ClinVar Genomic variation as it relates to human health

The NM_000152.5:c.266G>A variant in GAA is a missense variant predicted to cause substitution of arginine by histidine at amino acid 89 (p.Arg89His). Two individuals with this variant have been reported to exhibit clinical symptoms of infantile onset Pompe disease, one of them on enzyme replacement therapy (PMID: 22644586 25626711) (PP4_Moderate). One of them is homozygous for c.[266G>A, 546+45G>C] (PMID: 22644586), and the other is compound heterozygous for c.266G>A and c.104T>C (p.Phe35Ser) (PMID: 25626711). The clinical significance of c.[266G>A, 546+45G>C] and c.104T>C (p.Phe35Ser) is unknown. The variant was also found in compound heterozygosity with c.2237G>C (p.Trp746Ser) (LP based on classification by the ClinGen Lysosomal Diseases VCEP) by exome sequencing in a proband, confirmed in trans; three siblings have the same genotype. Three of the siblings have GAA activity consistent with late onset disease, and GAA activity was slightly above this range in the remaining sibling. However, none of them have developed clinical symptoms of Pompe disease yet (see ClinVar Variation ID: 283219; SCV003927052.1). Individuals identified by newborn screen, and so far clinically asymptomatic, have the genotype c.-32-13T>G/c.266G>A; c.1377C>G (PMID: 33202836); or are compound heterozygous for the variant and another variant in GAA that has been classified as pathogenic by the ClinGen LD VCEP, either c.1655T>C (p.Leu552Pro) (PMID: 33202836) or c.525delT (PMID: 28196920). Another individual, with dual molecular diagnoses of GSD 1, and Pompe is homozygous for c.266G>A (p.Arg89His) but no clinical details are available (PMID: 28600779). Therefore, without reports of any symptomatic individuals who have this variant in trans with a variant that has been classified as pathogenic or likely pathogenic by the ClinGen LD VCEP, PM3 is not met at the current time. The highest population minor allele frequency in gnomAD v2.1.1 is 0.001168 in the Ashkenazi Jewish population (12/10272; no homozygotes). The highest population minor allele frequency in a continental population in gnomAD v2.1.1. is 0.00016 (5/30550) in the South Asian population, which is lower than the ClinGen Lysosomal Diseases VCEP threshold (<0.001) for PM2_Supporting, meeting this criterion. The computational predictor REVEL gives a score of 0.819 which is above the threshold of 0.7, evidence that correlates with impact to GAA function (PP3). When expressed in COS cells, this variant was classified as Class E ("presumably nonpathogenic") by Kroos et al, 2012 (PMID:22644586). This includes 43.5% GAA activity in cells and 20% in medium, and is normally synthesized (though less than normal) and processed on Western blot (BS3_Supporting). Three other missense variants at the same position have been reported - c.265C>T (p.Arg89Cys) (VUS based on classification by the ClinGen LD VCEP; ClinVar Variation ID: 456412, SCV002817439.1), c.265C>A (p.Arg89Ser) (ClinVar Variation ID: 1989151), and c.266G>T (p.Arg89Leu) (ClinVar Variation ID: 2754338) (not yet classified by the ClinGen LD VCEP).There is a ClinVar entry for this variant (Variation ID: 283219). In summary, this variant meets the criteria to be classified as a Variant of Unknown Significance for Pompe disease. GAA-specific ACMG/AMP criteria met, based on the specification of the ClinGen Lysosomal Diseases VCEP: PP4_Moderate, PP3, PM2_Supporting, BS3_Supporting. (Classification approved by the ClinGen Lysosomal Diseases Variant Curation Expert Panel on April 16, 2024).

The p.Arg89His variant in GAA has been reported in 3 individuals with Glycogen Storage Disease II (PMID: 22644586, 28600779, 25626711), but it is unclear whether this variant causes Glycogen Storage Disease II. This variant has also been reported as a VUS by Invitae and EGL Genetic Diagnostics in ClinVar (Variation ID: 283219), and has been identified in 0.117% (12/10272) of Ashkenazi Jewish chromosomes, 0.016% (5/30550) of South Asian chromosomes, and 0.015% (19/127034) of European (Non-Finnish) chromosomes by the Genome Aggregation Database (gnomAD, http://gnomad.broadinstitute.org; dbSNP rs200586324). Although this variant has been seen in the general population, its frequency is low enough to be consistent with a recessive carrier frequency. One individual was homozygous for this variant and another reported pathogenic variant for Glycogen Storage Disease I in G6PC, p.Arg83Cys (PMID: 28600779; Variation ID: 11998). Another individual was homozygous for this variant and another VUS reported in the literature, c.546+45G>C (PMID: 22644586). This variant was also seen in trans with a rare VUS in GAA (PMID: 25626711). In vitro functional studies provide some evidence that the p.Arg89His variant may not impact protein function (PMID: 22644586). However, these types of assays may not accurately represent biological function. Computational prediction tools and conservation analyses suggest that this variant may impact the protein, though this information is not predictive enough to determine pathogenicity. In summary, while the clinical significance of the p.Arg89His variant is uncertain, these data suggest that it is more likely to be benign. ACMG/AMP Criteria applied: PP3, BS3_Supporting, BP2 (Richards 2015).

