Helsmoortel-Van der Aa Syndrome (ADNP)

ADNP gene-related neurodevelopmental disorder (ADNP syndrome)

Literature and regulatory information updated: 9 October 2026.

0. Initial Reference Table

1. Introduction

Helsmoortel-Van der Aa syndrome is an autosomal dominant neurodevelopmental disorder caused by heterozygous pathogenic variants in the ADNP gene (activity-dependent neuroprotector homeobox), located at 20q13.13. Its presentation combines mild to severe intellectual disability (ID), global developmental delay with marked expressive language delay, autism spectrum disorder (ASD) or autistic traits, hypotonia, recognisable facies and a range of multisystem comorbidities including gastrointestinal and feeding problems, visual abnormalities, congenital heart defects, sleep disturbances, recurrent infections and endocrine manifestations [1, 2]. The Orphanet definition describes it as a rare neurodevelopmental disease characterised by ID, autistic traits, gastrointestinal problems, hypotonia, speech delay, behavioural and sleep problems, insensitivity to pain, epileptic seizures, structural brain abnormalities, dysmorphic features and visual problems, and classifies it as a malformation syndrome [2].

1.1. Founding paper and molecular discovery

The entity was delineated in 2014 by Céline Helsmoortel, A. T. Vulto-van Silfhout, B. P. Coe, Geert Vandeweyer and colleagues, in a study published in Nature Genetics that combined data from three exome sequencing studies (240 patients) with targeted resequencing using molecular inversion probes and high-resolution melting analysis in 2891 patients with syndromic ASD [4]. Among 5776 patients analysed, ten truncating ADNP variants were identified in ten patients with ASD, ID and dysmorphic features; in all nine for whom parents were available, the variants were de novo. The variants, all heterozygous frameshift or nonsense variants, were located in the C-terminal portion of the last exon and were absent from the 1000 Genomes Project, 1728 unaffected siblings from the Simons Simplex Collection and 192 chromosomes from Belgian controls. Enrichment of truncating variants in patients, compared with the Exome Sequencing Project and the Simons collection siblings, yielded p = 0.001852 (odds ratio 13.25), and the probability of observing eight or more de novo truncating variants by chance was 2.65 × 10⁻¹⁸ [4]. The authors estimated that ADNP variants would explain at least 0.17% of ASD cases (95% binomial confidence interval 0.083% to 0.32%), placing it among the most frequent monogenic causes of autism [1, 4]. In the same year, ADNP variants also emerged in other large autism and ID cohort studies (Coe, De Rubeis, Pescosolido and Deciphering Developmental Disorders in 2015, cited by GeneReviews), and Vandeweyer and colleagues described the functional link between the ADNP transcriptional regulator and BAF complexes (BRG1/BRM-associated factor, SWI/SNF family) [1, 14].

1.2. Subsequent milestones

The first systematic clinical description came from the international cohort of 78 individuals assembled by the ADNP consortium between January 2014 and October 2016 (Van Dijck et al., Biological Psychiatry, 2019), through a literature review, collaborators and direct contact with families, using a structured questionnaire, photographs and interviews [15]. This cohort underpins the clinical description in GeneReviews and established the frequencies of the principal comorbidities [1]. In 2019 and 2020, two peripheral blood DNA methylation episignatures were described, dependent on variant position, making HVDAS the first known Mendelian disease with two distinct methylation profiles associated with the same gene [16, 17]. In 2023 and 2024, the Antwerp group documented variants leading to loss of function through non-truncating mechanisms, such as an intragenic inversion and a splice acceptor site variant detected only by epigenetic and transcriptomic analysis [18, 19]. In 2024, overlap between the class 2 episignature and that of SMARCA2-associated blepharophimosis–ID syndrome further established a blepharophimosis phenotype within the HVDAS spectrum [20]. In December 2025, a PROSPERO-registered systematic review collected 66 new individuals from the literature, who, added to the 78 from 2019, totalled 144 with clinical data [6]. In September 2026, the first disease-specific clinical recommendations were published, developed by expert consensus at the Seaver Autism Center (Icahn School of Medicine at Mount Sinai, New York) based on 209 published cases [3]. The 2025 edition of GeneReviews, which preceded that document, stated that no published clinical guidelines existed and based its recommendations on the authors' experience [1].

1.3. Nomenclature and entities within scope

The eponym includes the surname of the first author of the founding paper, Céline Helsmoortel, who also co-authored the 2016 to 2021 editions of GeneReviews, written by the University of Antwerp group; the term “ADNP syndrome” is used synonymously by patient organisations and by Orphanet in Portuguese [1, 2]. In 2025, GeneReviews adopted the name “ADNP-Related Helsmoortel-Van der Aa Syndrome”, following the convention of naming diseases by their relationship to the gene [1, 3]. Orphanet also distinguishes ADNP-related blepharophimosis–ID syndrome (ORPHA:700160) as a separate entity, with approximately 30 reported cases or families and a prevalence below 1/1,000,000, without its own OMIM number [8]. Clinically, this is a presentation within the HVDAS spectrum, associated with variants in the nuclear localisation signal region and the class 2 episignature, and is addressed in this article as a phenotypic subtype [6, 20, 21]. No germline allelic disorders distinct from HVDAS are known [1].

2. Molecular Genetics and Pathophysiology

2.1. The ADNP gene and its protein

The ADNP gene occupies band 20q13.13 (identifiers HGNC:15766, Entrez 23394, Ensembl ENSG00000101126, UniProt Q9H2P0) [22]. The MANE Select (Matched Annotation from NCBI and EMBL-EBI) transcript is NM_001282531.3, corresponding to ENST00000621696.5 [5]. According to GenBank annotation, NM_001282531.3 has six exons and the coding sequence starts in exon 4, so only the last three exons (4 to 6) are translated; transcript NM_015339.5, used in older literature and in GeneReviews Table 7, has five exons, with the coding sequence starting in exon 3, but coding nucleotide numbering (c.) is identical in both [5, 23]. This annotation difference explains why the founding paper refers to “five exons, of which the last three are translated” and the 2025 GeneReviews to six exons [1, 4]. The last exon (exon 6 in MANE) contains most of the coding sequence and harbours almost all recurrent truncating variants [6, 23].

The 1102-amino-acid ADNP protein contains nine zinc fingers, the NAP motif (neuroprotective octapeptide NAPVSIPQ) adjacent to a bipartite nuclear localisation signal (NLS), a DNA-binding homeodomain and an HP1-binding PxVxL motif (heterochromatin protein 1) [1, 4]. ADNP was identified as a gene responsive to vasoactive intestinal peptide (VIP), a neuropeptide with a neuroprotective role that is active in embryonic development, particularly neural tube closure, and protects injured neurons by inducing glial release of survival factors [1]. ADNP is a chromatin regulator: its C-terminal region interacts with the ARID1A, SMARCA4 and SMARCC2 subunits of the BAF (SWI/SNF) chromatin-remodelling complex, binds HP1 and regulates the expression of approximately 400 genes through chromatin remodelling [3, 4, 14]. ADNP positively modulates the WNT/β-catenin pathway by stabilising β-catenin (CTNNB1), and is required for neural induction and neuronal differentiation; it also interacts with proteins involved in microtubule dynamics, a function for which the NAP motif is essential [3, 22, 24].

2.2. Disease mechanism

The disease mechanism is not fully understood and sources differ in how they formulate it [1, 3]. Most pathogenic variants are nonsense or frameshift variants in the last coding exon, predicting protein truncation; because there is no downstream exon–exon junction, mutant messenger RNA (mRNA) escapes nonsense-mediated decay (NMD), and in the founding study mutant mRNA was detected in patients' lymphoblastoid cell lines [1, 4]. In these cells, total ADNP expression was increased by 41%, attributable to the mutant allele, which the authors interpreted as loss of an autoregulatory negative feedback mechanism, and expression of the paralogue ADNP2 was also increased [4]. Because all variants eliminated at least 166 C-terminal amino acids and the lost region mediates BAF complex binding, Helsmoortel and colleagues proposed a dominant-negative effect, whereby the truncated protein would compete with the normal protein [4]. The truncated protein was not detectable in patient samples using standard methods, but its presence was traced using targeted approaches, and a toxic gain-of-function effect cannot be excluded [1, 25–27]. In contrast, a splice acceptor site variant (c.-5-1_-4del in NM_001282531.3) caused exon 4 skipping and absence of truncated protein, suggesting haploinsufficiency as the mechanism in that particular case [1, 19]. The only known intragenic inversion also causes loss of function [18]. ClinGen assigns ADNP a haploinsufficiency score of 3 (sufficient evidence) [22]. The 2026 recommendations describe the disease as resulting from haploinsufficiency, attributing this interpretation to the founding paper, which does not match the dominant-negative hypothesis proposed in that paper [3, 4]. The 2025 systematic review considers that the existence of two opposite episignatures, dependent on variant position, challenges haploinsufficiency as the primary mechanism, and that the association of microdeletions including ADNP with HVDAS rests on limited evidence [6]. Taken together, the evidence suggests that different variant classes act through different mechanisms: distal truncations producing abnormal protein, loss of function in non-coding or structural variants, and possible gain of function in some missense variants [6, 28, 29].

2.3. Cellular and tissue consequences

ADNP variants cause chromatin-remodelling defects with widespread changes in DNA methylation affecting genes involved in cytoskeletal organisation, synaptic transmission, nervous system development, calcium binding and the WNT pathway, with defective neurogenesis, and also affect the autophagy pathway [1, 30]. Changes in methylation and mitochondrial gene expression were observed in the cerebellum of an autopsy case [31]. Distinct disturbances of cytoplasm–nucleus communication and nucleocytoskeletal abnormalities have been described in cells with different variants [26, 32]. In mouse models carrying variants reproducing human variants, marked sex differences in peripheral gene expression, synaptic and tau protein pathology were observed, and, in a model with a frameshift variant, disrupted Wnt signalling with chromatin–cytoskeletal defects and autism-like behaviour in males [27, 33]. These data derive from models and do not, by themselves, demonstrate the human pathophysiology of each manifestation.

2.4. Variant spectrum and nomenclature

In the 2025 systematic review, 105 individuals with pathogenic or likely pathogenic variants had 48 unique protein variants: 11 nonsense, 28 frameshift, 7 missense, one intragenic inversion (c.-89-3923_201+2793inv) and one splice acceptor site variant [6]. According to the authors of the 2026 recommendations, ClinVar contained 111 unique pathogenic and 49 likely pathogenic variants, including 98 frameshift and 45 nonsense variants, rare splice and untranslated region variants, and eight missense variants (six likely pathogenic and two pathogenic) [3]. Sequencing detects 99% of known pathogenic variants; a heterozygous deletion encompassing the entire gene has been reported only once, by chromosomal microarray [1, 34]. The founding paper also mentioned deletions of 313 kb to 3.31 Mb encompassing ADNP and 5 to 23 other genes in databases such as DECIPHER, four of them de novo [4]. The following table normalises the most frequent variants to the MANE transcript according to Human Genome Variation Society (HGVS) nomenclature (VariantValidator validation and ClinVar consultation on 9 October 2026).

