Cardiofacioneurodevelopmental Syndrome

1. Disease Information

Overview

Cardiofacioneurodevelopmental syndrome (CFNDS) is an autosomal recessive multiple-congenital-anomaly syndrome caused by biallelic loss-of-function variants in CCDC32. Its recognized core is global developmental delay plus bilateral cleft lip and palate, with a variable constellation of craniofacial dysmorphism, congenital heart disease, hindbrain (cerebellar vermian) hypoplasia, microcephaly, digital/nail anomalies, postnatal growth restriction, and — in one individual — laterality disturbance (abdominal situs inversus with asplenia).

The MedGen/OMIM clinical definition:

"Characterized by microcephaly, midline facial defects, developmental delay, and cerebellar hypoplasia. Variable cardiac defects may be present, including atrioventricular canal and ventricular septal defects. Heterotaxy has also been reported." — MedGen UID 1721861

Abdalla et al. proposed the two-feature core:

"We describe a core phenotype comprising developmental delay and bilateral cleft lip and palate in the three individuals with CFNDS." — PMID:35451546 (verbatim, abstract)

Key identifiers

Synonyms / alternative names

• CFNDS (standard abbreviation)
• CCDC32-related cardiofacioneurodevelopmental syndrome (usage in Albuainain et al. 2026, PMID:41639596)
• CCDC32 deficiency / CCDC32-related syndrome (informal)
• Historical gene alias in older literature: C15orf57-related syndrome

Provenance of information

Entirely aggregated disease-level and case-level literature. All data derive from individual published case reports and one review; there is no EHR-derived cohort, no patient registry, no natural-history study, and no biobank cohort for CFNDS. No ICEES/COHD-style co-occurrence data exist for this entity (it has no ICD code to key on).

2. Etiology

Disease causal factors

Purely monogenic and germline. CFNDS is caused by biallelic (homozygous, in all published families) loss-of-function variants in CCDC32. No environmental, infectious, or acquired etiology is known or plausible.

"Using whole exome sequencing, we identified homozygous frameshift CCDC32 variants in three affected individuals." — PMID:32307552 (verbatim, abstract)

All reported probands to date are homozygous, arising in the context of parental consanguinity in at least two of the founding families. Compound heterozygosity is theoretically expected but has not yet been reported.

Risk factors

Genetic (causal): - Biallelic CCDC32 LoF variants (frameshift, whole-gene/multi-exon deletion, nonsense) — see §4. - Carrier parents (obligate heterozygotes are unaffected; no heterozygous phenotype reported).

Non-genetic modifier of risk: - Consanguinity is the dominant epidemiological risk factor, as for essentially all ultra-rare AR disorders. Family A was a consanguineous Arab Muslim pedigree (first cousins once removed); Family B a consanguineous Iranian (Isfahan) pedigree [PMC7268788, full text]. - Founder/population endogamy may explain the apparently recurrent large deletion allele (§4).

Environmental risk factors: none identified. No toxin, teratogen, maternal exposure, occupational, dietary, or lifestyle risk factor has been implicated. Given complete genetic determinism at the causal level, none is expected.

Age / sex / family history: Disease is congenital; sex distribution is unremarkable in the tiny cohort (roughly balanced). Family history of consanguinity or an affected sib is the actionable risk signal.

Protective factors

None identified. No protective allele, modifier allele, or environmental protective factor has been reported. gnomAD contains no reported homozygous LoF individuals that would suggest incomplete penetrance or a protective background.

Gene–environment interactions

None identified. No GxE data exist for CFNDS. CTD contains no CCDC32–chemical–disease interaction of relevance to this phenotype.

Curation note / open question: Harel et al. explicitly raised genetic modifiers as a candidate explanation for the incomplete ciliopathy phenotype (absence of cystic kidney disease and polydactyly), citing possible "cis/trans genetic interactions" [PMC7268788, full text]. This is an appropriate KNOWLEDGE_GAP discussion item.

3. Phenotypes

3.1 HPO annotation set (authoritative, from HPOA)

Retrieved from the HPO API for OMIM:619123. Sole annotation source: PMID:32307552 (2 individuals). Frequencies are literal patient counts.

Note the hypertelorism/hypotelorism split (1/2 each) — this is genuine phenotypic discordance between the two index patients, not an annotation error: Individual A-II-1 had hypotelorism, Individual B-II-1 had hypertelorism [PMC7268788, full text]. This is mechanistically interesting (both are midline-patterning readouts in opposite directions) and worth a curation note.

3.2 Additional phenotypes from later reports (not yet in HPOA)

Suggested HPO terms below were verified against OLS:

Additional descriptive features from full text, without a crisply matching specific HP term, best captured as free-text preferred_term on a broader parent: stiff upper lip, vaulted palate, underdeveloped helices, broad nasal root, prominent large nose, nail clubbing, small hands and feet, abnormal cisterna magna, missing teeth (HP:0000670 Carious teeth / HP:0000668 Hypodontia — verify against the exact clinical description before assigning).

3.3 Phenotype characteristics

Type distribution: Predominantly congenital structural malformations (craniofacial, cardiac, CNS, limb) plus neurodevelopmental/behavioral features. There are no reported disease-specific laboratory abnormalities — no biomarker, no metabolic derangement, no characteristic biochemical signature. This is important: CFNDS has no biochemical diagnostic handle.

Age of onset: Congenital (HP:0003577), 2/2. Structural anomalies are present prenatally — the terminated fetus A-II-2 was ascertained on prenatal imaging with bilateral cleft lip, vermian hypoplasia, hypoplastic pons, and abnormal cisterna magna [PMC7268788, full text]. Developmental delay declares itself in infancy; feeding difficulties are an early-infancy presentation (2/2).

Severity: Variable. Structural anomalies range from lethal-in-utero-decision severity (fetus) through moderate (VSD + pulmonic stenosis, moderate motor/language delay) to comparatively mild growth findings (one patient at 80th centile height). Neurodevelopmental severity spans "moderately delayed motor and language development" to frank intellectual disability.

Progression: The malformations are static (non-progressive congenital structural defects). The neurodevelopmental phenotype is developmental, not neurodegenerative — no regression, no progressive neurological decline has been reported in any patient. Cerebellar and callosal hypoplasia are developmental (hypoplasia), not atrophic.

Frequency: See table. All frequencies are n/2 or n/6 and are not generalizable. Only DD and bilateral cleft lip/palate approach "obligate" status, and even that is a proposal from a three-patient series.

3.4 Quality-of-life impact

No CFNDS-specific QoL data exist. No EQ-5D, SF-36, PROMIS, or disease-specific PROM has been administered. The following are reasoned extrapolations from the constituent phenotypes and should be curated as such (not as CFNDS evidence):

• Bilateral cleft lip/palate (2/2 → likely near-obligate): feeding difficulty in infancy, speech/resonance impairment, recurrent otitis media and conductive hearing loss, multiple staged surgeries through childhood, facial-appearance psychosocial burden. Highest-burden single feature.
• Global developmental delay / ID: dominant long-term determinant of independence and caregiver burden.
• Congenital heart disease (AVSD/VSD/PS): surgical morbidity; AVSD in particular carries meaningful operative and long-term valve-function burden.
• Conductive hearing loss: compounds the speech impact of cleft palate.
• Feeding difficulties (2/2): early-infancy nutritional and caregiver burden.
• Asplenia (with heterotaxy): lifelong invasive-infection risk requiring prophylaxis — see §12/§13.

