KdVS is caused by a 500- to 650-kb heterozygous deletion at chromosome 17q21.31 that includes KANSL1 or a heterozygous intragenic mutation in KANSL1.
Mutations in KANSL1 are scattered throughout the gene (NM_001193466.1) and are predicted protein truncation, including nonsense variants, frame shift variants, and canonical splice site variants.
Genetic testing of the 17q21.31 genomic region can be challenging. The mapping and interpretation of the deletion breakpoints are confounded by the structural complexity and genomic variation of the 17q21.31 locus. The resolution of the different array platforms, for example, is limited because of the low probe coverage of the segments that are rich in LCRs. Moreover, the region consists of two (benign) copy-number polymorphic (CNP) segmental duplications, including a short duplication (CNP155 or α polymorphism duplication) and a long duplication (CNP205 or β polymorphism duplication) corresponding to the promoter and first exons of KANSL1. These CNPs are associated with the H2 and H1 haplotypes, respectively, and are significantly enriched in European/Mediterranean populations where frequencies can reach as high as 60%. These common copy number variants (CNVs) are listed in large numbers in the various databases as inherited and/or benign (ISCA benign variants (http://dbsearch.clinicalgenome.org/search/); Decipher (http://decipher.sanger.ac.uk); Database of Genomic Variants (http://dgv.tcag.ca/) and can give rise to ‘artefacts’ in genetic testing.
In addition to difficulties with the standard CNV analysis, the H2 haplotype and the CNP clusters encompassing exons 1–3 of KANSL1 contribute to difficulties in SNV calling such as loss-of-function variant ‘artefacts’ in the KANSL1 gene. As a result, in a number of variant databases, we have observed loss-of-function variants reported in non-affected individuals (the Exome Aggregation Consortium,http://exac.broadinstitute.org/; NHLBI GO Exome Sequencing Project, http://evs.gs.washington.edu/EVS/) corresponding specifically to the duplicated portion. This means that the detection of a truncating variant in exons 1–3 of the KANSL1 gene is not sufficient to make a diagnosis of KdVS. In these cases, next to a compatible clinical phenotype, variant analysis of the parental samples is of the utmost importance to make sure that the possibly pathogenic variant has originated ‘de novo’.