Table sources: [5–7, 19, 32, 35].

p.(Tyr719Ter) is the most frequent recurrent variant, and GeneReviews Table 7 describes it as being “in exon 4”, a designation that does not correspond to current annotation: in NM_001282531.3 codon 719 lies in exon 6, the last exon, as do the other recurrent truncating variants [1, 23]. p.(Val180GlyfsTer17), found in five unrelated individuals, constitutes an N-terminal hotspot identified in the 2025 review; VariantValidator validation confirmed termination at codon 17 of the new reading frame, rather than 81 as stated in one passage of the paper [5, 6]. Variant c.-5-1_-4del affects the splice acceptor site of exon 4 of the MANE transcript, where the coding sequence begins; GeneReviews records it as a de novo non-coding 3 bp deletion, with exon 4 skipping demonstrated by transcriptome analysis [1, 19].

Missense variants require cautious interpretation. The gene generally tolerates missense variation according to gnomAD constraint metrics, and rare missense variants are usually classified as uncertain or benign when inherited from an unaffected parent [3, 28]. The 2025 review lists p.Val619Phe (two individuals), p.Arg532Glu, p.Gln67His, p.Ser802Phe, p.Gln362Glu, p.Cys687Arg and p.Arg730Gly, whose pathogenic relevance it considers uncertain and dependent on functional validation [6]. Two functional studies were published in 2026: the de novo homeodomain variant c.2405C>T p.(Ser802Phe), in a 24-year-old Italian woman, with structural modelling suggesting greater ADNP affinity for DNA and a gain-of-function hypothesis, being only the fourth missense variant with a documented causal role; and the de novo p.(Cys687Arg) variant, in the ninth zinc finger, which altered subnuclear localisation, redistributed the normal protein, disrupted neuronal migration in in utero electroporation and increased bivalent marks at GABAergic differentiation genes in induced pluripotent stem cells, with gain-of-function-like effects in a single patient cell line [28, 29].

2.5. Inheritance, penetrance and expressivity

Inheritance is autosomal dominant and the great majority of variants arise de novo: in the 2025 review, 92 of 105 individuals (87.6%) had a confirmed de novo variant, 13 (12.4%) had unknown inheritance and none had demonstrated inheritance [6]. Penetrance is, however, below 100%. In the cohort of 78, there were two families in which the proband inherited the variant from an unaffected parent [1]. In 2026, the first Turkish series described a family with p.(Ile939SerfsTer4) in the terminal exon in which the 47-year-old father was almost asymptomatic (obesity only), the 17-year-old sister had very mild expression (no reported neurodevelopmental delay, no major anomalies, with oligohydramnios during gestation and early tooth eruption) and the 2.5-year-old proband had severe disease, confounded by stage 3 neonatal hypoxic–ischaemic encephalopathy [7]. The authors concluded that penetrance was reduced and intrafamilial expressivity variable, and recommend testing parents even when apparently asymptomatic [7]. Orphanet records inheritance as unknown, but OMIM and GeneReviews classify it as autosomal dominant [1, 2]. Postzygotic mosaicism in the proband is possible, and parental mosaicism had not been reported up to 2025 [1, 36].

2.6. Episignatures and genotype–phenotype correlations

Variant position determines two peripheral blood DNA methylation profiles. Class 1 (epi-ADNP-1), with approximately 6000 predominantly hypomethylated CpG dinucleotides, is associated with variants outside the c.2000–c.2340 interval, i.e. in the N-terminal or distal C-terminal portions; class 2 (epi-ADNP-2), with approximately 1000 predominantly hypermethylated CpGs, is associated with variants within that interval, encompassing the NLS and proximal portion of the homeodomain (approximately codons 667 to 780) [1, 16, 22]. In the original series of 22 patients, the classifier separated the two profiles with complete sensitivity and specificity, reclassified variants of uncertain significance and identified three patients among 1150 individuals with undiagnosed developmental delay [16, 22]. Independent replication in 24 patients confirmed the two classes but found limited phenotypic differences between them, without evidence that more methylation changes mean more severe disease, and advised against using the signatures for clinical stratification; in a prospective cohort of 32 individuals, only greater delay in walking distinguished class 2 [3, 17].

Correlation data are modest and sometimes contradictory. Using PhenoScore, which combines facial analysis and Human Phenotype Ontology (HPO) terms, recurrent infections, gastrointestinal problems (reflux, constipation, feeding difficulties) and short stature were two to three times more frequent in class 2, whereas neurodevelopmental problems were more frequent in class 1 according to GeneReviews [1, 37]. The 2025 systematic review, by contrast, associates class 2 with more frequent and more severe ASD and more marked motor delay [6]. In the dizygotic twin case, an N-terminal variant with a class 1 episignature coexisted with one of the most severe phenotypes in the series, challenging the idea that class 1 corresponds to milder forms [32]. In the cohort of 78, carriers of p.(Tyr719Ter) acquired walking later and had a higher pain threshold than the others; the paper's abstract describes them as more severely affected [1, 15]. p.(Tyr719Ter) and neighbouring variants are associated with blepharophimosis, and variants distal to the NLS, such as p.(Arg730Ter), with ptosis and oromotor dysfunction [6, 21]. Individuals with blepharophimosis share an episignature with SMARCA2-associated blepharophimosis–ID syndrome: in a 2024 study, 12 individuals with class 2 truncating variants and 15 with SMARCA2 variants had a common profile, termed “BIS-HVDAS” [6, 20]. A 2026 case with hearing loss and cochlear nerve deficiency led the authors to propose an audiological genotype correlation, still unconfirmed in series [38]. The conclusion of the 2026 recommendations is that known genotype–phenotype correlations do not justify different medical recommendations by genotype [3].

2.7. Allelic disorders and modifiers

No germline allelic disorders other than HVDAS have been described [1]. Microdeletions of 20q13.13–q13.2 encompassing ADNP together with other developmental genes, such as SALL4 and KCNB1, produce combined phenotypes and should not be equated with classical HVDAS [6]. No validated genetic modifiers have been described; the sex differences observed in mouse models do not yet have a demonstrated clinical counterpart [27, 39].

3. Clinical Spectrum and Manifestations by System

3.1. Data sources and what they measure

Known natural history comes from three partially overlapping datasets. The first is the international cohort of 78 individuals (2014–2016), on which GeneReviews is based, with percentages presented without item-specific denominators [1, 15]. The second is the Antwerp systematic review, which assembled 66 new individuals from 26 publications (15 countries; 29 male and 37 female; median age at assessment 9 years, range 8 months to 27 years) and added them to the 78 from 2019, presenting each trait with a numerator and denominator, the latter being the number of individuals in whom the trait was assessed or reported [6]. The third is the descriptive review underpinning the 2026 recommendations, with 209 cases from 28 publications up to July 2026, without quality or risk-of-bias assessment [3]. All these data describe children and adolescents, most already receiving medical care and therapies, and are subject to publication bias towards more severe cases; absence of a trait from a report does not mean that it was absent, and longitudinal follow-up is scarce [6]. There are no untreated patient cohorts that allow natural history to be rigorously distinguished from treated disease.

Table sources: cohort of 78 [1]; 2025 review [6]; 2026 review [3]. The frequencies in the two reviews are not independent, because both include many of the same patients. The difference in ASD (93% versus 75%) may result from different recording criteria (“autistic behaviour” in 2025, “ASD or autistic traits” in 2026) and heterogeneous assessment in the reports; the 73% short stature figure in 2026 rests on a denominator of 11 and contrasts with 27% in 99 individuals in the 2025 review [3, 6].

3.2. Development and cognitive profile

All individuals described have developmental delay and almost all have ID; in the cohort of 78, ID was severe in 52%, moderate in 36% and mild in 12% [1]. Infantile hypotonia is frequent and underlies motor delay: in the cohort of 78, mean age at unsupported sitting was 13 months and at walking 2.5 years, and in the 2025 review (n = 58 and n = 64) the means were 11.5 and 27 months, with 76 of 88 individuals (86.4%) acquiring independent walking [1, 6]. Expressive language is the most affected domain, ranging from no words to phrase use; first words emerged at a mean of 35.5 months (n = 49), approximately 29% of individuals had no speech, and in the new 2025 cohort 11 of 14 (mean age 8 years and 7 months) were non-verbal [1, 6]. Receptive language is relatively preserved compared with expressive language, and many use gestures or sign language; speech apraxia and oromotor dysfunction contribute to the deficit [3, 6]. In psychometric studies cited by the 2025 review, mean verbal quotient was 32.64 and non-verbal quotient 31.18, with global deficits on the Differential Ability Scales-II (n = 21) and Vineland-III (n = 4) [6]. Longitudinal data from nine individuals suggest slowing, plateauing or declining skills in adolescence and adulthood, and regression was reported in approximately one fifth of individuals assessed [3, 6]. Toilet training is delayed in more than 80% [1, 6].

3.3. Behaviour, sensory processing and sleep

ASD or autistic traits are characteristic, with stereotypies, impaired social interaction, intense sensory interests, licking behaviours and prolonged visual inspection of objects [1, 6]. ADHD, anxiety, obsessive–compulsive behaviours, mood disorders, aggression and tantrums are frequent, and self-injury may be severe: a 2026 case describes a non-verbal 10-year-old girl with ocular rupture due to self-injury [1, 3, 40]. The high pain threshold, present in approximately two thirds, may delay recognition of fractures, otitis or abdominal pain, and sensory processing abnormalities affect approximately 67% [6]. Sleep disturbances, with difficulty falling asleep, night waking and an irregular sleep–wake rhythm, affect approximately 70% [3, 6]. A 2026 study of 248 individuals with neurogenetic diseases, including patients with HVDAS in a number not specified in the abstract, identified five sleep profiles (low symptomatology, morning sleepiness, insomnia, high symptomatology and bedtime resistance), with the insomnia profile associated with more self-injury [41]. Many individuals described received antipsychotics or stimulants for behavioural problems [6].

3.4. Neurological manifestations and neuroimaging

Epileptic seizures affect 15 to 22% of individuals, without consistent semiology: absences, focal impaired-awareness seizures, epilepsy with continuous spike-and-wave during slow-wave sleep and unclassified seizures [1, 3]. Hypertonia occurs in a minority (6.5%), and gait is frequently atypical: ataxic gait was described in all seven reports in the 2025 review that assessed it, and a kinematic study of 12 individuals showed reduced speed, shorter steps and a wide base [6]. One individual had severe axonal motor polyneuropathy in all four limbs [6]. Brain magnetic resonance imaging (MRI) is abnormal in approximately half of those studied (63/131), without a specific pattern [3]. GeneReviews describes underdevelopment of the frontal lobes with a simplified gyral pattern and occasional hypoplasia of the olfactory bulb and optic chiasm, a thin or short corpus callosum, hypoplasia of the inferior vermis, abnormal and frequently asymmetric opercularisation of the Sylvian fissure, dilation of the lateral ventricles, particularly frontally, and enlarged perivascular spaces [1]. In pooled series, ventriculomegaly (33.8%), corpus callosum hypoplasia (19.3%), cerebral atrophy (21.4%), white matter abnormalities (15.2%), delayed myelination (14.5%) and cortical dysplasia (3.4%) are the most frequently reported abnormalities, generally mild and non-specific, and therefore do not suggest the diagnosis by themselves [1, 6]. Transient visual evoked potentials in 12 children with HVDAS, compared with 46 autistic children without a genetic cause and 19 typically developing children, showed lower amplitudes and longer latencies of P60 and N75 waves and altered coherence, proposed as a biomarker for trials [42].