4. Genetic / Molecular Information

4.1 Causal gene

⚠️ Transcript-nomenclature hazard. Published HGVS for the two founding frameshifts is on NM_001080791.2, but ClinVar reports the same alleles on NM_001080792.4 with different numbering. Curate both and note the transcript explicitly:

4.2 Pathogenic variants reported

Verbatim support for variant 3:

"We report a 9-year-old female patient with CFNDS caused by a homozygous 32,583-bp deletion affecting CCDC32. Independent of the affected CCDC32 transcript variant this deletion likely leads to loss of the encoded protein." — PMID:35451546 (verbatim, abstract)

🔬 Novel observation worth flagging (hypothesis, needs verification): ClinVar records VCV001690313 and VCV002431643 have near-identical breakpoints (g.40529942_40562524 vs g.40529939_40562522) yet correspond to patients reported independently from different centres (Abdalla et al., Egypt/Hamburg; Fernandes da Rocha et al., Porto). This is consistent with a recurrent, likely repeat-mediated (NAHR) deletion allele at 15q15.1, or possibly a founder allele. I could not confirm the Fernandes da Rocha breakpoints directly — that paper has no abstract in PubMed and the full text is paywalled. Curate this as a KNOWLEDGE_GAP/hypothesis, not as an established fact.

Variant classification (ACMG/AMP): ClinVar carries 4 Pathogenic CFNDS-associated CCDC32 alleles (two frameshift dups, two large deletions), all with the criteria-provided or no-assertion review status — none are expert-panel reviewed. One nonsense VUS exists. There are no ClinGen variant-curation assertions for CCDC32.

Variant class distribution: exclusively loss-of-function — frameshift duplications, whole-gene/multi-exon deletions, and (VUS) nonsense. No pathogenic missense variant has been reported. This matters mechanistically: the disease is a pure LoF/hypomorph disorder, not a gain-of-function or dominant-negative one.

Allele frequency: Both founding frameshifts were:

"Absent from gnomAD, TOPMed, Geno2MP, and GME Variome" [PMC7268788, full text]

Gene-level constraint (gnomAD v4.0): pLI = 0.19, LOEUF = 0.76. Both indicate CCDC32 is tolerant of heterozygous LoF — exactly the expected signature for a recessive disease gene, and a useful reassurance that carriers are unaffected. (LOEUF 0.76 sits above the conventional <0.6 Mendelian-dominant threshold.)

Somatic vs germline: Germline only. No somatic role. COSMIC/TCGA report no recurrent driver role for CCDC32.

Functional consequence: Loss of function. Mechanistic work (§6) shows the disease-associated truncations remove the α-helical AP-2-binding region:

"Disease-causing mutations…lack the α-helix encoded by residues 78–98…defective in AP2 binding" [PMC12768407, eLife, full text] "The CCDC32 mutant was defective in binding AP2 α and σ2 when expressed at similar levels as WT" [PMC11348294, PNAS, full text] "The AP2-regulating function of CCDC32 is disrupted by a disease-causing mutation" [PMC11348294, full text]

4.3 ClinGen gene–disease validity

This is a CGGV-citable structured assertion for dismech evidence. Note it is Moderate, not Definitive — appropriate epistemic humility for a 5-family disease. (Structured-source citation would take the form CGGV:<assertion_id>; the assertion ID must be resolved from the ClinGen CSV via just clingen-rebuild before citing.)

ClinGen dosage sensitivity: no curation exists (0 classifications) — so no CGDS citation is available.

4.4 Modifier genes

None identified. Harel et al. invoke unspecified modifiers to explain absent classical ciliopathy features, but no modifier locus has been mapped.

4.5 Epigenetic information

No data. No methylation episignature has been described for CFNDS. (This is a plausible future study — episignatures now exist for many Mendelian NDDs — and is a reasonable KNOWLEDGE_GAP.)

4.6 Chromosomal abnormalities

The pathogenic alleles include intragenic and whole-gene deletions at 15q15.1, i.e. CNVs rather than SNVs, in 3 of 5 families. Critically:

"This deletion was not detected in previous SNP array analyses and trio exome sequencing" [PMID:41639596]

No CFNDS case has been attributed to aneuploidy, translocation, inversion, or a contiguous-gene 15q15 microdeletion syndrome. Larger 15q duplications/gains overlapping CCDC32 appear in ClinVar but are unrelated pathogenic CNVs of other phenotypes.

5. Environmental Information

• Environmental factors: Not applicable. No toxin, radiation, pollutant, or occupational exposure is implicated. CTD contains no relevant CCDC32–chemical–disease axis for this phenotype.
• Lifestyle factors: Not applicable. No smoking, diet, exercise, or alcohol association. (Note: general periconceptional folate status modifies non-syndromic orofacial clefting risk in the population — this is not established as relevant to the syndromic, monogenic clefting in CFNDS and should not be curated as a CFNDS factor.)
• Infectious agents: Not applicable. No infectious etiology or trigger.

The only non-genetic factor with real epidemiological traction is consanguinity (§2, §9), which is a population-structure variable rather than an environmental exposure.

6. Mechanism / Pathophysiology

CFNDS has a genuinely contested and rapidly evolving mechanism, which is the most scientifically interesting aspect of this entry. Two models exist; the second has largely overtaken the first in molecular support, but the first carries the developmental-phenotype evidence.

6.1 Model A — Ciliopathy (Harel et al. 2020; the original proposal)

The founding paper framed CFNDS as a ciliopathy on the basis of phenotype overlap (laterality defect, cerebellar hypoplasia, craniofacial anomalies) plus direct functional data.

Full verbatim abstract (PMID:32307552) — safe for evidence snippets:

"Despite the wide use of genomics to investigate the molecular basis of rare congenital malformations, a significant fraction of patients remains bereft of diagnosis. As part of our continuous effort to recruit and perform genomic and functional studies on such cohorts, we investigated the genetic and mechanistic cause of disease in two independent consanguineous families affected by overlapping craniofacial, cardiac, laterality and neurodevelopmental anomalies. Using whole exome sequencing, we identified homozygous frameshift CCDC32 variants in three affected individuals. Functional analysis in a zebrafish model revealed that ccdc32 depletion recapitulates the human phenotypes. Because some of the patient phenotypes overlap defects common to ciliopathies, we asked if loss of CCDC32 might contribute to the dysfunction of this organelle. Consistent with this hypothesis, we show that ccdc32 is required for normal cilia formation in zebrafish embryos and mammalian cell culture, arguing that ciliary defects are at least partially involved in the pathomechanism of this disorder."

Supporting evidence: - Kupffer's vesicle (the zebrafish left-right organizer) cilia reduced in both number (P<0.001) and length (P<0.05) in ccdc32 crispants [MODEL_ORGANISM] - Ciliogenesis impaired in mouse IMCD3 cells on siRNA knockdown: reduced % ciliated cells (P<0.01) and reduced cilium length [IN_VITRO] - Human Protein Atlas independently localizes CCDC32 to the primary cilium, centrosome, and basal body (as well as plasma membrane and microtubules) - CCDC-family precedent: CCDC39, CCDC40, CCDC103, CCDC114 all cause primary ciliary dyskinesia

Authors' own caveats (important for balanced curation):

"Some of the hallmark ciliopathy pathologies were absent from the described individuals, such as cystic renal disease and polydactyly" [full text] "Whether this molecule also performs non-ciliary roles relevant to the human pathology remain unclear" [full text]

6.2 Model B — AP-2 adaptor assembly chaperone / clathrin-mediated endocytosis (2024–2026; now the dominant molecular model)

Three independent structural/cell-biological studies since 2024 have assigned CCDC32 a precise, non-ciliary molecular function: it is a dedicated assembly chaperone for the AP-2 clathrin adaptor complex.

