexibility (an OCD/rigidity analog), reduced grip strength (hypotonia analog), vertebral anomalies, and the core molecular lesion.

Model limitations: (a) no reported spontaneous seizures despite human epilepsy in ~14%; (b) enhanced LTP in mice is hard to map onto human cognition; (c) mouse cortex lacks human-specific outer radial glia/OSVZ biology relevant to an NSPC-proliferation disorder; (d) no leg-length-discrepancy analog; (e) the human catalytic-activity question is not resolved by the mouse.

Curation note: limitation (c) — and more broadly the question of whether murine NSPC proliferation phenotypes translate to human corticogenesis — is a textbook case for discussions with kind: HUMAN_MODEL_MISMATCH rather than generic KNOWLEDGE_GAP, since the evidence exists in the model and it is the translational validity that is open.

15.2 Zebrafish

setd5 CRISPR/Cas9 heterozygous mutants (PMID:36613611): defective aggregation and shoaling coordination, indifference to social stimuli; adult-brain downregulation of synaptic structure/function genes suggesting hypo-connectivity; risperidone rescues social interest. Positioned as "a promising setd5 haploinsufficiency model" for drug screening. Also used to confirm H3K36me3 loss (Sessa 2019). ZFIN is the resource database.

15.3 Invertebrate / amphibian

Drosophila UpSET and X. laevis setd5 (PMID:33819264): two-hit interaction platform; SETD5–MOSMO synergy producing axon-outgrowth defects. Databases: FlyBase, Xenbase.

15.4 Cellular / in vitro

• Mouse ESCs — reduced proliferation, increased apoptosis, impaired cell-cycle progression and cardiomyocyte differentiation (PMID:27864380); required for primordial-germ-cell specification genes via Tbl1xr1/Ctr9 (Yu et al. 2017, Cell Biochem Funct 35:247–253)
• Neural stem cells — H3K36me3 and RNA-elongation phenotypes (Sessa 2019)
• hiPSC-derived neurons and astrocytes — mitochondrial phenotypes (PMID:37264456); astrocyte IL-6/JAK-STAT (PMID:41993368, preprint)
• Mouse retinal explants + shRNA — Setd5, not Setd2, required for retinal cell survival/proliferation; SET-domain-dependent; the SETD5^S1257 separation-of-function allele (PMID:36349512*)
• High-throughput neural-development platform profiling shared ASD-gene impact on cell fate/differentiation (cached PMID_35197626)

15.5 Model databases

MGI (MGI:1920145), IMPC, IMSR, EuMMCR/EUCOMM, KOMP, EMMA, MMRRC, ZFIN, FlyBase, Xenbase, Alliance of Genome Resources, DepMap (cancer dependency), Cellosaurus.

Summary of high-priority knowledge gaps (candidates for discussions: entries)

Key reference list (with cache status)

Non-PubMed sources used: ClinGen (search.clinicalgenome.org, gene HGNC:25566 — GDV Definitive 2023-07-27; dosage HI=3/TS=0, 2014-11-06); HGNC REST (rest.genenames.org); EBI OLS4 (MONDO:0014336, GO term verification); HPO/Jax annotation API (ontology.jax.org, OMIM:615761); IMPC solr (Setd5 genotype-phenotype, MGI:1920145); SFARI Gene (gene.sfari.org); Genomics England PanelApp API; Orphadata 2025-12-09 snapshot via references_cache/ORPHA_404440.md and ORPHA_435638.md.

Sources: - PubMed E-utilities (esearch/efetch) - ClinGen — SETD5 (HGNC:25566) - HGNC REST — SETD5 - EBI OLS4 — MONDO:0014336 - HPO annotations — OMIM:615761 - IMPC — Setd5 phenotypes - SFARI Gene — SETD5 - Genomics England PanelApp — SETD5 - ClinVar RCV000114962 — SETD5 c.3001C>T - GTR — Intellectual disability-facial dysmorphism syndrome due to SETD5 haploinsufficiency - Kuechler et al. 2015, Eur J Hum Genet - Crippa et al. 2020, Front Neurol (PMC7393934) - GARD — Intellectual disability-facial dysmorphism syndrome due to SETD5 haploinsufficiency - Levy et al., DNA methylation episignature testing improves molecular diagnosis of Mendelian chromatinopathies - EpiSign v5 methylation array panel content