3.5. Facies, oral cavity and teeth

Characteristic facies were recognised in 97% of the cohort of 78: prominent forehead, high anterior hairline, ptosis, downslanting palpebral fissures, prominent eyelashes, a broad depressed nasal root, a short nose with a full upturned tip, long philtrum, thin upper vermilion, widely spaced teeth, pointed chin and small, low-set, prominent or cup-shaped ears [1]. GeneReviews Figure 1, reproduced from Van Dijck et al. (2019), shows these traits in frontal and profile photographs (https://www.ncbi.nlm.nih.gov/books/NBK355518/) [1]. In the 2025 review, the most frequent traits were thin upper vermilion (72.4%), prominent forehead (65.4%), broad or depressed nasal bridge (55.9%), bulbous or upturned nasal tip (55.3%), high hairline (54.9%), external ear anomalies (53.9%) and downslanting palpebral fissures (50%); oral findings seldom reported previously were also described, such as a tented upper lip, microstomia, retrognathia, arched palate, dental crowding and one case of suspected molar–incisor hypomineralisation [6]. Early tooth eruption is a distinguishing sign: in a 2017 parent survey, 44 of 54 children had almost complete primary dentition, including molars, in the first year of life, and in the 2025 review the sign was recorded in 44 of 73 individuals (60.3%) [6, 43]. The 2017 finding depends on parental report and is subject to recall bias [43].

3.6. Feeding and digestive system

Feeding difficulties begin in the neonatal period, with weak sucking and chewing, dysphagia, aspiration risk, gastro-oesophageal reflux, frequent vomiting or rumination, constipation and lack of satiety; some patients require gastrostomy [1, 3]. In the 2025 review, reflux affected 62%, constipation 51.2%, oromotor dysfunction 46%, hyperphagia 43.1%, dysphagia for liquids 32.8%, vomiting 31.8%, aspiration 23% and gastrostomy 13.4% [6]. The authors of the 2026 recommendations attribute digestive dysfunction to pelvic floor and smooth muscle hypotonia, colonic dysmotility, altered rectal and visceral sensitivity and autonomic dysfunction [3]. Obesity, particularly truncal obesity, was recorded in 8% of the cohort of 78 and in 32% of the 99 individuals with data in the 2026 review, a difference that may relate to the ages of included individuals and different definitions (truncal obesity versus body mass index above 30) [1, 3].

3.7. Vision

More than half of patients have visual problems, particularly hypermetropia and strabismus; cerebral visual impairment, documented in 41% of the cohort of 78 and in 45% of the 40 individuals with detailed assessment, is frequently underdiagnosed [1, 6]. Ptosis, myopia, astigmatism, nystagmus, coloboma, congenital cataracts, ectropion and everted or notched eyelids are also described [1, 6].

3.8. Ear, hearing and airway

Otological problems affected 32% of the cohort of 78; among these, 88% had narrow auditory canals, 86% frequent otitis media and 73% ventilation tubes, and hearing loss, present in 12%, was identified in early childhood [1]. In the 2026 review, hearing loss was recorded in 19% [3]. In 2026, the first case of unilateral hearing loss with ipsilateral cochlear nerve deficiency was described, in a child with p.(Asp877ValfsTer36) [38]. Obstructive sleep apnoea affects 11 to 13% of those assessed and warrants airway assessment [3, 6].

3.9. Heart

Approximately 38% of the cohort of 78 and 49 of 108 individuals with cardiac data in the 2025 review had structural congenital heart disease, with no reported acquired heart disease [1, 6]. Atrial septal defect is the most frequent (16 to 22%), followed by ventricular septal defect, patent ductus arteriosus, patent foramen ovale and mitral valve prolapse; tetralogy of Fallot and right aortic arch have been described in isolated cases [1, 3, 6].

3.10. Growth and endocrine system

Birth weight, length and head circumference are usually normal (mean weight 3061 g in 63 individuals; mean head circumference 34.6 cm) [1, 6]. Short stature (below −2 standard deviations) affects 23 to 27% of individuals aged 2 to 23 years, and growth hormone deficiency has been documented in 11 to 12% of those assessed [1, 6]. Thyroid abnormalities, particularly hypothyroidism, affect approximately 15 to 19%, and precocious puberty, with early pubic hair or menarche, has been described in 24 to 33% of those assessed, with small denominators [1, 3, 6]. A 24-year-old woman with a missense variant had hirsutism, hyperandrogenism and secondary amenorrhoea, and one case in the 2025 review had polycystic ovaries [6, 28].

3.11. Urinary and genital systems

Cryptorchidism affects approximately 42% of boys assessed, and small external genitalia 11% [6]. Congenital anomalies of the kidney and urinary tract (solitary kidney, bilateral vesicoureteric reflux, narrow ureters, duplicated collecting system) were recorded in 13 to 17% [1, 3, 32]. Recurrent urinary tract infections, enuresis and incontinence are frequent functional problems [1, 3].

3.12. Musculoskeletal system, hands and feet

Joint hypermobility affects approximately 38 to 39%; scoliosis, hip dysplasia or dislocation and Perthes disease, thoracic deformities (22% in the cohort of 78, including pectus excavatum, pectus carinatum or narrow chest) and cranial deformities (14%: plagiocephaly, trigonocephaly, brachycephaly) are described, with metopic craniosynostosis in two individuals [1, 3]. Hand and foot anomalies, present in 62% of the cohort of 78, include clinodactyly or short fifth fingers, polydactyly, persistent fetal finger pads, prominent interphalangeal joints and distal phalanges, a single transverse palmar crease, nail anomalies, toe malformations, flat feet and increased spacing between the first and second toes (sandal gap) [1, 6].

3.13. Infections, immunity and skin

Approximately half of patients have recurrent infections, mainly of the upper respiratory and urinary tracts; isolated cases of distal fibular and calcaneal osteomyelitis have been described [1, 6]. No underlying immune defect has been defined. In the skin, one case had histologically confirmed dermal thinning, periarticular hyperkeratosis and slow healing, with reduced keratinocyte proliferation and premature differentiation [6]. Prolonged retention of the umbilical stump with infection was reported in two unrelated probands in the Turkish series [7].

3.14. Blepharophimosis phenotype

A subgroup of patients, generally with truncating variants in the p.(Tyr719Ter) region and a class 2 episignature, presents with blepharophimosis and ID, a presentation that Orphanet distinguishes as ORPHA:700160 [8, 20, 21]. This phenotype overlaps with SMARCA2-associated blepharophimosis–ID syndrome, with which it shares an episignature, and enters the differential diagnosis of blepharophimosis–ptosis–epicanthus inversus syndrome: Smith's Recognizable Patterns of Human Malformation recommends investigating KAT6B and ADNP in individuals with blepharophimosis and ID without rearrangements involving FOXL2 [20, 44].

3.15. Prenatal period and adulthood

Prenatal abnormalities are rare and non-specific, and most diagnoses are made after birth [1]. Approximately 16% of individuals were born prematurely, a proportion similar to that in the general population [6]. Intrauterine growth restriction with proportional microcephaly, abnormal skull shape and white matter abnormalities have been described in a fetus diagnosed through genomic testing, as have skeletal malformations, congenital diaphragmatic hernia, ventriculomegaly and heart disease in other fetuses [6, 45]. Information on adults is scarce: the oldest individual known to the GeneReviews authors was 69 years old, the 24-year-old Italian woman with p.(Ser802Phe) continued to have ID, visual impairment, severely affected speech, truncal obesity, fragmented sleep and aggression, and the carrier father in the Turkish family was almost asymptomatic at 47 years [1, 7, 28].

4. Clinical Recognition and Warning Signs

There are no consensus clinical criteria, and HVDAS should be considered in any child with global developmental delay or ID, particularly when associated with ASD, hypotonia and multisystem comorbidities [1]. Findings that increase suspicion vary with age. In infancy, hypotonia, sucking and swallowing difficulties, reflux, congenital heart disease (particularly atrial septal defect), strabismus or lack of visual fixation due to cerebral visual impairment, cryptorchidism and, especially, early eruption of several primary teeth, including molars, before the first birthday are useful clues [1, 6, 43]. In the young child, marked expressive language delay with relatively better comprehension, late and atypical walking, autistic traits with strong sensory interests, apparent insensitivity to pain, sleep disturbance, constipation and recurrent respiratory or urinary infections complete the picture [1, 6]. The combination of a prominent forehead with high hairline, broad nasal root, short nose with an upturned tip, long philtrum, thin upper lip and widely spaced teeth is valuable when present, but the facial features are subtle in many patients [1]. Blepharophimosis with ID without a FOXL2 abnormality should prompt consideration of ADNP [8, 44]. At school age and in adolescence, regression, precocious puberty, short stature and obesity may prompt consultation [3, 6]. Family history is usually negative (an isolated case), although transmission from a mildly affected parent has been described [1, 7].

Warning signs requiring targeted assessment rather than routine follow-up alone are coughing or choking during meals, recurrent pneumonia or poor weight gain, suggesting aspiration; paroxysmal episodes or loss of skills, requiring an overnight or prolonged electroencephalogram (EEG); focal neurological signs, abnormal head circumference growth or rapid neurological deterioration, warranting MRI; abrupt changes in behaviour or sleep without an apparent cause, which may reflect pain (reflux, constipation, otitis, fracture) in a patient with a high pain threshold and limited communication; fever without a focus, requiring exclusion of urinary tract infection; and self-injury with a risk of severe injury, requiring urgent behavioural and psychiatric intervention [1, 3, 40]. There is no newborn screening for HVDAS because there is no biochemical marker [22].

5. Diagnosis: Criteria, Tests and Algorithms

5.1. Criteria

Diagnosis is molecular and is established in a proband with suggestive findings and a heterozygous pathogenic or likely pathogenic variant in ADNP, classified according to American College of Medical Genetics and Genomics and Association for Molecular Pathology (ACMG/AMP) standards [1, 46]. A variant of uncertain significance (VUS) neither establishes nor excludes the diagnosis [1]. There is no confirmatory biochemical, enzymatic or imaging biomarker [22].