The assembly pathway (PNAS 2024, PMID:39145939):

1. AAGAB binds the AP-2 α subunit, then recruits σ2 → AAGAB:α:σ2 ternary complex
2. CCDC32 recognizes AAGAB:α:σ2, displacing AAGAB → α:σ2:CCDC32 template complex
3. The template sequentially recruits μ2, then β2
4. β2 binding completes AP-2 and releases CCDC32

"AP2 assembly is controlled by a handover mechanism, switching from AAGAB-based initiation complexes to CCDC32-based template complexes" [PMC11348294, full text] "In mammalian cells deficient in AAGAB or CCDC32, all AP2 subunits are degraded" [full text] "CCDC32 is a general regulator of CME" [full text]

CCDC32 directly binds α, σ2, and μ2 — but not β2 (β2 binding is what evicts it). It is selective for AP-2 and does not regulate AP-1/AP-3/AP-4 assembly.

Cellular consequences of CCDC32 loss: AP-2 subunit degradation; loss of plasma-membrane AP-2 puncta; strongly reduced transferrin-receptor endocytosis with elevated surface TfR; impaired GLUT4 internalization.

Coated-pit dynamics (eLife 2026, PMID:41489497 / PMC12768407):

"siRNA-mediated knockdown of CCDC32 leads to the accumulation of unstable flat clathrin assemblies" [full text] "CCDC32 knockdown strongly inhibited CCP invagination" [full text] "CCDC32 interacts with AP2 via the α-AD [appendage domain]" [full text]

Notably, in this system knockdown "does not affect AP2 expression level" [full text] — an apparent tension with the PNAS degradation result, likely reflecting knockdown vs knockout depth. Worth curating as a hypothesis_groups distinction rather than smoothing over.

Structural mechanism and the membrane switch (Science Advances 2026, PMID:42234739):

"CCDC32 binds to α/σ2 in a multivalent manner, using at least three (extended FxDxF, dileucine, and AH1)" [full text] "CCDC32 uses a noncanonical WAPL (Wxxϕ) motif to bind in the same location as other tyrosine-containing cargo" [full text] "In solution, CCDC32 prevents complex assembly and actively disassembles AP-2 tetramers" [full text] "The presence of PIP2–containing membrane stabilizes the final stages of assembly" [full text] "Loss of CCDC32 significantly decreases the steady-state level of all four AP-2 subunits in HeLa cells" [full text]

So CCDC32 is a cargo-mimicking, membrane-gated chaperone: it occupies the cargo-binding sites of α/σ2 and μ2 to hold AP-2 in an assembly-competent but inactive state, and PI(4,5)P₂-containing membrane acts as the molecular switch that licenses final assembly at the plasma membrane. This elegantly couples AP-2 biogenesis to its site of action.

6.3 Reconciling the two models

These are not mutually exclusive, and the most defensible curated position is that AP-2/CME dysfunction is the primary molecular lesion, with ciliary phenotypes as a plausible downstream consequence:

• Clathrin-mediated endocytosis governs surface-receptor turnover, including receptors that traffic through the ciliary pocket and regulate Hedgehog/Wnt signalling — so CME failure can secondarily impair ciliogenesis and cilium-dependent signalling.
• The absence of classical ciliopathy features (no cystic kidney disease, no polydactyly, no retinal degeneration reported) argues against a primary, canonical ciliopathy.
• Independent genetic support for AP-2 hypofunction as a cause of exactly this phenotype spectrum: Ap2b1 (β2) null mice "survive until birth but then die shortly afterwards, the only obvious abnormality being that they have a cleft palate" [JCS review, full text]. Abdalla et al. made precisely this argument:

  "Cleft palate and cardiac defects observed in mice deficient of different AP2 subunits support a CCDC32 function in the AP2 complex." — PMID:35451546 (verbatim, abstract)

• AP2σ (σ2) and AP2μ (μ2) nulls are early embryonic lethal; complete AP-2 loss is embryonic lethal. CCDC32 loss is compatible with birth — consistent with CFNDS being a partial/hypomorphic AP-2 deficiency, i.e. residual AP-2 assembly persists (AAGAB-dependent, CCDC32-independent). The Sci Adv authors make this point:

  "Mutation of CCDC32 does not prevent embryonic development but results in patients with craniofacial malformations" [full text]

Another reported interaction: CCDC32 binds the C-terminus of annexin A2, itself a membrane–cytoskeleton and endocytosis protein [PMC7268788, full text] — consistent with the trafficking model.

6.4 Proposed causal chain (for a dismech pathograph)

Biallelic CCDC32 LoF (frameshift / multi-exon deletion)          [MOLECULAR]
  → Loss of CCDC32 α-helix aa78–98; failure to bind AP-2 α/σ2     [MOLECULAR]
    → Failure of the AAGAB→CCDC32 handover; no α:σ2:CCDC32 template [MOLECULAR]
      → Impaired AP-2 heterotetramer assembly; AP-2 subunit degradation [MOLECULAR]
→ Destabilized, flat, non-invaginating clathrin-coated pits [CELLULAR]
  → Reduced clathrin-mediated endocytosis / receptor internalization [CELLULAR]
    ├→ Impaired ciliogenesis (reduced cilium number and length)  [CELLULAR]  (hypothesis group: ciliary)
    │    → Defective left-right organizer (Kupffer's vesicle) function [TISSUE]
    │      → Aberrant southpaw/NODAL laterality signalling
    │        → Situs inversus, asplenia, cardiac looping defect  [ORGANISM]
    ├→ Impaired cranial neural crest / facial primordium morphogenesis [TISSUE]
    │    → Failure of lip and palatal shelf fusion
    │      → Bilateral cleft lip and palate; midline facial defects [ORGANISM]
    ├→ Impaired cardiac morphogenesis (septation / AV canal)     [TISSUE]
    │    → AVSD, VSD, pulmonic stenosis                          [ORGANISM]
    └→ Impaired cerebral/cerebellar growth and midline patterning [TISSUE]
 → Microcephaly, cerebellar vermis hypoplasia, hypoplastic
   corpus callosum, hypoplastic pons
     → Global developmental delay / intellectual disability [ORGANISM]

Upstream vs downstream: The AP-2 assembly failure is unambiguously upstream. The ciliary branch is contested in its position — Harel's model places cilia upstream of laterality/cerebellar defects directly; the newer model places cilia downstream of CME failure. Curate as two hypothesis_groups on the relevant edges (e.g. ciliary_primary vs cme_primary), with the AP-2 chain as the canonical/emerging model.

6.5 Ontology term suggestions for mechanism

GO biological process / cellular component (all verified via OLS):

Additional GO candidates to verify before use: protein-containing complex assembly (GO:0065003), chaperone-mediated protein complex assembly, phosphatidylinositol-4,5-bisphosphate binding, palate development, heart looping, cerebellum development.

CL cell types (verified):

Additional candidates (verify): ciliated epithelial cell, kidney collecting duct epithelial cell (the IMCD3 in-vitro model), cardiac myocyte, Purkinje cell, neuroepithelial cell.

UBERON (verified): UBERON:0004720 cerebellar vermis. Additional candidates to verify: secondary palate, upper lip, interventricular septum, atrioventricular canal, pulmonary valve, spleen, corpus callosum, pons, Kupffer's vesicle (zebrafish-specific).

CHEBI: phosphatidylinositol 4,5-bisphosphate (the membrane switch lipid) — verify the exact CHEBI ID before use.

