Submission Details

SMARCA4 was first reported in relation to autosomal dominant Coffin-Siris syndrome (CSS) in 2012 (Tsurusaki et al., PMID: 22426308). CSS is a syndromic form of cognitive and developmental disability most commonly associated with fifth finger hypoplasia and distinctive facial dysmorphisms, including a wide mouth with thick, everted lips, thick eyebrows, and a broad nasal bridge, consistent with facial coarseness. Congenital anomalies of the brain, kidney, and heart have been described but are less consistent across patients. Approximately 35% of all CSS patients have congenital heart disease, typically including ventricular septal defect, atrial septal defect, and tetralogy of Fallot (PMIDs: 23556151, 25298701, 31160358). Pathogenic variants in several genes encoding components of the BAF chromatin remodeling complex, including ARID1A, ARID1B, ARID2, SMARCA4, SMARCB1, SMARCC2, SMARCD1, SMARCE1, and DPF2, cause CSS (PMIDs: 25168959, 31530938). Heterozygous SMARCA4 pathogenic variants cause ~12% of CSS cases (PMID: 30123105).

Twenty-three unique de novo missense or in-frame variants (PMIDs: 22426308, 30792901, 31530938, 32686290, 38177409) and three frameshift or nonsense variants (PMIDs: 28608987, 32686290) that have been reported in 26 probands with CSS have been included in this curation. More evidence is available in the literature, but the maximum score for genetic evidence (12 points) has been reached. Most variants occur de novo. Almost all pathogenic missense variants reported in CSS cases are localized in the middle of the gene (within or near the HAS, DEXDc, and HELICc functional domains) and are expected to exert a dominant-negative or gain-of-function effect.

Experimental evidence supporting this gene-disease relationship includes the biochemical function of SMARCA4, protein interaction with ADNP and other BAF protein complex subunits, and mouse models recapitulating some of the neurological and cardiovascular phenotypes of CSS (PMIDs: 17878164, 23319608, 23698369, 26459759). SMARCA4 is involved in chromatin remodeling and transcriptional activation. Although pathogenic variants have been shown to alter transcriptional activation of development regulator genes (PMID: 23319608), the mechanism by which SMARCA4 pathogenic variants cause CSS is not clear.

In summary, there is definitive evidence supporting the relationship between SMARCA4 and autosomal dominant Coffin-Siris syndrome. This has been repeatedly demonstrated in both the research and clinical diagnostic settings, and has been upheld over time. This classification was initially approved by the ClinGen Intellectual Disability and Autism Gene Curation Expert Panel (GCEP) on June 17, 2020 (SOP Version 7). This gene-disease relationship was recurated by the ClinGen Congenital Heart Disease GCEP on June 9, 2025 with an updated SOP Version 11 to include more probands in the curation, however the classification remained the same (PMIDs: 31530938, 32686290, 38177409).

SMARCA4 was first reported in relation to autosomal dominant Coffin-Siris syndrome (CSS) in 2012 (Tsurusaki et al., PMID: 22426308). CSS is a syndromic form of cognitive and developmental disability most commonly associated with fifth finger hypoplasia and distinctive facial dysmorphisms, including a wide mouth with thick, everted lips, thick eyebrows, and a broad nasal bridge, consistent with facial coarseness. Congenital anomalies of the brain, kidney, and heart have been described but are less consistent across patients. Approximately 35% of all CSS patients have congenital heart disease, typically including ventricular septal defect, atrial septal defect, and tetralogy of Fallot (PMIDs: 23556151, 25298701, 31160358). Pathogenic variants in several genes encoding components of the BAF chromatin remodeling complex, including ARID1A, ARID1B, ARID2, SMARCA4, SMARCB1, SMARCC2, SMARCD1, SMARCE1, and DPF2, cause CSS (PMIDs: 25168959, 31530938). Heterozygous SMARCA4 pathogenic variants cause ~12% of CSS cases (PMID: 30123105).

Twenty-three unique de novo missense or in-frame variants (PMIDs: 22426308, 30792901, 31530938, 32686290, 38177409) and three frameshift or nonsense variants (PMID: 28608987, 32686290) that have been reported in 26 probands with CSS have been included in this curation. More evidence is available in the literature, but the maximum score for genetic evidence (12 points) has been reached. Most variants occur de novo. Almost all pathogenic missense variants reported in CSS cases are localized in the middle of the gene (within or near the HAS, DEXDc, and HELICc functional domains) and are expected to exert a dominant-negative or gain-of-function effect.

Experimental evidence supporting this gene-disease relationship includes biochemical function of SMARCA4, protein interaction with ADNP and other BAF protein complex subunits, and mouse models recapitulating some of the neurological and cardiovascular phenotypes of CSS (PMIDs: 17878164, 23319608, 23698369, 26459759). SMARCA4 is involved in chromatin remodeling and transcriptional activation. Although pathogenic variants have been shown to alter transcriptional activation of development regulator genes (PMID: 23319608), the mechanism by which SMARCA4 pathogenic variants cause CSS is not clear.

In summary, there is definitive evidence supporting the relationship between SMARCA4 and autosomal dominant Coffin-Siris syndrome. This has been repeatedly demonstrated in both the research and clinical diagnostic settings, and has been upheld over time. This classification was initially approved by the ClinGen Intellectual Disability and Autism Gene Curation Expert Panel (GCEP) on June 17, 2020 (SOP Version 7). This gene-disease relationship was recurated by the ClinGen Congenital Heart Disease GCEP on the meeting date June 9, 2025 with an updated SOP Version 11 to include more probands in the curation, however the classification remained the same (PMIDs: 31530938, 32686290, 38177409).