ClinVar Genomic variation as it relates to human health

Published functional studies demonstrate a damaging effect: reduced kinase activity compared to wild type (PMID: 19431188, 22071889); Observed with a second ATM variant in patients with clinical features of ataxia-telangiectasia but it is not known whether the variants occurred on the same (in cis) or opposite (in trans) alleles (PMID: 22071889, 30549301); In silico analysis supports that this missense variant has a deleterious effect on protein structure/function; Not observed at significant frequency in large population cohorts (gnomAD); This variant is associated with the following publications: (PMID: 22071889, 19781682, 23532176, 30549301, 33471991, 28975465, 19431188, 30303537, 35892882, 32295079, 26896183, 35893033, 33280026, 39669589, 38017116, 38313678)

This sequence change replaces glutamic acid, which is acidic and polar, with lysine, which is basic and polar, at codon 2039 of the ATM protein (p.Glu2039Lys). This variant is present in population databases (no rsID available, gnomAD 0.003%). This missense change has been observed in individuals with ataxia-telangiectasia and/or breast cancer (PMID: 19781682, 22071889, 26896183, 30303537, 30549301; internal data). ClinVar contains an entry for this variant (Variation ID: 219787). Invitae Evidence Modeling of protein sequence and biophysical properties (such as structural, functional, and spatial information, amino acid conservation, physicochemical variation, residue mobility, and thermodynamic stability) indicates that this missense variant is not expected to disrupt ATM protein function with a negative predictive value of 95%. Experimental studies have shown that this missense change affects ATM function (PMID: 19431188, 22071889). For these reasons, this variant has been classified as Pathogenic.

The p.E2039K variant (also known as c.6115G>A), located in coding exon 41 of the ATM gene, results from a G to A substitution at nucleotide position 6115. The glutamic acid at codon 2039 is replaced by lysine, an amino acid with similar properties. This variant has been identified in multiple individuals diagnosed with breast cancer (Tavtigian S et al. Am J Hum Genet. 2009 Oct;85(4):427-46; Siraj AK et al. Hum Genet, 2017 11;136:1431-1444; Girard E et al. Int J Cancer, 2019 04;144:1962-1974; Dorling et al. N Engl J Med. 2021 02;384:428-439; BenAyed-Guerfali D et al. Genes (Basel), 2022 Jul;13:; Bu R et al. Sci Rep, 2023 Nov;13:20924; Boujemaa M et al. Front Genet, 2024 Jan;15:1327894). Additionally, this variant has been identified in individuals diagnosed with prostate cancer (Mondschein R et al. Cancers (Basel), 2022 Jul;14:). This variant has been identified in the homozygous state and likely in trans with other ATM variants in individuals diagnosed with ataxia telangiectasia (Jacquemin V et al. Eur J Hum Genet, 2012 Mar;20:305-12; Schon K et al. Ann Neurol, 2019 02;85:170-180). Additional investigation has demonstrated stable, normally-localized ATM protein, but reduced kinase activity associated with p.E2039K compared to wild type (Barone G et al. Hum. Mutat. 2009 Aug; 30(8):1222-30. Jacquemin V et al. Eur. J. Hum. Genet. 2012 Mar; 20(3):305-12). This variant is considered to be rare based on population cohorts in the Genome Aggregation Database (gnomAD). This amino acid position is highly conserved in available vertebrate species. In addition, the in silico prediction for this alteration is inconclusive. Based on the majority of available evidence to date, this variant is likely to be pathogenic.

This missense variant replaces glutamic acid with lysine at codon 2039 of the ATM protein. Computational prediction is inconclusive regarding the impact of this variant on protein structure and function. Functional studies have reported that this variant does not affect protein expression but causes a reduction in kinase activity (PMID: 19431188). In a large international case-control study, this variant was reported in 2/60464 breast cancer cases and absent in 53461 controls (PMID: 33471991). This variant has been reported in additional individuals affected with breast cancer (PMID: 19781682, 28975465, 30303537, 35893033, 38017116) as well as in individuals affected with prostate cancer (PMID: 35892882). This variant has also been reported in an individual affected with variant ataxia-telangiectasia with a co-occurring pathogenic variant, c.8609_8610delAT (p.Asp2870GlufsTer10), in unknown phase (PMID: 26896183, 30549301). This variant was identified in a second individual affected with ataxia-telangiectasia with a co-occurring variant of uncertain significance, c.6491A4C (p.Glu2164Ala), in unknown phase (PMID: 22071889). This variant has been identified in 1/251392 chromosomes in the general population by the Genome Aggregation Database (gnomAD). Based on the available evidence, this variant is classified as Likely Pathogenic.

This variant is considered likely pathogenic. This variant has been reported in multiple individuals with clinical features of gene-specific disease [PMID: 22071889, 30549301, 27142713]. Functional studies indicate this variant impacts protein function [PMID: 19431188, 30549301].

The ATM c.6115G>A variant is predicted to result in the amino acid substitution p.Glu2039Lys. This variant has been observed in multiple individuals with breast cancer (Table S2 - Tavtigian et al. 2009. PubMed ID: 19781682; Table S4 - Siraj et al. 2017. PubMed ID: 28975465). Additionally, this variant has been reported with another variant in the compound heterozygous state in an individual with ataxia-telangiectasia (Jacquemin et al. 2012. PubMed ID: 22071889). Functional studies demonstrated reduced kinase activity (Barone et al. 2009. PubMed ID: 19431188) and defective response to ionizing radiation (Jacquemin et al. 2012. PubMed ID: 22071889). This variant is reported in 0.0033% of alleles in individuals of South Asian descent in gnomAD and has conflicting interpretations of pathogenicity of uncertain and likely pathogenic in ClinVar (https://www.ncbi.nlm.nih.gov/clinvar/variation/219787/). This variant is interpreted as likely pathogenic.

The c.6115G>A (p.Glu2039Lys) variant appears only once in the gnomAD v2.1.1 non-cancer dataset (0.0004% frequency), specifically in the South Asian subpopulation (PM2; http://gnomad.broadinstitute.org). This missense variant is not predicted to lead to a splicing alteration as per SPiCE predictor and no splicing site is created/activated according to at least 3 splicing predictors of the set SpliceSiteFinderlike - MaxEntScan - NNSplice – GeneSplicer, but it alters the protein function / structure on the in-silico prediction reports of REVEL and PROVEAN (PP3). Functional studies with the variant protein modelled in an ATM-null lymphoblastoid cell line and proper controls showed that the protein was stable but but retained relatively little kinase activity (no functional criterion met; PMID: 19431188). The variant was found in one ataxia telangiectasia proband with another missense variant (PS4_Supporting; PMID: 22071889). There is no other supporting data that meet criteria for consideration. Therefore, the clinical significance of this variant is uncertain. Adapted ACMG/AMP rules applied as defined by the Spanish ATM working group: PM2 + PP3 + PS4_Supporting (PMID: 33280026).