6.6 Other mechanism dimensions

• Metabolic changes: None reported. No metabolic phenotype. (Note the GLUT4 internalization defect in CCDC32-deficient cells is a cell-biological observation, not a reported clinical metabolic phenotype in patients — do not over-read it into a diabetes claim.)
• Immune involvement: No primary immune mechanism. However, asplenia (1/2 in HPOA) produces functional hyposplenism with encapsulated-organism susceptibility — a secondary, anatomically-mediated immunodeficiency, clinically important (§12/§13).
• Tissue damage mechanisms: Not applicable in the classical sense — CFNDS is a developmental morphogenesis disorder, not a tissue-injury disorder. There is no oxidative stress, ischemia, fibrosis, or necrosis mechanism.
• Biochemical abnormalities: No enzyme deficiency, no channelopathy, no receptor defect measurable in patient fluids. The defect is a protein-complex assembly failure.
• Molecular profiling: No CFNDS transcriptomic, proteomic, metabolomic, or lipidomic patient study exists. The one transcriptomic application is diagnostic, not mechanistic — clinical RNA-seq of patient cells detecting exon 3–4 skipping (PMID:41639596).
• Single-cell / spatial / multi-omics: None for CFNDS. CCDC32 shows "low tissue specificity" and "low cell type specificity" in Human Protein Atlas, with mass-spec enhancement in lymphoid tissue and single-cell enhancement in fallopian tube — none of which maps onto the disease phenotype, reinforcing that the phenotype specificity comes from developmental context, not expression restriction.
• Functional genomics screens: No published CRISPR/RNAi screen result specific to CFNDS. CCDC32 is not a common-essential gene in DepMap-style screens (consistent with LOEUF 0.76 and viability of KO cell lines used in the mechanism papers).

7. Anatomical Structures Affected

Organ level

Primary (directly malformed): | System | Structures | HPO/UBERON anchor | |---|---|---| | Craniofacial | Upper lip, primary and secondary palate, mandible, external ear, orbits/interorbital distance, nose | HP:0410030, HP:0000175, HP:0000347, HP:0000411 | | Central nervous | Cerebellum (vermis), pons, corpus callosum, cerebral cortex (volume — microcephaly) | HP:0001320, HP:0012110, HP:0002079, HP:0000252; UBERON:0004720 | | Cardiovascular | Atrioventricular canal, interventricular septum, pulmonary valve | HP:0006695, HP:0001629, HP:0001642 | | Skeletal / limb | Digits (5th finger, phalanges), nails, spine (kyphosis), hands/feet size | HP:0004209, HP:0001156, HP:0012385, HP:0008386, HP:0002808 | | Genitourinary | Testis (descent) | HP:0000028 |

Secondary / complication-driven: - Spleen — asplenia (HP:0001746), in the context of the laterality defect - Abdominal viscera — situs inversus (HP:0003363) - Middle ear — conductive hearing loss, mechanistically secondary to cleft palate/eustachian dysfunction (HP:0000405) - Upper GI / feeding apparatus — feeding difficulties (HP:0008872), largely secondary to cleft and neurological status - Growth axis — postnatal growth restriction / short stature (HP:0004322)

Notably spared: Kidneys (normal renal ultrasound in both index patients), eyes/retina (normal ophthalmology in both), EEG normal in individual B-II-1 [PMC7268788, full text]. The renal and retinal sparing is the key argument against classical ciliopathy.

Body systems involved: cardiovascular, nervous, musculoskeletal, digestive, integumentary (nails/dermatoglyphics), reproductive, immune (via asplenia), sensory (auditory).

Tissue and cell level

• Cranial neural crest–derived mesenchyme of the frontonasal and maxillary/mandibular prominences — the presumptive lesion site for cleft lip/palate and the facial/ear/jaw dysmorphism (CL:0000008)
• Cardiac neural crest and second-heart-field derivatives — outflow/septation defects (CL:2000073)
• Left-right organizer ciliated epithelium (Kupffer's vesicle in fish; embryonic node in mammals) — laterality
• Cerebellar and pontine neuroepithelium / rhombic lip derivatives — vermian and pontine hypoplasia
• Palatal shelf epithelium and mesenchyme — fusion failure
• Epithelial tissue broadly, given CME is a general epithelial/all-cell process

Subcellular level (GO cellular component)

CCDC32 and its pathway localize to: - Plasma membrane and clathrin-coated pit / clathrin-coated vesicle (GO:0030136) — the principal site of action - AP-2 adaptor complex (GO:0030122) itself - Primary cilium, centrosome, basal body — per Human Protein Atlas - Microtubules - Cytosol (site of the AAGAB/CCDC32 pre-assembly intermediates)

Localization / lateralization

• Facial clefting is bilateral in the reported patients ("bilateral cleft lip and palate" is the emphasized core feature) — a midline/bilateral pattern, not unilateral.
• Digital anomalies are bilateral (bilateral camptodactyly and clinodactyly of fifth fingers).
• The laterality defect is, by definition, an asymmetry disorder: abdominal situs inversus with asplenia, plus abnormal cardiac looping in the zebrafish model with "bilateral or right-sided" southpaw expression replacing normal left-sided expression.

This combination — bilateral/midline structural defects plus a stochastic laterality defect — is characteristic and diagnostically useful.

8. Temporal Development

Onset

• Congenital (HP:0003577), 2/2. All structural anomalies are established in embryogenesis.
• Detectable prenatally: the terminated fetus (A-II-2) was identified antenatally with bilateral cleft lip, vermian hypoplasia, hypoplastic pons, and abnormal cisterna magna — establishing that CFNDS can be recognized on second-trimester ultrasound/fetal MRI in an at-risk family.
• Onset pattern: chronic/static from birth. Not acute, not subacute, not insidious.
• Sequence of clinical declaration: prenatal structural findings (if imaged) → neonatal cleft and cardiac diagnosis → infantile feeding difficulty (2/2) → toddler-age developmental delay → school-age intellectual disability, hearing, and growth concerns.

Progression

• Disease stages: No staging system exists and none is applicable — CFNDS is not a staged/progressive disease.
• Progression rate: Non-progressive. The malformations are fixed. Developmental delay evolves into a stable intellectual disability profile rather than deteriorating.
• Course pattern: Static/stable with age-dependent emergence of developmental phenotypes. No episodic, relapsing-remitting, or degenerative course has been reported in any patient.
• Duration: Lifelong, chronic. Not self-limited.

Patterns

• Remission: Not applicable — no spontaneous remission; "treatment-induced improvement" refers only to surgical correction of individual anomalies (cleft repair, cardiac repair), not disease remission.
• Critical periods:
• Weeks 4–7 post-conception — lip fusion; weeks 6–12 — secondary palate fusion. These windows are already passed at diagnosis; no prenatal intervention is possible.
• Weeks 3–4 — left-right axis determination and cardiac looping.
• Postnatal intervention windows are the actionable ones: cleft lip repair ~3–6 months, palate repair ~9–18 months (speech-outcome critical period), cardiac repair timed to lesion (AVSD typically 3–6 months), early intervention/speech therapy from infancy, and hearing surveillance from the neonatal period because unrecognized conductive loss compounds the cleft-related speech deficit.

Longitudinal data: No natural-history study, no registry, no longitudinal cohort exists. The oldest reported patient is 9 years old. Adult outcomes for CFNDS are entirely unknown. This is arguably the single largest knowledge gap.

9. Inheritance and Population

Epidemiology

• Prevalence: Not established. With ~6 living individuals reported worldwide from 5 families since 2020, the observed prevalence is far below 1 in 1,000,000.
• For structured curation: measure_type: CASES_IN_LITERATURE, prevalence_class: BELOW_1_IN_1000000 (or ULTRA_RARE), population: Worldwide. Do not assign a rate_per_100000 — no denominator-based estimate exists.
• Supporting statement: "CFNDS has only been described in four living individuals and one terminated fetus from four families" (PMID:41639596) — that count precedes the two siblings in the same report.
• Incidence: Not established. No birth-prevalence estimate exists.
• Ascertainment caveat: True prevalence is almost certainly higher than reported. Three of five families' variants are deletions, and one was missed by both SNP array and trio exome sequencing — meaning conventional diagnostic pipelines systematically under-ascertain this disease (§10).