5.2. Choice of tests

Because the phenotype is non-specific, most patients are diagnosed through comprehensive genomic testing. The 2021 ACMG guideline, based on evidence graded using GRADE methodology, strongly recommends considering exome or genome sequencing as a first- or second-line test in paediatric patients with congenital anomalies arising before one year of age or developmental delay or ID arising before 18 years, emphasising the higher yield and possible cost-effectiveness when ordered early [9]. The 2025 American Academy of Pediatrics (AAP) clinical report proposes phenotype-guided testing when a specific syndrome is suggested and, in other cases, an agnostic approach chosen according to diagnostic yield, complexity and impact on management, without replacing neurological, developmental paediatric or clinical genetics assessment [10]. GeneReviews accepts chromosomal microarray or a multigene ID and ASD panel as alternatives, and reserves single-gene testing (ADNP sequencing followed, if negative, by targeted deletion and duplication analysis) for the rare cases in which findings are sufficiently distinctive [1]. If exome sequencing does not identify the variant, genome sequencing is recommended; this identified the only known intragenic inversion after negative exome and microarray results [1, 18, 47].

Table sources: [1, 18, 19, 34]. The 99% figure refers to known pathogenic variants detected by sequencing, rather than the clinical sensitivity of testing for the syndrome, which depends on laboratory coverage, variant class and the existence of abnormalities not yet detectable [1]. The genes included in panels and their sensitivity vary between laboratories and over time [1].

5.3. Episignature and complementary tools

Episignature analysis of peripheral blood DNA, including stored DNA, can clarify the diagnosis in people with suggestive clinical findings and a VUS in ADNP or non-diagnostic molecular testing [1]. It helps reclassify VUS and was decisive in the c.-5-1_-4del case, in which exome sequencing did not detect the non-coding deletion and the episignature guided transcriptome analysis, which demonstrated exon 4 skipping [1, 19]. It nevertheless has limitations: the class 2 episignature is shared with SMARCA2-associated blepharophimosis–ID syndrome and is therefore not absolutely gene-specific; a negative episignature does not exclude variant pathogenicity; and it must be interpreted alongside sequence and phenotype, never as a replacement [6, 20, 22]. PhenoScore, which combines facial photograph analysis with HPO terms, supported the diagnosis in the splice case and quantifies phenotypic similarity to the syndrome (0.92 in the twin case described in section 10, versus 0.11 in the unaffected brother) [19, 32, 37]. The Antwerp group is available to review VUS in ADNP [1].

5.4. Practical algorithm

For a child with global developmental delay, ID or syndromic ASD, exome sequencing (preferably as a trio) or genome sequencing is requested as a first- or second-line test [9, 10]. If a truncating variant in the last exon of ADNP is identified, absent from population databases and de novo, the diagnosis is established; parents are always tested, even if apparently unaffected, because confirmation of de novo status strengthens classification and incomplete penetrance occurs [3, 7]. If a VUS is identified (for example, a missense variant), segregation studies, episignature analysis and, where available, functional assessment are used [1, 6]. If exome sequencing is negative and clinical suspicion remains (early tooth eruption, typical facies and comorbidities), genome sequencing, episignature analysis and RNA sequencing (RNA-seq) are considered, as these may reveal structural or splice variants [1, 18, 19]. A negative result does not exclude the disease.

5.5. Report content and classification

The report should state the gene, transcript and version (NM_001282531.3, MANE Select), the variant in HGVS nomenclature at coding DNA and protein levels, genomic coordinates with reference genome version, zygosity, ACMG/AMP classification with the criteria applied and, when known, de novo or inherited status [5, 46]. Because older literature uses NM_015339 and non-standardised notations (e.g. “c.2156dupA” for codon 719 variants), correspondence with the MANE transcript should be checked before comparing a variant with the literature or ClinVar [5, 6]. Location within the class 2 region (c.2000–c.2340) can be recorded, without implying a different clinical recommendation [3].

5.6. Prenatal diagnosis

Prenatal diagnosis is rare because ultrasound findings are inconsistent and non-specific [1, 6]. When fetal exome sequencing identifies an ADNP variant, the prenatal phenotype does not allow reliable variant classification, and growth restriction with proportional microcephaly described in one fetus is not typical of postnatal disease [6, 45]. In a pregnancy terminated at 14 weeks and 3 days for ventriculomegaly, abnormal choroid plexuses, bilateral absence of thumbs, club hand and atrioventricular septal defect, the cause was a de novo 20q13.13–q13.2 deletion encompassing 38 genes, including SALL4, KCNB1 and ADNP, with a probably combined phenotype [6]. Targeted prenatal testing and preimplantation genetic testing are possible when the familial variant is known (section 8) [1].

6. Differential Diagnosis

The differential diagnosis of HVDAS encompasses virtually all causes of syndromic ID with ASD and in practice is resolved by genomic testing. The closest differentials are SWI/SNF-related intellectual disability disorders (SSRIDD), or BAFopathies, because ADNP interacts with BAF complex subunits and manifestations overlap: ID, speech and motor delay and distinctive facies [1, 48, 49]. Coffin–Siris syndrome is usually distinguished by coarser facial features, sparse scalp hair or hypertrichosis [1, 49]. Nicolaides–Baraitser syndrome, associated with SMARCA2, deserves particular attention because a subgroup of patients with class 2 ADNP variants shares both the episignature and blepharophimosis phenotype with SMARCA2-associated blepharophimosis–ID syndrome [1, 20]. In blepharophimosis with ID, blepharophimosis–ptosis–epicanthus inversus syndrome (FOXL2) and KAT6B-related disorders are classical differentials, and Smith's Recognizable Patterns of Human Malformation recommends investigating KAT6B and ADNP when there is no FOXL2 rearrangement [21, 44]. Microdeletions of 20q13.13–q13.2 encompassing ADNP and contiguous genes, such as SALL4 and KCNB1, produce combined presentations; a fetus with a 38-gene deletion had bilateral absence of thumbs and club hand, abnormalities that do not belong to HVDAS [4, 6]. GeneReviews also refers to the OMIM phenotypic series for autosomal dominant (AD), autosomal recessive (AR) and X-linked ID, both syndromic and non-syndromic [1].

Table sources: [1, 6, 20, 34, 44, 49].

7. Treatments and Risk Management Strategies

7.1. Principles and sources of recommendations

There is no authorised targeted treatment for HVDAS, and the approach is symptomatic, multidisciplinary and coordinated by a designated lead physician, following the American Academy of Pediatrics medical home model [1, 3]. Three documents guide management. GeneReviews (2025) presents recommendations based on the Antwerp authors' personal experience, as no published guidelines existed at that time [1]. The Antwerp systematic review (December 2025) adds clinical recommendations without evidence grading [6]. The Seaver Autism Center recommendations (September 2026) are the first dedicated to HVDAS and result from a descriptive review of 209 cases and consensus among specialists in genetics, psychiatry, psychology, neurology, gastroenterology, cardiology, endocrinology and nephrology, with gynaecology and primary care adapted from general standards and review by a parent representative; there is no formal grading of recommendation strength or evidence level, and the authors declare conflicts of interest (the parent representative chairs the ADNP Kids Research Foundation, and the senior author consults for several pharmaceutical companies) [3]. No European scientific society has published its own recommendations.

7.2. Initial assessment after diagnosis

7.3. Development, communication and education

All patients should be referred at diagnosis for developmental assessment and early intervention therapies (speech and language therapy, occupational therapy, physiotherapy and behavioural intervention) [3]. The 2026 recommendations emphasise that therapies should be frequent and intensive, that weekly sessions are typically insufficient, that provision should continue for 12 months without a summer break, and that lack of progress or regression should prompt reassessment of the type, frequency or service provider rather than discontinuation of intervention [3]. Applied behaviour analysis (ABA), delivered by a certified behaviour analyst, and structured teaching with visual schedules and social stories are the recommended interventions for ASD; in severe cases, a behavioural specialist should assess the need for one-to-one support [1, 3]. For language, therapy should start early and be frequent (the recommendations give the example of 30-minute sessions five days a week), include oromotor assessment (chewing, swallowing, speech apraxia) and use PROMPT when there are oromotor difficulties [3]. Augmentative and alternative communication (AAC), from picture exchange to speech-generating devices, should be introduced when the child has few or no words; it does not hinder speech but supports it, and should be coordinated with the vision specialist when cerebral visual impairment is present [1, 3]. Few patients achieve full independence in activities of daily living, and training in hygiene, feeding, dressing and toileting should be part of the plan from school age [3]. GeneReviews describes the United States of America (USA) framework: early intervention from 0 to 3 years, developmental preschool from 3 to 5 years, an individualised education plan reviewed annually, a 504 plan, and transition planning from adolescence, with school services up to 21 years; these frameworks vary between countries [1].

7.4. Behaviour and mental health

Behavioural interventions are first line for aggression and self-injury, using ABA, systematic data collection and plan adjustment [3]. There are no studies of drug efficacy in HVDAS beyond the ketamine trial described below, and the 2026 recommendations advise adapting approaches used in other neurodevelopmental disorders, bearing in mind that these patients may be more sensitive to standard doses and more vulnerable to paradoxical effects, requiring close adverse-effect monitoring [3]. The table summarises the classes mentioned in the sources. No source consulted gives HVDAS-specific doses; dosing follows each drug's summary of product characteristics (SmPC) and child psychiatry practice, and all these uses are off-label with respect to HVDAS.

The naltrexone-treated case was a non-verbal 10-year-old girl with ASD and chronic self-injury causing ocular rupture, in whom several antipsychotics and benzodiazepines had provided limited benefit [40]. This is very low-level evidence and does not support a recommendation. For severe aggression or destructive behaviour, GeneReviews recommends referral to child psychiatry [1].

7.5. Sleep

Sleep hygiene is the first step [3]. When insufficient, the 2026 recommendations list melatonin, clonidine, trazodone, doxepin and mirtazapine as drugs to consider, without doses or HVDAS studies; use of these drugs for paediatric insomnia associated with neurodevelopmental disorders is generally off-label [3]. Suspected obstructive sleep apnoea, present in 11 to 13% of patients, warrants airway assessment by otorhinolaryngology [3, 6]. Sleep did not improve in the ketamine trial [11].

7.6. Epilepsy

Seizures are treated with antiepileptics selected by an experienced neurologist according to seizure type; many drugs may be effective and none has demonstrated disease-specific efficacy [1]. EEG is performed only when seizures are suspected, and an overnight or 24-hour recording is warranted for regression or loss of skills, given the possibility of epilepsy with continuous spike-and-wave during slow-wave sleep [1, 3]. Educating caregivers about seizures and their recognition is part of management [1]. In girls and women, carbamazepine and topiramate may reduce oral contraceptive efficacy [3].

7.7. Motor function, tone and musculoskeletal system

Physiotherapy aims to maximise mobility and reduce the risk of late orthopaedic complications, such as contractures, scoliosis and hip dislocation, and includes assistive devices (wheelchair, walking frame, bath chair, orthoses, adapted pushchairs); occupational therapy addresses fine motor skills and activities of daily living [1, 3]. For hypertonia or dystonia, GeneReviews mentions specialist consultation to consider baclofen, tizanidine, botulinum toxin (Botox®), antiparkinsonian drugs or orthopaedic surgery, without giving doses [1]. Cranial deformities, scoliosis, thoracic deformities and hip abnormalities receive standard treatment from a craniofacial surgeon or orthopaedic surgeon [1].