Variant summary: GAA c.266G>A (p.Arg89His) results in a non-conservative amino acid change located in the P-type trefoil domain (IPR000519) of the encoded protein sequence. Five of five in-silico tools predict a damaging effect of the variant on protein function. The variant allele was found at a frequency of 0.00014 in 248182 control chromosomes (gnomAD). It has also been reported at a frequency of 0.0016 in the Turkish population, including one homozygote whose clinical status was not specified (Kars_2021). These frequencies are not significantly higher than expected for a pathogenic variant in GAA causing Glycogen Storage Disease, Type 2 (Pompe Disease) (0.0042), allowing no conclusion about variant significance. c.266G>A has been reported in the literature in individuals affected with Glycogen Storage Disease, Type 2 (Pompe Disease) and also in individuals suspected of the late onset form of the disease, but without a definitive clinical diagnosis (e.g. Hahn_2015, Lin_2017, Nair_2018, Tang_2020, Ficicioglu_2020, Bertoli-Avella_2021). In these reports it has been found in the compound heterozygous state together with pathogenic variants, but also with variants of uncertain significance, and in the heterozygous state where a second allele has not been identified or not specified. Therefore these reports do not provide unequivocal conclusions about association of the variant with Pompe Disease. At least one publication reports experimental evidence evaluating an impact on protein function. It found that the variant results in 43.5% of WT activity in vitro and described the variant as presumably non-pathogenic (Kroos_2012). Nine submitters have provided clinical-significance assessments for this variant to ClinVar after 2014 with conflicting assessments. The majority classified the variant as VUS (n=7) and the other classified the variant as likely pathogenic (n=2). Based on the evidence outlined above, the variant was classified as VUS.

This variant was observed as part of a predisposition screen in an ostensibly healthy population. A literature search was performed for the gene, cDNA change, and amino acid change (where applicable). Publications were found based on this search. However, the evidence from the literature, in combination with allele frequency data from public databases where available, was not sufficient to rule this variant in or out of causing disease. Therefore, this variant is classified as a variant of unknown significance.

The classification of the variant was performed according to the recommendations of ClinGen lysosomal storage disorders variant curation expert panel (Guidelines version 2, specific for the GAA gene). The NM_000152.5(GAA):c.266G>A p.(Arg89His) variant is a missense variant that replaces a highly conserved arginine residue (Arg89) by histidine (p.Arg89His). This variant is present at an extremely low frequency in gnomAD databases (ƒExomes = 0.0000141, ƒgenomes = 0.0000657), only in heterozygosity and meets the PM2_Supporting criterion (1pt). The c.266G>A variant confirmed to be in trans with the pathogenic variant NM_000152.5:c.2237G>C p.(Trp746Ser) in our case. Moreover, it has been identified in compound heterozygosity with the c.546+45G>C variant in individuals with clinical features of Pompe disease [PMID: 22644586], thus meets the PM3_strong criterion (2pts). The REVEL score is 0.819 which is higher than the LSD VCEP threshold for PP3 (>0.7) and therefore meets this criterion (1pt). An alternative variant NM_000152.5(GAA):c.265C>A (p.Arg89Ser) is classified as Likely Pathogenic [1 star, ClinVar, PM5_supporting (1pt)]. The four individuals, with the c.[266G>A];[2237G>C] genotype (2-13years), were found to have GAA activity clearly within the patient range (1.1-4.9nmol/h/mg) in leukocytes using 4-methylumbelliferyl-α-D-glucoside as a substrate. Using glycogen as a substrate, the GAA activity was found to be within the range for LOPD patients (3-10nmol/h/mg) for one subject and slightly above the upper limit for the three subjects. The presence of the KM-mutant polymorphism c.271G>A and the pseudodeficiency variant c.[1726G>A (p.Gly576Ser); 2065G>A (p.Glu689Lys)] was excluded. Based on the ClinGen LSD VCEP's specifications, this data meets PP4_Moderate rule (2pts). In summary, this variant meets the criteria to be classified as a likely pathogenic variant (total 7pts) for Pompe disease.

PP3, PP4, PM2

This sequence change replaces arginine, which is basic and polar, with histidine, which is basic and polar, at codon 89 of the GAA protein (p.Arg89His). This variant is present in population databases (rs200586324, gnomAD 0.1%). This missense change has been observed in individuals with clinical features of Pompe disease (PMID: 22644586, 25626711, 32860008, 34602496; Invitae). ClinVar contains an entry for this variant (Variation ID: 283219). Advanced modeling of protein sequence and biophysical properties (such as structural, functional, and spatial information, amino acid conservation, physicochemical variation, residue mobility, and thermodynamic stability) performed at Invitae indicates that this missense variant is expected to disrupt GAA protein function with a positive predictive value of 95%. Experimental studies are conflicting or provide insufficient evidence to determine the effect of this variant on GAA function (PMID: 22644586). For these reasons, this variant has been classified as Pathogenic.

Reported in the homozygous state in the presence of an additional homozygous GAA variant (c.-32-13T>G, a common pathogenic variant in individuals with adult-onset GSDII) in an infant severely affected with GSDII; functional studies indicated only minimally reduced enzyme activity and the R89H variant was classified as presumably non-pathogenic by the authors (Kroos M et al., 2012); Reported in the homozygous state in a patient with a dual diagnosis of GSDII and GSDI who also harbored a homozgyous pathogenic variant in the G6PC gene (Monies D et al., 2017); Reported in additional patients with GSDII with either a variant of uncertain significance in trans or for whom a second variant was not identified (Hahn A et al., 2015; Nair P et al., 2018; Bertoli-Avella AM et al., 2021); In silico analysis supports that this missense variant has a deleterious effect on protein structure/function; Reported in ClinVar (ClinVar Variant ID# 283219; ClinVar); This variant is associated with the following publications: (PMID: 25626711, 30293248, 31342611, 33202836, 32860008, 22644586, 27927596, 28600779, 34426522, 34602496, Goomber2022[abstract], 33073007, 31254424)