Genetic epidemiology

Population demographics

• Reported ancestries: Arab Muslim (Israel/Palestine region), Iranian (Isfahan), Egyptian (Abdalla cohort context), Portuguese, Dutch (Rotterdam sibling pair). This is a broad, non-clustered geographic spread — no single ethnic predilection can be claimed, though ascertainment is biased toward centres with consanguineous referral populations and toward European tertiary genetics services.
• Geographic distribution: No endemic region. Cases from the Middle East, North Africa, Southern Europe, and Northwestern Europe.
• Variant geography: The two founding frameshifts are each private to one Middle Eastern family. The ~32.6-kb deletion allele appears in at least two unrelated Southern-European/North-African-context patients — the only candidate for a geographically structured allele.
• Sex ratio: Approximately balanced in a cohort far too small to estimate (reported: female, male, female, plus the 2024 patient and sibling pair). No sex bias is expected for an autosomal recessive disorder; do not curate a sex ratio.
• Age distribution of affected individuals: All reported living patients were children at the time of report (ages 3, 6, 9 years, plus the sibling pair). No adult patient has been described. This reflects the disease's recent delineation, not necessarily a survival ceiling — but it means adult phenotype and survival are unknown.

10. Diagnostics

Clinical tests

• Laboratory tests: No disease-specific laboratory test exists. There is no enzyme assay, no metabolite, no biochemical marker for CFNDS. Standard labs are non-diagnostic.
• The one lab investigation with real value is functional asplenia screening in patients with laterality defects: peripheral smear for Howell-Jolly bodies, correlated with abdominal imaging.
• Biomarkers: None. No FDA-listed or research biomarker. Neither diagnostic, prognostic, nor monitoring biomarkers exist. (Speculatively, surface transferrin receptor levels or AP-2 subunit abundance in patient fibroblasts could serve as a functional assay — this has been demonstrated in engineered cells but never applied to a patient sample. This is a concrete, actionable research gap.)
• Imaging studies (essential, high-yield):
• Brain MRI — required. Detects cerebellar vermis hypoplasia (HP:0001320), hypoplastic corpus callosum (HP:0002079), hypoplastic pons (HP:0012110), abnormal cisterna magna. This is the highest-yield single imaging study.
• Echocardiography — required in all patients. Detects AVSD, VSD, pulmonic stenosis, and cardiac looping/positional abnormality.
• Abdominal ultrasound / cross-sectional imaging — for situs and spleen presence. Given asplenia carries life-threatening infection risk, this is not optional in a patient with any laterality clue.
• Renal ultrasound — normal in reported patients, but reasonable to exclude the ciliopathy differential.
• Skeletal survey / hand radiographs — brachydactyly, clinodactyly, phalangeal anomalies, kyphosis.
• Fetal ultrasound / fetal MRI — for prenatal detection in at-risk pregnancies (proven effective: the A-II-2 fetus).
• Functional tests: Audiometry / tympanometry — mandatory and repeated, given bilateral conductive hearing loss and cleft palate. Feeding/swallow evaluation (videofluoroscopy) for the 2/2 infantile feeding difficulty.
• Electrophysiology: EEG was normal in individual B-II-1. Seizures are not a reported CFNDS feature — a useful negative that distinguishes CFNDS from the AP2M1-related developmental and epileptic encephalopathy (see differential below). ECG accompanies echocardiography.
• Biopsy / pathology: No characteristic histopathology. No diagnostic biopsy exists. Nasal brush biopsy for ciliary ultrastructure/beat frequency (the PCD workup) has not been reported in CFNDS and there is no evidence of a motile-ciliary/PCD-type respiratory phenotype — do not curate a PCD-style workup as indicated.

Genetic testing — the only definitive diagnostic modality

Recommended approach, in order of yield:

1. Trio exome sequencing (ES) with explicit CNV calling, or preferably
2. Genome sequencing (GS) — better structural-variant resolution, and
3. RNA sequencing (RNA-seq) as a complementary second-tier test when DNA testing is negative.

The single most important practical lesson from the literature is that conventional testing fails in this disease:

"This deletion was not detected in previous SNP array analyses and trio exome sequencing" — PMID:41639596 "highlighting the complementary value of RNA-seq" — PMID:41639596

• WES utility: Established — it discovered the gene ("Using whole exome sequencing, we identified homozygous frameshift CCDC32 variants," PMID:32307552). But ES alone missed a causal intragenic deletion in one family. ES must be paired with CNV analysis, and a negative ES does not exclude CFNDS.
• WGS utility: Superior for the deletion alleles that constitute the majority of reported pathogenic variation (3/5 families). Preferred first-line where available.
• RNA-seq: Proven diagnostic in CFNDS — exon 3–4 skipping revealed the biallelic deletion the DNA tests missed. This makes CFNDS a genuine exemplar of RNA-seq's complementary diagnostic value.
• Gene panels: CCDC32 should be included on orofacial clefting, syndromic congenital heart disease, heterotaxy/laterality, cerebellar hypoplasia/pontocerebellar, and intellectual disability/multiple congenital anomaly panels. Verify inclusion before relying on a panel — as a 2020 gene with Moderate ClinGen validity, panel coverage is inconsistent.
• Single-gene testing: Reasonable only for targeted familial-variant testing (carrier testing of relatives, prenatal diagnosis) once the family's variant is known. Not appropriate as a primary diagnostic given phenotypic nonspecificity.
• Chromosomal microarray (CMA)/SNP array: Documented to fail — the exon 3–4 deletion escaped SNP array. CMA remains a reasonable first-tier test for MCA/DD generally, but a normal CMA does not exclude CFNDS.
• Homozygosity mapping: High value in consanguineous families — it was the discovery route (9.06 Mb and 5.23 Mb ROH). Should be run alongside ES/GS in consanguineous pedigrees.
• Karyotyping / FISH: No role. No CFNDS case involves a visible chromosomal rearrangement.
• Mitochondrial DNA testing: Not applicable.
• Repeat expansion testing: Not applicable.

Omics-based diagnostics

• RNA sequencing: the one validated omics diagnostic (above).
• Proteomics / metabolomics / epigenomics / liquid biopsy: No established role. No episignature. No metabolomic signature.

Clinical criteria

• No standardized diagnostic criteria, no consensus guideline, no society statement exists for CFNDS. Diagnosis is molecular: biallelic pathogenic CCDC32 variants in a compatible phenotype.
• Proposed clinical gestalt prompting testing (from Abdalla's core phenotype): developmental delay + bilateral cleft lip and palate, with any of — microcephaly, cerebellar vermis hypoplasia, congenital heart defect, digital/nail anomalies, laterality anomaly, postnatal growth restriction. Consanguinity substantially raises prior probability.

Differential diagnosis

Screening

• Newborn screening: Not applicable and not feasible — no biochemical marker. CFNDS is not on any NBS panel.
• Carrier screening: CCDC32 is not on standard expanded carrier screening panels. Given the ultra-rare status, population carrier screening is not indicated. Targeted carrier testing of relatives is indicated once a family variant is known.
• Cascade screening: Appropriate for at-risk relatives in a known family (siblings of probands; extended family in consanguineous pedigrees).

11. Outcome / Prognosis

Global caveat: there is no natural-history study, no survival analysis, and no adult patient reported. Everything below is either directly observed in ≤6 children or explicitly labelled as extrapolation.

Survival and mortality

• 5-year / 10-year survival: Not established. No survival data.
• Life expectancy: Unknown. All reported living patients were alive at last report at ages 3, 6, and 9 years, plus the sibling pair. No death of a liveborn CFNDS patient has been reported.
• Mortality rate / disease-specific mortality: Not established.
• One pregnancy was electively terminated following prenatal detection of anomalies (fetus A-II-2) — this is a reproductive decision, not a measure of intrinsic lethality, and should not be curated as a mortality statistic.
• Reasoned prognostic drivers of mortality risk (extrapolated, not CFNDS-observed): severity of the congenital heart defect (AVSD carries the greatest operative burden), and overwhelming post-splenectomy-type sepsis risk in asplenic patients — the latter is a preventable cause of death and the single most important actionable prognostic factor.