7.8. Feeding, nutrition and digestive system

Oromotor dysfunction should be assessed at every visit, and choking, poor weight gain, frequent respiratory infections or food refusal warrant a low threshold for clinical feeding assessment or a radiological swallow study [1]. Thickening or cooling food may help; in severe cases, a nasogastric tube or gastrostomy ensures nutrition, and some patients require gastrojejunostomy; with jejunal feeding, copper and vitamin B12 deficiencies should be monitored [1, 3]. The 2026 recommendations propose early assessment by a gastroenterologist experienced in neurodevelopmental disorders, neurogastroenterology or motility studies (gastric emptying, antroduodenal manometry) when gastroparesis or dysmotility is suspected, and digestive assessment for growth delay, feeding problems, abdominal pain or unexplained behavioural or sleep changes [3]. Constipation is prevented and treated with osmotic or stimulant laxatives and soluble fibre, reflux according to usual standards, and rumination requires intensive behavioural intervention [1, 3]. Obesity and hyperphagia warrant dietetic follow-up and metabolic monitoring [3].

7.9. Vision, hearing, heart, endocrine system, kidney and gynaecology

Visual abnormalities and strabismus receive standard ophthalmological treatment; cerebral visual impairment benefits from early intervention (education, habilitation, occupational therapy and low-vision consultation) to stimulate visual development, and school vision services should join the education team [1]. Hearing loss is treated according to usual standards; it may be conductive or sensorineural and fluctuate or progress, warranting reassessment [1, 6]. Congenital heart disease is treated by paediatric cardiology; in the twin case, the atrial septal defect and a right aortic arch with a vascular ring were surgically corrected [1, 32]. Hypothyroidism, precocious puberty and obesity receive standard endocrine treatment, and confirmed growth hormone deficiency can be treated with growth hormone by an endocrinologist; the sources do not describe HVDAS-specific outcomes of this treatment [1, 6]. Orchidopexy is indicated for cryptorchidism, and renal abnormalities or recurrent urinary tract infections warrant nephrology or urology assessment; for fever without a focus, a catheter or midstream urine sample should be obtained, and enuresis is treated with healthy voiding habits and correction of constipation [1, 3]. Gynaecological recommendations adapt the 2025 American College of Obstetricians and Gynecologists (ACOG) standards for people with disabilities: cervical cancer screening every five years between 25 and 65 years, human papillomavirus vaccination in adolescence, exclusion of thyroid disease, polycystic ovary syndrome and hyperprolactinaemia (for example, due to antipsychotics) in menstrual irregularity, and consideration of menstrual suppression or hormonal contraception when menstruation causes behavioural or hygiene problems [3].

7.10. Recurrent infections

Recurrent infections are treated according to usual standards, with referral to infectious diseases or immunology [1]. The sources describe neither a standardised immunological work-up nor specific prophylaxis.

7.11. Low-dose ketamine

Ketamine was proposed as a therapy for HVDAS because rodent studies indicated that it induces ADNP expression, preclinical evidence identified with the mediKanren biomedical reasoning system [11]. GeneReviews describes it as an “NMDA receptor agonist with pro-glutamatergic activity”, but the Food and Drug Administration (FDA) label for Ketalar® defines racemic ketamine as a non-selective, non-competitive N-methyl-D-aspartate (NMDA) receptor antagonist; the correct description is therefore antagonist [1, 51]. The only clinical trial with results in HVDAS is NCT04388774, open-label, single-arm and phase 1/2 in the registry (GeneReviews calls it phase II), sponsored by Alexander Kolevzon at the Icahn School of Medicine at Mount Sinai and partly funded by the ADNP Kids Research Foundation [1, 52, 53]. Between August 2020 and May 2021, ten children aged 5 to 12 years with a molecular diagnosis were enrolled (3 girls and 7 boys; mean age 9.5 ± 2.3 years; mean total developmental quotient 28.81 ± 15.42; ASD in 4 and ADHD in 7; frameshift variants in 6 and nonsense variants in 4) [11]. Each child received a single dose of racemic ketamine hydrochloride 0.5 mg/kg intravenously (IV), infused by pump over 40 minutes, a dose selected from paediatric sedation standards and trials in depression and post-traumatic stress disorder, under an FDA Investigational New Drug (IND) authorisation, with vital signs monitored during infusion and at all visits, electrocardiography and laboratory tests, and visits on day 1 and at weeks 1, 2 and 4 [11, 52].

The primary safety endpoint was the number of participants with adverse events up to week 4: nine of ten [52]. All events were mild to moderate, none was serious and none required intervention; during infusion, only nausea and vomiting occurred, in one participant with a prior history [11]. The most frequent were elation or silly behaviour in 5 of 10 (all with a similar history), drowsiness or fatigue in 4, aggression in 4 (present in all at baseline), decreased appetite in 3 (new) and increased anxiety in 3; irritability, nausea or vomiting and restlessness each occurred in 2 [11, 52]. Efficacy analysis (baseline versus week 1, Wilcoxon test, without correction for multiple comparisons) showed nominally significant improvements in several domains, mostly sustained at weeks 2 and 4, without sleep improvement or changes in receptive and expressive vocabulary or visuomotor integration standard scores [11].

Table source: [11]. In the trial registry, CGI improvement was 1.5 (SD 0.53) at week 2 and 1.0 (SD 0.52) at week 4 [52]. In electrophysiological biomarkers, auditory steady-state response phase coherence increased in the 40 Hz gamma band and decreased in the 20 Hz beta band (p = 0.05; 7 and 8 analysable participants), and joint attention eye tracking, analysable in only four, showed a non-significant trend (p = 0.068) [11]. In peripheral blood from the same participants, ketamine produced immediate, profound transcriptomic changes, particularly in monocyte pathways (increased immune and inflammatory processes, decreased RNA processing and metabolic functions), which returned to baseline between 24 hours and one week; the hypothesis of increased ADNP mRNA was not confirmed because expression variance was too high to draw conclusions [1, 11, 54]. In patient lymphoblastoid cell lines, only toxic ketamine concentrations increased ADNP protein, with changes in α-tubulin and β-actin [55].

Limitations acknowledged by the authors are the small number of participants, lack of placebo, use of scales not validated in HVDAS and absence of correction for multiple comparisons; the authors themselves propose a randomised placebo-controlled trial with repeated doses and more patients [11]. In an open-label design with assessors and parents aware of the intervention, improvements in caregiver-completed scales may reflect expectancy effects and regression to the mean, and the study does not establish efficacy [11, 22]. The 2026 recommendations, written by the same team, present the trial as showing “significant improvements”, wording that should be read in light of these limitations [3]. Ketamine should not be used to treat HVDAS outside a trial. The FDA label restricts indications to anaesthetic use, states that safety and efficacy below 16 years have not been established, and warns of transient increases in blood pressure, heart rate and cardiac index, emergence reactions with confusion, respiratory depression with overdose or rapid administration, and possible cognitive deficits after exposures longer than three hours in children up to three years, based on neuronal loss in young animals exposed to NMDA-blocking anaesthetics [51]. The trial consent also mentioned that repeated ketamine use is associated with brain changes, cognitive problems and increased inflammatory cytokines [11].

7.12. Davunetide (NAP, CP201)

Davunetide is the octapeptide NAPVSIPQ, the NAP motif of ADNP itself, formulated for intranasal administration [50, 56]. In cells with ADNP variants it corrects microtubule and nucleocytoplasmic transport abnormalities, and enters cell nuclei [24, 26]. In aged mice with ADNP variants, its effects on motor function and circadian rhythm depended on mutation and sex, and the same research group describes oestrous cycle-dependent intranasal bioavailability, arguing for sex-specific doses [39, 57]. All these studies are preclinical and come from the same research group (Gozes); the peptide licence belongs to Ramot, Tel Aviv University, and was successively licensed to Coronis Neurosciences (2017–2021) and ExoNavis Therapeutics [24, 39, 56, 57]. In adults, Allon Therapeutics studied davunetide in mild cognitive impairment, schizophrenia, tauopathies and progressive supranuclear palsy; the phase 2/3 trial in the latter (313 patients, 30 mg intranasally twice daily versus placebo for one year) was negative for all primary, secondary and exploratory endpoints and halted development in that indication [56, 58].

In the European Union, davunetide received orphan medicinal product designation for treatment of ADNP syndrome (EU/3/24/3007), with a designation date of 13 December 2024 and publication on 13 January 2025, in the name of AdRes EU B.V.; the European Medicines Agency (EMA) notes that orphan designation does not mean the medicine is available or authorised [12]. An earlier orphan designation for progressive supranuclear palsy (EU/3/10/728, 2010) was withdrawn in 2013 [12, 22]. In the USA, orphan drug designation no. DRU-2017-6243 for treating ADNP syndrome with CP201, granted to Coronis Neurosciences, is cited by a 2020 review authored by the company's chief scientific officer and by the ADNP Kids Research Foundation, which dates it to 2018 alongside a pre-IND meeting; the official FDA orphan designation database could not be consulted for this review, so the date and current status of the US designation rely solely on these secondary sources [50, 53, 56]. According to the foundation, the drug never entered clinical trials with Coronis, dissolved in 2021, and the foundation states that it does not support the project or associated companies [53]. Alzforum, citing the successor company ExoNavis Therapeutics website, reports that a phase 3 trial in 97 children with ADNP variants began in October 2024, but this trial is not listed on ClinicalTrials.gov, where the October 2026 search found no davunetide trial in HVDAS; other registries (European and Israeli) were not searched [56, 58]. Without registration, a published protocol or results, that trial cannot be taken as evidence.

7.13. Other investigational approaches

The remaining therapeutic targets have been studied only in animal or cellular models. Inhibition of the histone demethylase LSD1 improved synaptic deficits in mice with Adnp variants; everolimus improved cognitive and synaptic deficits in mice with ADNP knockdown in the prefrontal cortex; and a mouse model with a frameshift variant identified the Wnt pathway as a potential target [33, 59, 60]. There are no registered gene therapy or antisense oligonucleotide trials for HVDAS, which does not exclude unregistered preclinical work [58]. The Seaver Autism Center observational study (NCT03718936; 30 planned participants; genetic, medical and neuropsychological measures; completion expected in March 2027) and the Simons Searchlight registry (NCT01238250) are ongoing, include patients with HVDAS and are preparing outcome measures for future trials [58, 61]. Visual evoked potentials are among the candidate biomarkers [42].

7.14. Regulatory status

Table sources: [3, 12, 50, 51, 53, 58]. The European ketamine SmPC was not consulted; ketamine indications in the EU are authorised through national procedures.

7.15. Trials and clinical studies

Table sources: [11, 40, 52, 54, 56, 58]. There are no randomised trials in HVDAS. Results from an open-label trial with ten participants cannot be compared with those of other interventions, and reductions in mean scores do not equate to a proportion of patients with clinically relevant benefit. The negative davunetide trial in adults with progressive supranuclear palsy does not predict its effect in a disease with a different mechanism, but shows that peptide tolerability and activity did not translate into clinical benefit in that indication.

