ClinVar Genomic variation as it relates to human health

Variant summary: ABCC9 c.4570_4572delinsAAAT (p.Leu1524LysfsX5) results in a premature termination codon, predicted to cause a truncation of the encoded protein or absence of the protein due to nonsense mediated decay, which are commonly known mechanisms for disease. This variant was reported as two separate changes ((c.4571_4572delTA (p.Leu1524CysfsTer4) and c.4569_4570insAAA (p.Ile1523_Leu1524insLys)) in gnomAD and the variant allele was found at a frequency of 0.00051 in 283858 control chromosomes, predominantly at a frequency of 0.00085 within the Non-Finnish European subpopulation in the gnomAD database. The observed variant frequency within Non-Finnish European control individuals in the gnomAD database is approximately 54 fold of the estimated maximal expected allele frequency for a pathogenic variant in ABCC9 causing Dilated Cardiomyopathy phenotype (1.6e-05), strongly suggesting that the variant is a benign polymorphism found primarily in populations of Non-Finnish European origin. c.4570_4572delinsAAAT has been reported in the literature in individuals affected with idiopathic Dilated Cardiomyopathy, Brugada syndrome, very early onset Atrial Fibrillation as well as in individuals who were analyzed for arrhythmia gene panel and diagnosis was unknown arrhythmia (Bienengraeber_2004, Hu_2013, Mellor_2017, van Lint_2019, Goodyer_2019). At least one functional study reports experimental evidence evaluating an impact on protein function and this variant results in reducing the catalytic activity of NBD2 and physiological activation of the channel (Bienengraeber_2004). Seven ClinVar submitters (evaluation after 2014) cite the variant as uncertain significance. Based on the evidence outlined above, the variant was classified as uncertain significance until additional clinical and functional evidence become available.

The c.4570_4572delTTAinsAAAT variant, located in coding exon 38 of the ABCC9 gene, results from the deletion of 3 nucleotides and insertion of 4 nucleotides causing a translational frameshift with a predicted alternate stop codon (p.L1524Kfs*5). This variant was reported in two patients with dilated cardiomyopathy, one of whom also harbored a TTN frameshift alteration, an individual with Brugada syndrome, and an individual from a cardiac arrest cohort (Bienengraeber M et al. Nat. Genet., 2004 Apr;36:382-7; Hu D et al. Int. J. Cardiol., 2014 Feb;171:431-42; Cuenca S et al. J. Heart Lung Transplant., 2016 May;35:625-35; Mellor G et al. Circ Cardiovasc Genet. 2017 Jun;10(3):e001686). This alteration has also been reported as a secondary cardiac variant in an exome cohort (Ng D et al. Circ Cardiovasc Genet, 2013 Aug;6:337-46). This variant was detected in a cardiomyopathy/arrhythmia genetic testing cohort; however, clinical details were limited, and additional cardiac variants were detected in some cases (van Lint FHM et al. Neth Heart J, 2019 Jun;27:304-309). In vitro functional studies suggest that this variant causes aberrant K-ATP channel gating, but the impact of this aberrant gating on cardiomyocyte function in vivo is unknown (Bienengraeber M et al. Nat. Genet., 2004 Apr;36:382-7). This alteration is expected to result in loss of function by premature protein truncation. However, loss of function of ABCC9 has not been clearly established as a mechanism of disease, and similar truncating variants have been reported in control populations in the gnomAD dataset. Since supporting evidence is limited at this time, the clinical significance of this alteration remains unclear.

Reported in individuals affected with DCM, Brugada syndrome, or idiopathic cardiac arrest (PMID: 28600387, 15034580, 24439875, 26899768, 35495129); Functional studies suggest that this variant may lead to abnormal ATP-sensitive potassium channel function; however, additional functional evidence and segregation studies are needed to clarify the role of this variant in human disease (PMID: 15034580); Frameshift variant predicted to result in protein truncation, as the last 26 amino acids are replaced with 4 different amino acids in a gene for which loss of function is not an established mechanism of disease; Located in an alternate transcript of the gene.; This variant is associated with the following publications: (PMID: 20033705, 21846889, 24439875, 26899768, 23861362, 30215858, 30847666, 35495129, 28600387, 35276540, 31638414, 15034580, 36129056)

Variant classified as Uncertain Significance - Favor Benign. The p.Leu1524fs var iant in ABCC9 has been reported in 2 individuals with DCM (Bienengraeber 2004, C uenca 2016) and 1 individual with Brugada Syndrome (Hu 2014), but has also been identified in 0.08% (107/126560) of European chromosomes by the Genome Aggregati on Database (gnomAD, http://gnomAD.broadinstitute.org; dbSNP rs139703258 and rs7 61784169). In vitro functional studies provide some evidence that the p.Leu1524f s variant may impact protein function (Bienengraeber et al. 2004); however, thes e types of assays may not accurately represent biological function. This variant is predicted to cause a frameshift, which alters the protein?s amino acid seque nce beginning at position 1524 and leads to a premature termination codon 5 amin o acids downstream. This termination codon occurs within the last exon of the ge ne and is likely to escape nonsense mediated decay, resulting in a truncated pro tein. In summary, while the clinical significance of the p.Leu1524fs variant is uncertain, the frequency data suggests that it is more likely to be benign. ACMG /AMP Criteria applied. BS1, PS3_Supporting.