Morbidity and function

• Morbidity: Substantial and multi-domain — cognitive, speech, hearing, feeding, cardiac, and surgical.
• Disability outcomes: Global developmental delay in 2/2 and frank intellectual disability in the oldest reported patient (age 9). Long-term functional independence is unknown but likely to require support. No ICF-coded outcome data exist.
• Quality of life: No measured QoL data. No EQ-5D/SF-36/PROMIS administration reported.

Disease course and complications

Observed and anticipated complications: - Cleft-related: feeding failure in infancy (observed 2/2, "severe feeding difficulties" in one), velopharyngeal insufficiency, speech disorder, recurrent otitis media, bilateral conductive hearing loss (observed), dental anomalies (missing teeth observed) - Cardiac: heart failure from unrepaired AVSD/VSD, right-ventricular consequences of pulmonic stenosis, operative and post-operative morbidity - Neurological: developmental delay → intellectual disability; cerebellar signs (ataxia/coordination) plausible from vermian hypoplasia but not explicitly reported; seizures not reported - Infectious: encapsulated-organism sepsis in the asplenic patient — the highest-acuity preventable complication - Growth: postnatal growth restriction/short stature in some, but not universal (one patient at 80th centile height, another at 97th) - Behavioral: hyperactivity reported in one patient

Recovery potential

• Structural anomalies are surgically correctable, not spontaneously recoverable. Cleft repair and cardiac repair yield good anatomical outcomes by general paediatric standards.
• Neurodevelopmental impairment is not recoverable — it is amenable to habilitation/early intervention but not to reversal.
• No disease-modifying therapy exists, so there is no "with vs without treatment" survival comparison to report.

Prediction

• Prognostic factors: No validated prognostic model, no prognostic biomarker, no clinical calculator exists. Reasoned (unvalidated) determinants: severity of CHD; presence of asplenia; degree of microcephaly and hindbrain hypoplasia; adequacy/timing of cleft and hearing management.
• Genotype–phenotype correlation: None established. All reported variants are complete LoF (frameshift or deletion) yet produce a variable phenotype — arguing that modifiers or stochastic developmental variation, not allele severity, drive expressivity. This is a well-posed research question.

12. Treatment

There is no disease-specific, disease-modifying, or targeted therapy for CFNDS. Management is entirely symptomatic, anatomical, and habilitative, delivered by a multidisciplinary craniofacial/genetics/cardiology team. No treatment guideline exists for CFNDS specifically; care follows the guidelines for each constituent anomaly.

Pharmacotherapy

• No disease-specific pharmacotherapy.
• Antibiotic prophylaxis in asplenia is the one pharmacological intervention with a strong, specific indication in the subset with asplenia/heterotaxy: penicillin prophylaxis plus a low threshold for empiric treatment of febrile illness (standard asplenia protocols).
• Peri-operative and heart-failure medications as dictated by the specific cardiac lesion (diuretics, afterload reduction) — lesion-directed, not disease-directed.
• Pharmacogenomics: No CPIC or PharmGKB guidance applies to CCDC32. No PGx relevance identified.

Suggested NCIT annotation:

- name: Antimicrobial Prophylaxis for Asplenia
  treatment_term:
    preferred_term: Pharmacotherapy
    term: {id: NCIT:C15986, label: Pharmacotherapy}
  therapeutic_modality: SMALL_MOLECULE

Advanced therapeutics

• Gene therapy: None. Not in development. No preclinical program exists. (Conceptually challenged: the pathology is established during embryogenesis and is structural — postnatal gene replacement could not correct a cleft palate or a septal defect.)
• Cell therapy: None.
• RNA-based therapies (ASO/siRNA/mRNA): None. Note that despite the exon 3–4 deletion, exon-skipping/splice-modulating ASO strategies are not applicable — the lesion is a genomic deletion producing loss of protein, not a correctable splice defect, and the therapeutic window is prenatal.
• Targeted therapy: None. No druggable node has been proposed. (The AP-2 assembly pathway is not currently a therapeutic target for any indication.)
• Immunotherapy: Not applicable.

Surgical and interventional (the principal therapeutic modality)

Suggested NCIT annotations (verified via OLS unless noted):

- name: Cleft Palate Repair
  treatment_term:
    preferred_term: palatorrhaphy
    term: {id: NCIT:C168380, label: Palatorrhaphy}
  therapeutic_modality: SURGERY

- name: Congenital Heart Defect Surgical Repair
  treatment_term:
    preferred_term: surgical procedure
    term: {id: NCIT:C15329, label: Surgical Procedure}   # safe, reachable from NCIT:C25218
  therapeutic_modality: SURGERY

⚠️ NCIT:C157806 "Cardiac Surgery" also exists but may sit outside the NCIT:C25218 (Clinical Intervention or Procedure) subtree used by the dismech TreatmentTerm dynamic enum — validate with just validate-terms before using it; NCIT:C15329 is the safe fallback.

Supportive and rehabilitative

Experimental treatments

No clinical trials exist for CFNDS. A ClinicalTrials.gov search yields no interventional or observational study recruiting CFNDS or CCDC32 patients. No NCT identifiers to report. Do not curate a clinical_trials: block for this entry.

Treatment outcomes

• Response rates: No CFNDS-specific outcome data. Outcomes of cleft repair and CHD repair in CFNDS patients have not been separately reported.
• Side effects / adverse events: No disease-specific pharmacovigilance signal (no disease-specific drug). Surgical risks are the standard risks of the respective procedures.

Treatment strategy

• Algorithm: No CFNDS-specific pathway. Practical approach: (1) confirm molecular diagnosis; (2) complete the baseline evaluation — echocardiogram, brain MRI, abdominal imaging for situs/spleen, audiology, ophthalmology, renal ultrasound, feeding assessment, developmental assessment; (3) refer to a multidisciplinary craniofacial team; (4) stage cleft and cardiac surgery per standard protocols; (5) institute early intervention and hearing surveillance; (6) if asplenic, start antibiotic prophylaxis and asplenia vaccination immediately; (7) genetic counselling for the family.
• Combination therapies: Not applicable in the pharmacological sense; management is inherently multimodal/multidisciplinary.
• Personalized/genotype-guided treatment: None available. No genotype-guided management stratification exists — and with all reported alleles being complete LoF, none is currently plausible.

13. Prevention

Prevention levels

• Primary prevention (preventing occurrence): The only effective primary prevention is reproductive — genetic counselling, carrier testing, and reproductive options in at-risk (typically consanguineous) families. There is no modifiable exposure to target. Population-level consanguinity counselling programs reduce the aggregate burden of AR disease but are not CFNDS-specific.
• Secondary prevention (early detection): Prenatal ultrasound/fetal MRI in at-risk pregnancies is proven to detect the phenotype (fetus A-II-2). Targeted prenatal molecular testing (CVS/amniocentesis for the known familial variant) is definitive. Postnatally, early molecular diagnosis enables timely baseline evaluation.
• Tertiary prevention (preventing complications in affected individuals) — this is where the highest-value, concrete actions lie:
• Asplenia protocol: lifelong antibiotic prophylaxis + immunization + fever action plan. Prevents the most likely preventable death.
• Audiology surveillance from birth: prevents hearing loss compounding speech/language deficit.
• Timely palatoplasty within the speech-critical window: prevents durable velopharyngeal speech impairment.
• Cardiac surveillance and timely repair: prevents pulmonary vascular disease and heart failure.
• Feeding/nutrition management: prevents failure to thrive.
• Early developmental intervention: optimizes attainable function.

Immunization

Highly relevant in the asplenic subset. Standard functional-asplenia immunization: pneumococcal (conjugate + polysaccharide), meningococcal (ACWY and B), and Haemophilus influenzae type b, plus annual influenza. Otherwise, routine childhood immunization per national schedule.