7.16. Developments in the last 12 months

Between October 2025 and October 2026, the following were published: the first dedicated clinical recommendations (expert consensus, September 2026) [3]; the Antwerp systematic review with 144 patients (systematic review of case reports, December 2025) [6]; the in vitro study showing that ketamine increases ADNP only at toxic concentrations (preclinical, October 2025) [55]; the dizygotic twin case with multi-omic characterisation (case report) [32]; the first Turkish series, with reduced intrafamilial penetrance (case series) [7]; an Indian series published in a supplement, for which the abstract was unavailable [62]; two functional studies of missense variants (case report and experimental study) [28, 29]; visual evoked potentials as a biomarker (observational study) [42]; sleep profiles in neurogenetic diseases (observational study) [41]; cochlear nerve deficiency (case report) [38]; naltrexone for self-injury (case report) [40]; and several animal model studies of davunetide, LSD1, everolimus and Wnt (preclinical) [33, 39, 57, 60]. The European orphan designation of davunetide (December 2024) is a regulatory eligibility decision, not an authorisation [12]. There were no authorisations or opinions recommending authorisation for HVDAS during this period from any agency consulted.

7.17. Precautions, pregnancy, perioperative care and transition

No drug is specifically contraindicated in HVDAS, but sources draw attention to increased sensitivity to standard doses and paradoxical effects in neurodevelopmental disorders, metabolic monitoring with second-generation antipsychotics and the interaction of some antiepileptics with oral contraceptives [3]. There are no published data on pregnancy in women with HVDAS or reproduction in affected individuals [1]. The sources also describe no syndrome-specific anaesthetic complications; documented comorbidities relevant to perioperative planning are aspiration risk, obstructive sleep apnoea, congenital heart disease, hypotonia and a high pain threshold, which makes postoperative pain assessment difficult in a patient with limited communication [1, 3]. Transition to adult care should be prepared during adolescence: in the USA, school services continue up to 21 years under the Individuals with Disabilities Education Act, and the recommendations advise legal consultation about trust funds and eligibility for support; psychiatric surveillance should be at least annual in adulthood [1, 3]. Ongoing assessment of palliative care, home nursing and caregiver respite needs forms part of follow-up [1].

7.18. Differences between sources

Table sources: [1, 3, 6]. In practice, the more recent, age-specific 2026 recommendations underpin the follow-up described in section 11, with the caveat that they come from a single US centre and ungraded consensus.

8. Reproductive Options and Family Counselling

8.1. Recurrence risks

Most probands whose parents have been tested have a de novo variant, and sibling recurrence had not been reported up to 2025 [1, 6]. Parental testing is always recommended after the variant is identified, even if parents appear unaffected, because penetrance is incomplete and there are families in which the variant was inherited from a parent without apparent disease or with minimal expression [1, 3, 7]. Sibling risk depends on the result.

Table sources: [1, 7, 63]. If the variant is not detected in either parent, it may have arisen de novo in the proband, constitutionally or postzygotically, or one parent may have germline mosaicism, either isolated or associated with somatic mosaicism; leucocyte DNA may not detect all somatic mosaicism and does not detect exclusively germline mosaicism, and a parent with somatic and germline mosaicism may theoretically have mild or minimal manifestations [1, 36]. Risk to other relatives depends on parental status and is presumably low [1].

8.2. Cascade testing, minors and relatives

There is no recommendation to test at-risk relatives to anticipate medical interventions, because no presymptomatic surveillance or treatment changes prognosis [1]. The sources consulted do not specifically address testing asymptomatic minor siblings of a proband with an inherited variant; as there is no presymptomatic intervention, its value is mainly for future reproduction, and counselling should address the variable expressivity demonstrated in families with reduced penetrance [1, 7].

8.3. Prenatal diagnosis and preimplantation genetic testing

The ideal time to discuss risk and reproductive options is before pregnancy, and counselling should be offered to parents of affected children [1]. Once the familial variant has been identified, prenatal molecular testing and preimplantation genetic testing are possible [1]. In the de novo variant scenario, the approximately 1% residual risk due to germline mosaicism is the factor to weigh in this decision [1, 63]. Ultrasound is not a reliable detection method because prenatal abnormalities are rare and non-specific [1, 6]. Views on using prenatal diagnosis and preimplantation testing differ among healthcare professionals and families, and the decision is personal [1].

8.4. Counselling content

Genetic counselling should be offered to all families and cover disease characteristics, variable expressivity, recurrence risk, family planning and opportunities for research participation, such as patient registries and natural history studies [1, 3]. It should also explain that, with current data, neither episignature class nor the specific variant allows reliable prediction of individual severity [3, 17].

9. Special Considerations: Population Diversity and Epidemiology

9.1. Prevalence and what each number measures

There is no validated estimate of HVDAS population prevalence. Orphanet classifies worldwide point prevalence as unknown (June 2026 Orphadata) [2]. The most frequently cited figure, 0.17% (95% CI 0.083–0.32%), is the proportion of patients with ASD carrying an ADNP variant in the 2014 discovery cohort, which included 5776 patients from European and North American cohorts; it measures the contribution of ADNP to ASD in an ascertained population, rather than disease frequency in the general population [1, 4]. The 2026 recommendations cite a prevalence of 1/20 000 to 1/100 000 “depending on the source” and approximately 0.2% of ASD cases, without describing the population estimation method [3]. Published case numbers grew from 10 in 2014 and 78 in 2019 to 144 with clinical data in 2025 and 209 in 2026 [3, 4, 6, 15]. As exome and genome sequencing have become first-line tests in ID, any current count tends to underestimate the true frequency [9, 22]. ADNP-related blepharophimosis–ID syndrome has approximately 30 reported cases or families and a prevalence below 1/1 000,000 according to Orphanet [8].

9.2. Geographical distribution

The disease is pan-ethnic and no founder variants have been described; recurrent variants, such as p.(Tyr719Ter), arise de novo in unrelated individuals, reflecting recurrent mutational hotspots rather than a founder effect [4, 6]. The cohort of 78 was worldwide, and the 2025 review included patients from 15 countries [6, 15]. North American data come mainly from the Seaver Autism Center, which conducted the ketamine trial and observational study and produced the 2026 recommendations [3, 11]. In Europe, the scientific reference centre is the Antwerp group, which authors GeneReviews, and publications have included cases from Italy and the first national series from Turkey, in 2026 [1, 7, 28]. Series from China and India have also been published [29, 62]. The sources consulted did not include data from Portugal or Brazil, which does not demonstrate absence of diagnosed patients in those countries. In the ketamine trial, 8 of 10 participants were White and 2 Hispanic, limiting generalisability [11].

9.3. Special populations

Adults are poorly represented in series, and declining or plateauing skills in adolescence and adulthood are known only from longitudinal data on nine individuals [6]. The sex differences observed in mouse models in peripheral gene expression, response to davunetide and motor and circadian behaviour do not yet have a demonstrated clinical counterpart [27, 39, 57]. In girls and women, precocious puberty, menstrual irregularities and contraceptive needs require gynaecological attention adapted to disability [3].

9.4. Reference networks and patient organisations

In the European Union, the relevant European Reference Network is ERN ITHACA, dedicated to rare congenital anomalies and intellectual disability (https://ern-ithaca.eu/) [13]. The patient organisations listed by GeneReviews are the ADNP Kids Research Foundation (USA; https://www.adnpfoundation.org/), which maintains an international patient and natural history registry, ADNP Kids-HVDAS Belgium, Amis de ADNP France, ADNP Italia, ADNP Asociace (Czech Republic) and the Mickcare Foundation (Netherlands) [1, 3]. The Simons Searchlight registry (https://www.simonssearchlight.org/) collects data from people with genetic changes associated with autism, including ADNP [58, 61]. The sources consulted did not identify specific organisations in Portugal or Brazil; Orphanet serves as a directory of European organisations [2].

10. Illustrative Clinical Case

D'Incal and colleagues published in 2025 the case of a pair of male dizygotic twins discordant for HVDAS, followed in the USA and studied in Antwerp [32]. The affected twin was born at 36 weeks by caesarean section, indicated by a positive contraction stress test and reduced fetal movements. In the neonatal period he had macrocephaly, facial dysmorphic features, a small chest, undescended testes and a heart murmur; echocardiography revealed an atrial septal defect and right aortic arch with a vascular ring, both surgically corrected. Feeding difficulties led to gastrostomy placement in the first months, and follow-up documented silent aspiration, gastro-oesophageal reflux and chronic constipation. By one year he had complete dentition, including molars. Ophthalmological assessment showed strabismus, cortical visual impairment, irregular astigmatism and hypermetropia, and the patient had epileptic seizures. Urogenital findings were bilateral cryptorchidism with testicular atrophy, congenital inguinal hernia, a duplicated renal collecting system and a right renal abscess; endocrine findings were premature adrenarche and subclinical hypothyroidism. Development was globally delayed, with mixed receptive and expressive language disorder, speech apraxia and oromotor problems; at 8 years he was non-verbal, communicating only through vocalisations. Severe ID, ASD, sensory processing disorder and ADHD were diagnosed, with behavioural abnormalities, an irregular sleep–wake rhythm and obstructive sleep apnoea. On examination (aged 15 years at the time of the study) he had motor planning difficulties, hypotonia and global muscle atrophy, elbow contractures, reduced knee and hip extension during walking, leg length discrepancy and bilateral flat feet. His twin brother developed normally [32].

Exome sequencing identified the heterozygous de novo variant c.1046_1047delTG, p.Leu349ArgfsX49 (normalised as NM_001282531.3:c.1046_1047del, p.(Leu349ArgfsTer49)), confirmed by Sanger sequencing and absent in his brother; this N-terminal variant eliminates the NAP motif, NLS and homeodomain [5, 32]. PhenoScore was 0.92 in the patient and 0.11 in his brother. Methylation analysis showed a class 1 episignature, with predominant hypomethylation, and RNA-seq of peripheral blood mononuclear cells showed 2329 upregulated and 822 downregulated genes, with enrichment in actin organisation, Wnt signalling, cardiac development and immune pathways [32]. MRI at 6 months showed delayed myelination and slightly enlarged subarachnoid spaces; at 9 years there were no major abnormalities, enlarged subarachnoid spaces persisted and there were non-specific punctate blooming foci in the right temporal region, left posterior putamen and ventricular margin of the left caudate (calcifications or residual microhaemorrhages), without signs of neurodegeneration [32]. Figure 1 of the paper depicts the protein with the variant location, Sanger electropherogram and protein consequence (https://pmc.ncbi.nlm.nih.gov/articles/PMC12599069/figure/Fig1/), and Figure 4 shows MRI at 6 months and 9 years (https://pmc.ncbi.nlm.nih.gov/articles/PMC12599069/figure/Fig4/); the figures were consulted through their respective legends [32].

