Comprehensive study of 28 individuals with *SIN3A*-related disorder underscoring the associated mild cognitive and distinctive facial phenotype

Witteveen-Kolk syndrome (OMIM 613406) was first described in 2016 and is characterized by distinctive facial features, microcephaly, short stature, and usually mild intellectual disability with delayed cognitive and motor development and subtle MRI brain anomalies. Frameshift and missense variants in SIN3A have been reported in larger neurodevelopmental cohorts with an overall mild clinical picture, and additional single cases have been described with relatively mild ID and normal growth. SIN3A resides at 15q24, within the shortest region of overlap of several 15q24 microdeletions, and is thought to be the critical gene for the atypical 15q24 microdeletion syndrome. SIN3A encodes a transcriptional regulatory scaffold protein within the core histone deacetylase complex. Prior work by the authors implicated SIN3A in cortical neurogenesis, supporting that variants adversely affecting SIN3A function cause a range of developmental and neurological problems. The aim of this study was to identify additional individuals with SIN3A variants and comprehensively define the clinical, developmental, and behavioural phenotype, and explore genotype–phenotype correlations.

- Initial description in 2016 detailed a syndrome with distinctive facial features, microcephaly, short stature, mild ID, and subtle MRI anomalies. - Frameshift and missense SIN3A variants have been reported in larger neurodevelopmental disorder cohorts with overall mild phenotypes, and individual case reports add to this spectrum. - SIN3A is within the critical region for atypical 15q24 microdeletion syndrome, supporting its dosage sensitivity as the key gene. - Prior functional work indicated a role for SIN3A in cortical neurogenesis, providing a mechanistic basis for neurodevelopmental manifestations. - Existing reports did not establish clustering of missense variants within a specific hotspot; this study further evaluates missense impacts within functional domains (notably PAH1).

- Case ascertainment: 28 unpublished individuals with rare SIN3A variants identified during routine diagnostic testing were assembled via collaborations through the Deciphering Developmental Disorders (DDD) study, GeneMatcher, and DECIPHER. A DDD Complementary Analysis Project enabled access to anonymized details; clinicians then recruited patients with consent. - Inclusion criteria: Individuals harboring SIN3A variants classified as pathogenic or likely pathogenic (ACMG class 4 or 5), with clinical presentation of developmental delay/ID or related phenotypes, and without confounding additional genetic diagnoses or chromosomal anomalies explaining the phenotype. - Exclusion criteria: Additional proven genetic diagnoses deemed explanatory, chromosomal anomalies likely to explain the phenotype, or SIN3A variants of uncertain significance without convincing clinical correlation. - Clinical data collection: Retrospective review of medical records and routine clinical evaluations covering medical history, physical examination, developmental and behavioural observations; multidisciplinary clinical input. - Genomic methods: Exome sequencing and variant filtering performed per routine diagnostic protocols at participating institutes. Variant pathogenicity was interpreted using ACMG guidelines and subsequent reviews. Variants and phenotypes were deposited in DECIPHER. - Variant classification: Predominantly truncating variants were classified pathogenic using PVS1, PM2, and PS2 where applicable. Missense variants were evaluated with structural modeling of the PAH1 domain (based on mouse SIN3A PAH1 bound to SAP25 SID) and applied PM1 at moderate level for missense within residues 119–189 when meeting PM2 and conservation criteria; computational evidence (PP3) and benign predictors (BP4) were considered as relevant.

- Cohort: 28 individuals (17 males, 11 females); mean age 8.2 years (range 0.6–67 years). - Neurodevelopment: - Global developmental delay in 15/28 (56%). - Among 16 patients aged ≥8 years (one not formally assessed): mild ID in 8 (50%), moderate ID in 4 (22%), severe ID in 1 (6%), and no cognitive impairment in 4 (25%). - Motor delay reported in 13/21 (62%) and language delay in 16/21 (76%) among those aged ≥5 years. - IQ formally measured in 6/28, ranging 60–100, suggesting low-normal intelligence in parts of the cohort. - ASD diagnosed in 3/23 (13%) patients aged ≥2 years; ADHD diagnosed in 4/23 (17%) aged ≥2 years. - Neurology and behaviour: - Seizures in 4/28; hypotonia in 12/28; behaviour problems present in a minority. - Growth and feeding: - Microcephaly (OFC ≤ −2 SD) in 13/28 (46%). - Weight ≤ −2 SD in 8/28 (29%); height ≤ −2 SD in 5/28 (18%). - Feeding difficulties in 15/28 (54%); at least two required nasogastric feeding in the neonatal period. - Craniofacial: 14/28 (50%) reported craniofacial dysmorphism; a consistent facial gestalt was noted (high forehead, small pointed chin, down-slanting palpebral fissures) across illustrated cases. - Molecular genetics: - 27/28 variants novel; one recurrent (c.3310C>T; p.Arg1104*). - 24 truncating variants, three missense, and one large in-frame gain including exons 10–12 (reported as c.1318_1737dup; p.Val580Lysfs*35/36 in figures/tables). - Most were de novo; absence from gnomAD. - Missense variants not clustered in a hotspot; two missense (p.Ala126Val, p.Lys155Glu) mapped to the PAH1 SID-binding surface with structural rationale supporting pathogenicity; p.Met1106Thr in the C-terminus had benign computational predictions (REVEL 0.105) but was de novo with consistent phenotype and no other variants, leading to a likely pathogenic classification without PM1. - Classification framework: Application of ACMG criteria (PS2, PVS1, PM1, PM2, PP3, BP4) detailed; proposal to apply PM1 at moderate level for missense variants within residues 119–189 when conserved and absent in other species and meeting PM2.

This study expands the clinical and genetic spectrum of SIN3A-related Witteveen-Kolk syndrome by detailing 28 additional individuals, largely with de novo loss-of-function variants. The clinical data support a predominantly mild neurodevelopmental phenotype with variable cognitive outcomes ranging from low-normal intelligence to mild/moderate ID, along with frequent language and motor delays. Consistent craniofacial features and substantial rates of microcephaly and early feeding difficulties further delineate the phenotype. Neurological features such as hypotonia and seizures occur in a subset; ASD and ADHD are present in a minority of patients aged ≥2 years. Molecularly, most pathogenic variants are truncating and distributed across the gene, consistent with haploinsufficiency as the disease mechanism. Missense variants are less common and not hotspot-clustered; structural modeling indicates that variants within the PAH1 SID-binding surface can disrupt critical protein–protein interactions, supporting application of ACMG PM1 at a moderate level for conserved residues 119–189 when other criteria are met. Overall, the findings address the research aim by defining a consistent, generally mild neurodevelopmental and craniofacial phenotype associated with SIN3A haploinsufficiency and by providing guidance for variant interpretation, thereby informing diagnosis and counseling.

The cohort of 28 individuals with SIN3A variants broadens the variant spectrum (27 novel) and refines the associated clinical phenotype, which typically includes mild intellectual disability or low-normal cognition, characteristic facial features, microcephaly, and early feeding difficulties, with hypotonia and seizures in a subset. Missense variants can be pathogenic when affecting functional domains such as PAH1, supporting the moderated use of ACMG PM1 for residues 119–189 under defined conditions. Clinical management should consider a multidisciplinary team including genetics, paediatrics, and neurology.