Based on the classification scheme VCGS_Germline_v1.3.4, this variant is classified as VUS-3C. Following criteria are met: 0103 - Loss of function and gain of function are known mechanisms of disease in this gene. Loss of function variant have been reported to cause dilated cardiomyopathy, 1O (MIM#608569) and ABCC9-related Intellectual disability Myopathy Syndrome (AIMS) (PMID: 31575858). While gain of function variant have been reported to cause Cantu syndrome (MIM#239850). (I) 0107 - This gene is associated with autosomal dominant disease. (I) 0205 - Variant is predicted to result in a truncated protein with less than 1/3 of the protein sequence affected. (SP) 0219 - This variant is non-coding in an alternative transcript. This variant is non-coding in three out of the four ABCC9 transcripts. Exon expression data indicates that this variant is coding in the skeletal muscle transcript and non-coding in the main cardiac tissue transcript (GTEx Portal). (I) 0251 - This variant is heterozygous. (I) 0302 - Variant is present in gnomAD (v2) <0.001 for a dominant condition (145 heterozygotes, 0 homozygotes). (I) 0710 – Other truncating variants comparable to the one identified in this case have inconclusive previous evidence for pathogenicity. Three other truncating variants have conflicting or uncertain interpretations of pathogenicity (ClinVar). (I) 0808 - Previous reports of pathogenicity for this variant are conflicting. This variant has been reported with conflicting interpretations of pathogenicity in individuals with DCM, atrial fibrillation, Brugada syndrome, arrhythmias, and idiopathic cardiac arrest (PMID: 15034580, 26899768, 31638414, 24439875, 23861362, 28600387, 30847666). This variant has seven VUS entries in ClinVar. (I) 0905 - No published segregation evidence has been identified for this variant. (I) 1002 - This variant has moderate functional evidence supporting abnormal protein function. Patch clamp functional studies involving site-directed mutagenesis and Xenopus laevis oocytes, indicated that this variant might cause a structural defect of the potassium channel and aberrant ATP-dependent channel gating (PMID: 15034580). However, patch clamp assays have been shown to be unreliable, therefore results from these studies are used with caution during variant classification. (I) 1206 - This variant has been shown to be paternally inherited (by segregation analysis). (I) Legend: (SP) - Supporting pathogenic, (I) - Information, (SB) - Supporting benign

ABCC9: PM2

iven the lack of case data and weak genotype-phenotype correlation, we also consider this variant to be of uncertain significance and we do not feel it is suitable for assessing risk in healthy relatives ("predictive genetic testing"). The ABCC9 gene is associated with dilated cardiomyopathy and atrial fibrillation. Specific variants have also been associated with Hypertrichotic Osteochondrodysplasia. There is weak case data available for this variant. The variant has been seen in at least 5 unrelated cases of dilated cardiomyopathy (not including this patient's family). We have another patient at our center with this variant who also had early-onset atrial fibrillation (perhaps even as a child) and a cardiac arrest. This other patient also had left ventricular non-compaction (LVNC). This variant has been reported twice in ClinVar, classified as variant of uncertain significance by Blueprint Genetics and the Genome Clinic of Geneva. GeneDx also classifies it as a variant of uncertain significance. A poster by GeneDx indicates that they have seen this variant in 2 patients. Bienengraeber et al. (2004) scanned for mutations in the KATP channel genes of 323 individuals with idiopathic dilated cardiomyopathy. Two heterozygous variants were identified in exon 38 of the ABCC9 gene, one being the c.4570_4572delTTAinsAAAT variant found in our patient. Exon 38 of the ABCC9 gene encodes the C-terminal domain of the protein, which contributes to KATP channel trafficking. Functional studies of these variants found aberrant KATP pore regulation and reduced ATP hydrolytic activities, ultimately reducing the ability of these pores to function properly under stress. The two individuals in this study with ABCC9 variants showed pronounced dilated cardiomyopathy with compromised contractile function (ejection fraction <25%) and ventricular tachycardia. The leucine at codon 1524 is highly conserved across species, as are neighboring amino acids. Other variants (missense and frameshift) have been reported in this codon and a nearby codons (1525); however, none of these have been associated with disease. In total the variant has not been seen in 200 published controls (Hu et al. 2014) and individuals from publicly available population datasets. This variant is not listed in the Genome Aggregation Consortium dataset (http://gnomAD.broadinstitute.org/), which currently includes variant calls on ~140,000 individuals of European, African, Latino, Asian and Ashkenazi Jewish descent (as of 7/2016). However, another frameshift variant at this codon, p.Leu1524Cysfs*4, is present in 139 out of 138,493 individuals in gnomAD (highest MAF is in Europeans at 0.08%). The dataset is comprised of multiple cohorts, some of which were recruited from the general population, others were enriched for common cardiovascular disease. Note that other variants with strong evidence for pathogenicity have been seen at similar frequencies in datasets like this so this does not necessarily rule out pathogenicity (Pan et al 2012). Furthermore, according to the Exome Aggregation Consortium Dataset (http://exac.broardinstitute.org/), which includes ~64,000 variant calls on European, African, Latino and Asian descent, the ABCC9 gene appears fairly tolerant to loss of function variation (pLI=0.00).