Suggested annotation: treatment_term: NCIT:C15346 (Vaccination), therapeutic_modality: VACCINE.

Screening and early detection

• Population screening programs: None applicable. Not newborn-screenable (no biochemical marker); too rare for population carrier screening.
• Genetic screening in families: Carrier testing of at-risk relatives; preimplantation genetic testing for monogenic disease (PGT-M) and prenatal diagnosis (CVS/amniocentesis) are both technically available once the familial variant is characterized. For families whose variant is a deletion, ensure the prenatal assay is deletion-capable (targeted MLPA/ddPCR or breakpoint-spanning PCR — a standard variant-specific Sanger assay will fail).
• Risk stratification: Prior risk is driven by consanguinity and by an affected sibling (1 in 4 recurrence).

Behavioral interventions

Not applicable to disease occurrence. Standard periconceptional care (folic acid, avoidance of teratogens) is appropriate general practice but has no established effect on CFNDS risk and should not be curated as a CFNDS protective factor.

Counseling

Genetic counselling is the cornerstone of prevention. Content should include: - Autosomal recessive inheritance; 25% recurrence risk for each pregnancy of carrier parents - Carrier parents are unaffected (supported by gnomAD constraint: pLI 0.19, LOEUF 0.76) - Extended-family carrier risk in consanguineous pedigrees; offer cascade testing - Availability of PGT-M and prenatal diagnosis - Honest communication of prognostic uncertainty — with ~6 published patients and no adult data, families must be told that long-term outcome is genuinely unknown - Consanguinity counselling for the wider kindred where culturally appropriate

NCIT: NCIT:C15240 Genetic Counseling.

Public health / environmental interventions

Not applicable. No environmental determinant. The only population-level lever is consanguinity-related genetic services and premarital/preconception carrier screening programs in high-consanguinity populations — a general AR-disease intervention, not CFNDS-specific.

Prophylaxis

• Antibiotic prophylaxis for asplenia — the one strongly indicated prophylactic medication (see above).
• Endocarditis prophylaxis per standard cardiology guidelines for the specific repaired/unrepaired cardiac lesion.

14. Other Species / Natural Disease

Taxonomy of species with relevant biology

Breed

Not applicable. No breed-associated CFNDS-equivalent disorder has been described. No VBO identifiers apply.

Orthologous genes

• Zebrafish ccdc32: a single ortholog, with 46% amino-acid identity and 64% similarity to human CCDC32 [PMC7268788, full text]. Modest conservation, but functionally sufficient — depletion recapitulates the human phenotype.
• Mouse Ccdc32: ortholog exists (used for the IMCD3 siRNA experiments, targeted via siCcdc32). MGI record exists; I could not retrieve an IMPC phenotype page (404), and no IMPC null-allele phenotype data were located.
• Human NCBI Gene ID 90416 is the anchor for ortholog lookup (Alliance of Genome Resources / HomoloGene).

Natural disease in other species

None known. No OMIA entry, no veterinary case series, no spontaneous animal disorder corresponding to CFNDS has been reported. There is no companion-animal or wildlife counterpart, and no veterinary health importance.

Comparative biology

• Evolutionary conservation of mechanism: Strong. Harel et al. concluded that "CCDC32 plays an evolutionarily conserved role in cilia formation in the vertebrate left/right organizing center" [full text]. The AP-2 adaptor complex and its assembly chaperones (AAGAB, CCDC32) are deeply conserved across eukaryotes, and the AP-2 assembly mechanism described in human cells is expected to be broadly conserved.
• Comparative pathology — concordances: zebrafish ccdc32 crispants reproduce reduced head size (≈ microcephaly), altered facial cartilage morphology (≈ craniofacial dysmorphism/clefting), cerebellar hypoplasia, disrupted cardiac looping, and laterality randomization (≈ situs inversus).
• Comparative pathology — divergences: the zebrafish model shows no global developmental delay (normal body length), and clefting per se cannot be modelled in fish (no secondary palate) — only the homologous pharyngeal cartilage patterning.

Transmission

Not applicable. CFNDS is a germline monogenic disorder. No zoonotic potential, no cross-species transmission, no infectious component.

15. Model Organisms

15.1 Zebrafish (Danio rerio, NCBITaxon:7955) — the primary and best-characterized model

Model type: vertebrate, mammalian-adjacent developmental model; F0 CRISPR crispant (mosaic knockout).

Construction: Two distinct sgRNAs (sgRNA1, sgRNA2) targeting non-overlapping regions of exon 2, injected with Cas9 protein at the one-cell stage. Editing efficiency: 85% (sgRNA1) and 70% (sgRNA2) mosaic alterations in F0 crispants [PMC7268788, full text]. Two independent guides is good practice and substantially strengthens the specificity of the result.

Expression pattern in zebrafish: ccdc32 detected as early as 1 hour post-fertilization; localized to the developing head and neural tube throughout embryogenesis, and particularly concentrated in the Kupffer's vesicle region — a spatial pattern that closely prefigures the human phenotype (head, CNS, laterality).

Phenotypes observed [all MODEL_ORGANISM evidence]:

"ccdc32 depletion recapitulates the human phenotypes" — PMID:32307552 (verbatim, abstract)

Phenotype recapitulation quality: Good for craniofacial, cerebellar, cardiac-looping, and laterality domains. This is a genuinely convergent model — four independent human phenotype domains reproduced.

Limitations: - F0 crispants are mosaic, not stable germline nulls — a stable mutant line has not been reported and is an obvious next step. - Cleft lip/palate cannot be modelled — zebrafish have no secondary palate; only homologous pharyngeal cartilage patterning is assessable. - Intellectual disability/global developmental delay cannot be modelled; body length was explicitly normal. - Situs inversus and asplenia are assessed indirectly (spaw expression, cardiac looping) rather than as terminal organ situs. - Crispant phenotypes can carry p53-dependent off-target/toxicity artefacts; the two-independent-guide design mitigates but does not eliminate this. - Modest human–fish protein identity (46%) limits inference about specific residues/motifs.

Resource: ZFIN (the ZFIN publication record ZDB-PUB-220423-8 exists for the Abdalla paper; I was unable to retrieve the ccdc32 gene record — ZFIN returned a CAPTCHA/traffic page — so the ZFIN gene ID should be looked up before curation).

15.2 Mammalian cell models

mIMCD3 5-HT6-GFP (mouse inner medullary collecting duct, ciliated reporter line) [IN_VITRO] - siRNA against Ccdc32 (5 nM) vs non-targeting control; 24 h serum starvation to induce ciliogenesis; knockdown validated by qRT-PCR with three primer sets - Result: significantly reduced % ciliated cells (P<0.01) and reduced cilium length (P<0.05 to P<0.001) - Rigor: >500 cells per replicate, 4 replicates, >1000 cilia measured, imaging blinded to condition — a well-controlled experiment

"Cilia formation is similarly impaired in ciliated mouse inner medullary collecting duct cells" [full text]

HeLa and other human cell lines (2024–2026 mechanism studies) [IN_VITRO] - CCDC32 knockout HeLa cells: loss of all four AP-2 subunits at steady state; loss of plasma-membrane AP-2 puncta; strongly reduced transferrin-receptor endocytosis with elevated surface TfR; impaired GLUT4 internalization - siRNA knockdown: accumulation of unstable flat clathrin assemblies; inhibited CCP invagination; reduced TfnR uptake - Structure–function: deletion constructs (e.g. CCDC32(1-54), mimicking the patient truncations) fail to rescue CCP stabilization; patient-mutant CCDC32 is defective in binding AP-2 α and σ2 — this is the closest thing to a direct functional validation of patient alleles - In vitro reconstitution with PIP2-containing liposomes — the membrane-switch experiments (Sci Adv 2026)

Patient-derived cells: None reported beyond the RNA-seq performed on patient material for diagnosis (PMID:41639596). No patient fibroblast, iPSC, organoid, or neuronal model of CFNDS exists. This is a major and readily addressable gap.