The case brings together almost the entire disease spectrum in one patient: hypotonia, feeding difficulties with aspiration, congenital heart disease, early tooth eruption, cerebral visual impairment, epilepsy, urogenital anomalies, endocrine abnormalities, severe ID with ASD and absence of speech. It also illustrates two interpretative points. Phenotypic severity despite a class 1 episignature challenges the association of this class with milder forms, reinforcing the recommendation not to use episignature class to predict individual prognosis [3, 17, 32]. Discordance between dizygotic twins, who share the intrauterine and family environment, is consistent with a de novo variant restricted to the affected brother. Methylation, transcriptome and PhenoScore analyses were performed in a research setting and are not routine clinical practice; surveillance described in the case broadly corresponds to the general recommendations in section 11.

11. Prognosis and Long-Term Follow-Up

11.1. Prognosis

HVDAS is a chronic neurodevelopmental disease with permanent ID, generally moderate to severe, and variable dependence in activities of daily living; few patients attain full independence [1, 3]. Most acquire walking (86% in the 2025 review), but approximately 29% do not develop spoken language [6]. Longitudinal data, limited to nine individuals, suggest slowing, plateauing or declining skills in adolescence and adulthood, and regression was reported in approximately one fifth of individuals assessed [3, 6]. There are no survival data. The oldest individual known to the GeneReviews authors was 69 years old, and the adults described reached adulthood with childhood comorbidities, which GeneGuidance interprets as a disease that is not life-limiting in the usual sense; however, there are no data allowing life expectancy to be quantified [1, 22]. The comorbidities with the greatest morbidity potential described in the sources are aspiration, recurrent infections, epilepsy, complications of congenital heart disease, obesity and behavioural problems, most of which are treatable [1, 3].

11.2. Surveillance

The table combines the Seaver Autism Center age-specific recommendations (2026) with the GeneReviews surveillance table (2025). Where they differ, both are stated.

GeneReviews places neurological surveillance on an annual schedule, with seizure follow-up according to clinical findings, but its summary includes seizures among items monitored at every visit; in practice, questions about seizures should form part of all visits, and formal neurological examination should be performed at least annually [1, 3].

12. Summary and Final Considerations

Helsmoortel-Van der Aa syndrome results from heterozygous ADNP variants, almost always truncating, de novo and located in the last exon, and is one of the most frequent monogenic causes of syndromic ASD, present in approximately 0.17% of patients with ASD in the discovery cohort. The presentation combines ID, almost always moderate to severe, marked expressive language delay, autistic traits, hypotonia, recognisable facies and multisystem comorbidities: feeding difficulties and aspiration, reflux and constipation, cerebral visual impairment and strabismus, congenital heart disease, recurrent infections, sleep disturbances, epileptic seizures, short stature, thyroid abnormalities and precocious puberty. Early eruption of primary dentition is a useful clue. Diagnosis is made through exome or genome sequencing during investigation of developmental delay, and the methylation episignature, with its two classes, helps interpret variants of uncertain significance and identify non-coding variants, without allowing individual severity to be predicted. The molecular mechanism remains debated between dominant-negative effects, gain of function and haploinsufficiency, and missense variants require functional validation. There is no targeted treatment: management relies on intensive, continuous early intervention, augmentative communication, behavioural intervention, judicious symptomatic treatment and multisystem surveillance, now organised by age in the 2026 recommendations. Low-dose ketamine was well tolerated in an open-label trial of ten children, without established efficacy, and davunetide has European orphan designation without a registered trial in the disease. Counselling should consider incomplete penetrance, parental testing and residual germline mosaicism risk. The main gaps are the absence of population prevalence studies, adult cohorts and controlled trials.

13. References

1. Harutyunyan L, Kooy RF, D'Incal CP, Van Dijck A. ADNP-Related Helsmoortel-Van der Aa Syndrome. 2016 Apr 7 [updated 2025 Aug 21]. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Available from: https://www.ncbi.nlm.nih.gov/books/NBK355518/
2. Orphanet. Helsmoortel-Van der Aa syndrome (ORPHA:404448) [Internet]. Paris: INSERM; [cited 2026 Oct 9]. Available from: https://www.orpha.net/en/disease/detail/404448
3. Fastman J, Kolevzon A, Siper P, Levy T, Saland JM, Frank Y, et al. ADNP-Related Helsmoortel-Van der Aa Syndrome: A Review of the Literature and Clinical Recommendations for Assessment and Monitoring. Am J Med Genet A. 2026 Sep 17 [Epub ahead of print]. doi:10.1002/ajmg.a.70288.
4. Helsmoortel C, Vulto-van Silfhout AT, Coe BP, Vandeweyer G, Rooms L, van den Ende J, et al. A SWI/SNF-related autism syndrome caused by de novo mutations in ADNP. Nat Genet. 2014;46(4):380-4. doi:10.1038/ng.2899.
5. Freeman PJ, Hart RK, Gretton LJ, Brookes AJ, Dalgleish R. VariantValidator: Accurate validation, mapping, and formatting of sequence variation descriptions. Hum Mutat. 2018;39(1):61-8. doi:10.1002/humu.23348.
6. Harutyunyan L, D'Incal CP, Jansen AC, Meuwissen M, Van Dijck A, Kooy RF. A Systematic Review Illustrates the Expanding Clinical and Molecular Landscape of Helsmoortel-Van der Aa Syndrome. Brain Sci. 2025;16(1):4. doi:10.3390/brainsci16010004.
7. Kablan A, Sezer A, Deniz A, Taşdelen E, Dinçsoy Bir F, Kolkıran A, et al. ADNP-Related Neurodevelopmental Disorder: The First Turkish Case Series with Novel Variants and Reduced Intrafamilial Penetrance. Diagnostics (Basel). 2026;16(17):2879. doi:10.3390/diagnostics16172879.
8. Orphanet. ADNP-related blepharophimosis-intellectual disability syndrome (ORPHA:700160) [Internet]. Paris: INSERM; [cited 2026 Oct 9]. Available from: https://www.orpha.net/en/disease/detail/700160
9. Manickam K, McClain MR, Demmer LA, Biswas S, Kearney HM, Malinowski J, et al. Exome and genome sequencing for pediatric patients with congenital anomalies or intellectual disability: an evidence-based clinical guideline of the American College of Medical Genetics and Genomics (ACMG). Genet Med. 2021;23(11):2029-2037. doi:10.1038/s41436-021-01242-6.
10. Rodan LH, Stoler J, Chen E, Geleske T. Genetic Evaluation of the Child With Intellectual Disability or Global Developmental Delay: Clinical Report. Pediatrics. 2025;156(1):e2025072219. doi:10.1542/peds.2025-072219.
11. Kolevzon A, Levy T, Barkley S, Bedrosian-Sermone S, Davis M, Foss-Feig J, et al. An open-label study evaluating the safety, behavioral, and electrophysiological outcomes of low-dose ketamine in children with ADNP syndrome. HGG Adv. 2022;3(4):100138. doi:10.1016/j.xhgg.2022.100138.
12. European Medicines Agency. Davunetide — orphan designation EU/3/24/3007 for treatment of activity-dependent neuroprotective protein (ADNP) syndrome [Internet]. Amsterdam: EMA; 2025 [designation 2024 Dec 13; published 2025 Jan 13; cited 2026 Oct 9]. Available from: https://www.ema.europa.eu/en/medicines/human/orphan-designations/eu-3-24-3007
13. European Reference Network ITHACA (Intellectual disability, TeleHealth, Autism and Congenital Anomalies) [Internet]. [cited 2026 Oct 9]. Available from: https://ern-ithaca.eu/
14. Vandeweyer G, Helsmoortel C, Van Dijck A, Vulto-van Silfhout AT, Coe BP, Bernier R, et al. The transcriptional regulator ADNP links the BAF (SWI/SNF) complexes with autism. Am J Med Genet C Semin Med Genet. 2014;166C(3):315-26. doi:10.1002/ajmg.c.31413.
15. Van Dijck A, Vulto-van Silfhout AT, Cappuyns E, van der Werf IM, Mancini GM, Tzschach A, et al. Clinical Presentation of a Complex Neurodevelopmental Disorder Caused by Mutations in ADNP. Biol Psychiatry. 2019;85(4):287-297. doi:10.1016/j.biopsych.2018.02.1173.
16. Bend EG, Aref-Eshghi E, Everman DB, Rogers RC, Cathey SS, Prijoles EJ, et al. Gene domain-specific DNA methylation episignatures highlight distinct molecular entities of ADNP syndrome. Clin Epigenetics. 2019;11(1):64. doi:10.1186/s13148-019-0658-5.
17. Breen MS, Garg P, Tang L, Mendonca D, Levy T, Barbosa M, et al. Episignatures Stratifying Helsmoortel-Van Der Aa Syndrome Show Modest Correlation with Phenotype. Am J Hum Genet. 2020;107(3):555-563. doi:10.1016/j.ajhg.2020.07.003.
18. Georget M, Lejeune E, Buratti J, Servant E, le Guern E, Heron D, et al. Loss of function of ADNP by an intragenic inversion. Eur J Hum Genet. 2023;31(8):967-970. doi:10.1038/s41431-023-01323-x.
19. D'Incal CP, Annear DJ, Elinck E, van der Smagt JJ, Alders M, Dingemans AJM, et al. Loss-of-function of activity-dependent neuroprotective protein (ADNP) by a splice-acceptor site mutation causes Helsmoortel-Van der Aa syndrome. Eur J Hum Genet. 2024;32(6):630-638. doi:10.1038/s41431-024-01556-4.
20. Sarli C, van der Laan L, Reilly J, Trajkova S, Carli D, Brusco A, et al. Blepharophimosis with intellectual disability and Helsmoortel-Van Der Aa Syndrome share episignature and phenotype. Am J Med Genet C Semin Med Genet. 2024;196(4):e32089. doi:10.1002/ajmg.c.32089.
21. Takenouchi T, Miwa T, Sakamoto Y, Sakaguchi Y, Uehara T, Takahashi T, et al. Further evidence that a blepharophimosis syndrome phenotype is associated with a specific class of mutation in the ADNP gene. Am J Med Genet A. 2017;173(6):1631-1634. doi:10.1002/ajmg.a.38126.
22. GeneGuidance. ADNP: gene and Helsmoortel-Van der Aa syndrome condition record [database]. Version of 26 September 2026 [cited 2026 Oct 9].
23. National Center for Biotechnology Information. Nucleotide: NM_001282531.3 and NM_015339.5 (Homo sapiens ADNP) [Internet]. Bethesda (MD): NLM [cited 2026 Oct 9]. Available from: https://www.ncbi.nlm.nih.gov/nuccore/NM_001282531.3
24. Ganaiem M, Gildor ND, Shazman S, Karmon G, Ivashko-Pachima Y, Gozes I. NAP (Davunetide): The Neuroprotective ADNP Drug Candidate Penetrates Cell Nuclei Explaining Pleiotropic Mechanisms. Cells. 2023;12(18):2251. doi:10.3390/cells12182251.
25. D'Incal CP, Cappuyns E, Choukri K, De Man K, Szrama K, Konings A, et al. Tracing the invisible mutant ADNP protein in Helsmoortel-Van der Aa syndrome patients. Sci Rep. 2024;14(1):14710. doi:10.1038/s41598-024-65608-x.
26. Ganaiem M, Karmon G, Ivashko-Pachima Y, Gozes I. Distinct Impairments Characterizing Different ADNP Mutants Reveal Aberrant Cytoplasmic-Nuclear Crosstalk. Cells. 2022;11(19):2994. doi:10.3390/cells11192994.
27. Karmon G, Sragovich S, Hacohen-Kleiman G, Ben-Horin-Hazak I, Kasparek P, Schuster B, et al. Novel ADNP Syndrome Mice Reveal Dramatic Sex-Specific Peripheral Gene Expression With Brain Synaptic and Tau Pathologies. Biol Psychiatry. 2022;92(1):81-95. doi:10.1016/j.biopsych.2021.09.018.
28. Benvenuto M, Giacomo MCD, Piepoli A, Carella M, D'Addetta P, Palumbo O, et al. The De Novo p.(Ser802Phe) Variant Causes Helsmoortel-Van der Aa/ ADNP Syndrome in a 24-Year-Old Woman and Is Predicted to Perturb ADNP-DNA Affinity. Am J Med Genet A. 2026;200(2):468-474. doi:10.1002/ajmg.a.64264.
29. Chen Q, Liu X, Wu M, Sun D, Ding X, Zhou M, et al. ADNP missense variant p.C687R disrupts chromatin regulation and GABAergic differentiation in Helsmoortel-Van der Aa syndrome. Mol Autism. 2026;17(1):19. doi:10.1186/s13229-026-00713-4.
30. D'Incal CP, Van Rossem KE, De Man K, Konings A, Van Dijck A, Rizzuti L, et al. Chromatin remodeler Activity-Dependent Neuroprotective Protein (ADNP) contributes to syndromic autism. Clin Epigenetics. 2023;15(1):45. doi:10.1186/s13148-023-01450-8.
31. D'Incal C, Van Dijck A, Ibrahim J, De Man K, Bastini L, Konings A, et al. ADNP dysregulates methylation and mitochondrial gene expression in the cerebellum of a Helsmoortel-Van der Aa syndrome autopsy case. Acta Neuropathol Commun. 2024;12(1):62. doi:10.1186/s40478-024-01743-w.
32. D'Incal CP, Van Dijck A, Annear DJ, Harutyunyan L, Elinck E, Dingemans AJM, et al. A frameshift variant in activity-dependent neuroprotective protein (ADNP) causes nucleocytoskeletal alterations in a dizygotic male twin: a case study. Clin Epigenetics. 2025;17(1):185. doi:10.1186/s13148-025-01998-7.
33. D'Incal CP, Cappuyns E, Paldi F, van der Lei MB, Annear DJ, Alastruey CM, et al. An Adnp frameshift variant disrupts Wnt signalling inducing chromatocytoskeletal defects and autism-related behaviour in male mice. EBioMedicine. 2026;128:106309. doi:10.1016/j.ebiom.2026.106309.
34. Huynh MT, Boudry-Labis E, Massard A, Thuillier C, Delobel B, Duban-Bedu B, et al. A heterozygous microdeletion of 20q13.13 encompassing ADNP gene in a child with Helsmoortel-van der Aa syndrome. Eur J Hum Genet. 2018;26(10):1497-1501. doi:10.1038/s41431-018-0165-8.
35. National Center for Biotechnology Information. ClinVar: VCV000139635, VCV000279598, VCV000139632 (ADNP) [Internet]. Bethesda (MD): NLM [cited 2026 Oct 9]. Available from: https://www.ncbi.nlm.nih.gov/clinvar/
36. Mohiuddin M, Kooy RF, Pearson CE. De novo mutations, genetic mosaicism and human disease. Front Genet. 2022;13:983668. doi:10.3389/fgene.2022.983668.
37. Dingemans AJM, Hinne M, Truijen KMG, Goltstein L, van Reeuwijk J, de Leeuw N, et al. PhenoScore quantifies phenotypic variation for rare genetic diseases by combining facial analysis with other clinical features using a machine-learning framework. Nat Genet. 2023;55(9):1598-1607. doi:10.1038/s41588-023-01469-w.
38. Carratu K, Crocker K, Chen S, Moore C, Patel NA, Krantz ID. Expanding the Audiological Phenotype Associated With Activity-Dependent Neuroprotective Protein (ADNP) Syndrome: A Case Report and Literature Review Suggesting a Genotype/Phenotype Correlation. Am J Med Genet A. 2026;200(8):1799-1806. doi:10.1002/ajmg.a.70150.
39. Lobyntseva A, Guz LS, Galushkin A, Gozes I. Dramatic sex differences leading to different brain disease presentation: The requirement for sex-specific medications with ADNP/davunetide as a case study. Front Neuroendocrinol. 2026;83:101274. doi:10.1016/j.yfrne.2026.101274.
40. Durmaz GN, Kahrama NN, Coşkun M. Severe self-injury in a young girl with autism spectrum disorder and Helsmoortel-van der Aa syndrome responding to naltrexone. Psychiatr Genet. 2026 [Epub ahead of print, 18 Aug 2026]. doi:10.1097/YPG.0000000000000429. PMID: 42704315.
41. Reyes IC, Frazier TW, Hardan AY, Chetcuti L, Chawner SJRA, Spackman E, et al. Sleep profiles in individuals with rare neurogenetic syndromes. Dev Med Child Neurol. 2026 [Epub ahead of print]. doi:10.1111/dmcn.70337.
42. Levy T, Silver H, Benrey N, Siegel A, Layton C, Zemon V, et al. Transient visual evoked potential abnormalities in ADNP syndrome. J Neurodev Disord. 2026;18(1):25. doi:10.1186/s11689-026-09688-y.
43. Gozes I, Van Dijck A, Hacohen-Kleiman G, Grigg I, Karmon G, Giladi E, et al. Premature primary tooth eruption in cognitive/motor-delayed ADNP-mutated children. Transl Psychiatry. 2017;7(2):e1043. doi:10.1038/tp.2017.27.
44. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 326–327 (Blepharophimosis-Ptosis-Epicanthus Inversus Syndrome).
45. Rosenblum J, Van der Veeken L, Aertsen M, Meuwissen M, Jansen AC. Abnormal fetal ultrasound leading to the diagnosis of ADNP syndrome. Eur J Med Genet. 2023;66(11):104855. doi:10.1016/j.ejmg.2023.104855.
46. Richards S, Aziz N, Bale S, Bick D, Das S, Gastier-Foster J, et al. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet Med. 2015;17(5):405-24. doi:10.1038/gim.2015.30.
47. D'Incal CP, Kooy RF. ADNP in reverse gear. Eur J Hum Genet. 2023;31(8):849-850. doi:10.1038/s41431-023-01360-6.
48. Kosho T, Okamoto N, Ohashi H, Tsurusaki Y, Imai Y, Hibi-Ko Y, et al. Clinical correlations of mutations affecting six components of the SWI/SNF complex: detailed description of 21 patients and a review of the literature. Am J Med Genet A. 2013;161A(6):1221-37. doi:10.1002/ajmg.a.35933.
49. Bögershausen N, Wollnik B. Mutational Landscapes and Phenotypic Spectrum of SWI/SNF-Related Intellectual Disability Disorders. Front Mol Neurosci. 2018;11:252. doi:10.3389/fnmol.2018.00252.
50. Gozes I. The ADNP Syndrome and CP201 (NAP) Potential and Hope. Front Neurol. 2020;11:608444. doi:10.3389/fneur.2020.608444.
51. Par Health USA. KETALAR (ketamine hydrochloride) injection, for intravenous or intramuscular use [prescribing information]. NDA016812. Revised 2026 Mar [cited 2026 Oct 9]. Available from: https://api.fda.gov/drug/label.json?search=openfda.brand_name:%22KETALAR%22 (DailyMed)
52. Icahn School of Medicine at Mount Sinai (Kolevzon A). Low-Dose Ketamine in Children With ADNP Syndrome. ClinicalTrials.gov identifier: NCT04388774 [Internet]. Bethesda (MD): National Library of Medicine; 2020 [cited 2026 Oct 9]. Available from: https://clinicaltrials.gov/study/NCT04388774
53. ADNP Kids Research Foundation. Research and drug development timeline [Internet]. [undated; cited 2026 Oct 9]. Available from: https://www.adnpfoundation.org/research-timeline.html
54. Buxbaum Grice AS, Sloofman L, Levy T, Walker H, Ganesh G, Rodriguez de Los Santos M, et al. Transient peripheral blood transcriptomic response to ketamine treatment in children with ADNP syndrome. Transl Psychiatry. 2024;14(1):307. doi:10.1038/s41398-024-03005-8.
55. D'Incal CP, Mateiu L, De Man K, Elinck E, Gozes I, Kooy RF. The ADNP-Mediated Transcriptome Response to Ketamine Impairs the Cytoskeletal Protein Axis. J Mol Neurosci. 2025;75(4):136. doi:10.1007/s12031-025-02420-w.
56. Alzforum. Therapeutics: Davunetide [Internet]. Alzforum; updated 2025 Jun 9 [cited 2026 Oct 9]. Available from: https://www.alzforum.org/therapeutics/davunetide
57. Galushkin A, Giladi E, Gozes I. Aging ADNP syndrome mice exhibit mutation/sex-dependent disruption of motor behavior and circadian rhythmicity. Neurobiol Dis. 2026;227:107509. doi:10.1016/j.nbd.2026.107509.
58. National Library of Medicine. ClinicalTrials.gov — search “ADNP OR Helsmoortel OR davunetide” (NCT03718936, NCT01238250, NCT00404014, NCT00422981, NCT00505765, NCT01056965, NCT01110720) [Internet]. [cited 2026 Oct 9]. Available from: https://clinicaltrials.gov/
59. Lin CH, Ren Y, Tam KW, Conrow-Graham M, Yan Z. Synaptic Deficits in Adnp-Mutant Mice Are Ameliorated by Histone Demethylase LSD1 Inhibition. Autism Res. 2025;18(7):1342-1355. doi:10.1002/aur.70069.
60. Xiang Y, Zhang Z, Jiang Y, Wei H. Everolimus ameliorates cognitive deficits and synaptic dysfunction in mice with prefrontal cortical ADNP knockdown. Neurosci Lett. 2026;871:138460. doi:10.1016/j.neulet.2025.138460.
61. Simons Searchlight [Internet]. Simons Foundation [cited 2026 Oct 9]. Available from: https://www.simonssearchlight.org/
62. Kafley P, Saxena D, Moirangthem A. Clinical and Genetic Characterization of Patients with ADNP Related Helsmoortel-Van der Aa Syndrome. Indian J Pediatr. 2026;93(Suppl 1):137-139. doi:10.1007/s12098-025-05894-z.
63. Rahbari R, Wuster A, Lindsay SJ, Hardwick RJ, Alexandrov LB, Turki SA, et al. Timing, rates and spectra of human germline mutation. Nat Genet. 2016;48(2):126-133. doi:10.1038/ng.3469.