15.3 Mouse (Mus musculus, NCBITaxon:10090)

• No published Ccdc32 whole-animal knockout or knock-in mouse model exists. I found no IMPC phenotype data (the queried IMPC gene page returned 404) and no MGI-recorded allele phenotype for a Ccdc32 null.
• A commercially available conditional (flox) allele exists — a Ccdc32-flox C57BL/6 line (Cyagen Ccdc32em1flox) is catalogued — but no phenotype has been published from it. Curate as "resource available, phenotype unpublished," not as a characterized model.
• Highly informative surrogate models — the AP-2 subunit knockouts:
• Ap2b1 (β2) null: "mice lacking the single-copy AP-2 β subunit gene AP2B1 survive until birth but then die shortly afterwards, the only obvious abnormality being that they have a cleft palate" [JCS review, full text] — strikingly convergent with the CFNDS core phenotype. Survival is attributed to partial redundancy with the AP-1 β paralog.
• Ap2s1 (σ2) null: early embryonic lethal (~E3.5–E9.5).
• Ap2m1 (μ2) null: early embryonic lethal.
• Complete AP-2 loss: "Complete loss of AP-2 is early embryonic lethal" [JCS review, full text]
• Interpretive value: these establish a dosage/severity gradient in which CFNDS sits at the mild, viable end — consistent with CCDC32 loss producing partial rather than complete AP-2 deficiency, and directly supporting the cleft palate and cardiac phenotypes as AP-2-attributable. ⚠️ Do not conflate the AP-2 adaptor complex α subunit (AP2A1/AP2A2) with the AP-2α transcription factor (TFAP2A) — searches readily mix them, and TFAP2A knockouts also produce cleft palate and cardiovascular defects for entirely unrelated reasons.

15.4 Applications and gaps

What current models support: ciliogenesis assays; left-right patterning; craniofacial cartilage morphometry; cardiac looping; AP-2 assembly biochemistry and structural biology; CCP dynamics by TIRF; transferrin-uptake functional readouts; direct testing of patient alleles in rescue assays.

What no current model supports: the neurodevelopmental/cognitive phenotype; cleft lip and palate morphogenesis in a mammal; longitudinal/adult natural history; therapeutic testing.

Highest-value next models: (1) a stable germline zebrafish ccdc32 mutant line; (2) a constitutive or neural-crest-conditional mouse Ccdc32 knockout using the existing flox allele — with explicit assessment of palate, cardiac septation, situs, and brain; (3) patient-derived iPSC neural crest cells and cerebral/cardiac organoids; (4) patient fibroblast AP-2 abundance and transferrin-uptake assays as a candidate functional diagnostic for VUS resolution.

Model databases: ZFIN (zebrafish), MGI / IMPC / IMSR / KOMP (mouse), Alliance of Genome Resources (ortholog integration), Cellosaurus (HeLa, mIMCD3 lines).

Appendix A — Consolidated Reference List

Structured / database sources: MONDO:0030873 (OLS/EBI); MedGen UID 1721861; HPO API annotations for OMIM:619123; HGNC:28295; ClinGen gene-disease validity (Syndromic Disorders GCEP, 2024-10-18, MODERATE); ClinVar (VCV001690313, VCV002431643, VCV000988600, VCV000988601, VCV002580223); gnomAD v4.0 constraint; Human Protein Atlas ENSG00000128891.

Appendix B — Explicit "No Data" Register

For honest curation, these are confirmed absences, not unsearched areas:

Appendix C — Suggested High-Priority discussions Entries for the KB Record

1. kind: KNOWLEDGE_GAP — Adult phenotype, survival, and natural history are entirely unknown (oldest reported patient age 9).
2. kind: KNOWLEDGE_GAP — Are the two ~32.6-kb deletions (ClinVar VCV001690313 / VCV002431643) the same recurrent, repeat-mediated allele? Breakpoint and haplotype analysis needed. Proposed experiment: breakpoint-junction sequencing and SNP-haplotype comparison across reported deletion carriers.
3. kind: HUMAN_MODEL_MISMATCH — The zebrafish ccdc32 crispant supports a primary ciliary mechanism, but the 2024–2026 human-cell structural/biochemical work assigns CCDC32 a primary AP-2 assembly function, and patients lack the cardinal ciliopathy features (cystic kidney disease, polydactyly, retinal dystrophy). Whether the ciliary defect is primary or secondary to CME failure is unresolved. Proposed experiments: test ciliogenesis in CCDC32-KO human cells with AP-2 rescue; assay ciliary receptor trafficking; characterize a stable zebrafish mutant line and a neural-crest-conditional mouse KO.
4. kind: KNOWLEDGE_GAP — All reported alleles are complete LoF yet expressivity is markedly variable (hypertelorism vs hypotelorism; AVSD+heterotaxy vs isolated VSD; fetus with no cardiac defect). Modifiers, stochastic developmental variation, or both?
5. kind: KNOWLEDGE_GAP — No functional assay exists to classify CCDC32 VUS (e.g. the p.Tyr157Ter ClinVar VUS). Proposed experiment: validate patient-fibroblast AP-2 subunit abundance and transferrin-uptake as a clinical-grade functional readout.

Suggested conforms_to candidates: none of the existing dismech modules is a clean fit. pharyngeal_arch_patterning_serial_homology is a partial conceptual neighbour (cranial-neural-crest-derived multi-element craniofacial malformation) but CFNDS's lesion is a trafficking/assembly chaperone rather than an arch-patterning or ribosome/spliceosome lesion, and the CFNDS bundle is not confined to arch derivatives — do not force conformance. If a module is created, the natural one is a new "AP-2 adaptor assembly / clathrin-mediated endocytosis deficiency" module, which would already have three worked conformers across the pathway (CCDC32→CFNDS, AP2M1→DEE, AP2S1→FHH3, AAGAB→punctate PPK type 1) — a genuinely reusable conserved mechanism.

Sources: - OMIM #619123 — CARDIOFACIONEURODEVELOPMENTAL SYNDROME; CFNDS - OMIM *618941 — CCDC32 - Harel et al. 2020, Hum Mol Genet (PMID:32307552) · full text PMC7268788 · publisher - Abdalla et al. 2022, Am J Med Genet A (PMID:35451546) · Wiley - Fernandes da Rocha et al. 2024, Clin Dysmorphol (PMID:38818818) · journal - Albuainain et al. 2026, Eur J Hum Genet (PMID:41639596) - Wan et al. 2024, PNAS — AAGAB-to-CCDC32 handover (PMC11348294) - Yang et al. 2026, eLife — CCDC32 stabilizes clathrin-coated pits (PMC12768407) - Sloan et al. 2026, Sci Adv — CCDC32 collaborates with the membrane (PMC13267310) - Stepwise assembly of the AP2 endocytic clathrin adaptor complex, PNAS 2024 (PMC11420168) - Adaptor protein complexes and disease at a glance, J Cell Sci 2019 - MedGen UID 1721861 — CFNDS - HPO annotations for OMIM:619123 - MONDO:0030873 via EBI OLS4 - HGNC:28295 — CCDC32 - ClinGen curation results for CCDC32 (HGNC:28295) - ClinVar — CCDC32 variants - gnomAD v4.0 gene constraint - Human Protein Atlas — CCDC32 (ENSG00000128891) - GeneCards — CCDC32 - Cyagen Ccdc32-flox mouse model - AAGAB mutations in punctate palmoplantar keratoderma (PMC4282079) - Mutations in AP2S1 cause familial hypocalciuric hypercalcemia type 3, Nat Genet - Modeling AP2M1 developmental and epileptic encephalopathy in Drosophila, DMM