ClinVar Genomic variation as it relates to human health

ACMG categories: PS3,PM1,PM3,PP3

A homozygous missense variant was identified, NM_000153.3(GALC):c.334A>G in exon 4 of the GALC gene. This substitution is predicted to create a minor amino acid change from a threonine to an alanine at position 112 of the protein; NP_000144.2(GALC):p.Thr112Ala. The threonine at this position has very high conservation (100 vertebrates, UCSC), and is located within the Glycosyl hydrolase family 59 functional domain (NCBI, PDB). In silico software predicts this variant to be damaging (PolyPhen2, SIFT, CADD, Mutation Taster). The variant is present in the gnomAD population database at a global population frequency of 0.25% (701 heterozygotes, 4 homozygotes) with a European sub-population frequency of 0.40%. This variant has been previously reported as likely pathogenic in patients with Krabbe disease (Nashabat, M., et al. (2019)), and as a VUS (ClinVar, Orsini, J. J., et al. (2016)). It has also been shown to segregate with disease in one family (Shao, Y. H. et al. (2016)). In addition, functional studies show reduced GALC activity but the effect of this reduction is unknown (Orsini, J. J., et al. (2016), Saavedra-Matiz, C. A. et al. (2016)). Based on information available at the time of curation, this variant has been classified as a VUS.

The GALC c.334A>G (p.Thr112Ala) variant, which was previously referred to as p.Thr96Ala, is a missense variant that has been reported in association with Krabbe disease in several studies. Shao et al. (2016) reported the variant in five members of a French-Canadian family who all presented with adult-onset disease with predominant cerebellar ataxia and mild spasticity. Nashabat et al. (2019) described a 6-year-old female with early-onset Krabbe disease who was homozygous for the p.Thr112Ala variant and showed low galactocerebrosidase enzyme activity in cultured fibroblasts but normal levels in peripheral blood. This variant was also found to be enriched among infants with GALC activity <12% of normal who were identified through newborn screening (Orsini et al. 2016) and has also been reported in a compound heterozygous state in several affected individuals who were also heterozygous for common enzyme activity-lowering variants, including p.Ile562Thr (previously p.Ile546Thr) (Luzi et al. 1996; Debs et al. 2013). Notably, this variant is reported at a frequency of 0.004022 in the European (non-Finnish) population of the Genome Aggregation Database, which also includes four homozygous carriers. Functional studies in COS1 cells showed the p.Thr112Ala-containing enzyme showed approximately 30% of wild type activity; residual activity was even lower when an additional pseudodeficiency variant was co-expressed (Luzi et al. 1996; Saavedra-Matiz et al. 2016). Marshall et al. (2018) asserted that the p.Thr112Ala variant should be considered a disease-causing allele when in cis with a common enzyme activity reducing variant or when in trans with a severe allele. Based on the collective evidence and application of the ACMG criteria, the p.Thr112Ala variant is classified as likely pathogenic for Krabbe disease.

Reported previously in association with later-onset Krabbe disease in two individuals, one of whom had 5 additional coding variants in GALC and the second had 2 additional coding variants, the phase of the other variants in these cases was unknown (PMID: 8687180, 23197103); In silico analysis supports that this missense variant has a deleterious effect on protein structure/function; This variant is associated with the following publications: (PMID: 27276562, 24082139, 23891399, 27533158, 30609409, 23197103, 27638593, 23509109, 30202406, 31664448, 29481565, 31053700, 24388568, 26915362, 34426522, 32860008, 34065072, 31980526, 8687180, 26795590, 33742171, 36964991, 36781956, 34531397, 32483926, 31847883)

BS1, PP3

Variant summary: The variant, GALC c.334A>G (p.Thr112Ala), also reported as p.Thr96Ala in the literature, results in a non-conservative amino acid change in the encoded protein sequence. Four of five in-silico tools predict a damaging effect of the variant on protein function. The variant allele was found at a frequency of 0.0025 in 248976 control chromosomes, predominantly at a frequency of 0.0041 within the Non-Finnish European subpopulation in the gnomAD database, including 1 homozygote. The observed variant frequency within Non-Finnish European control individuals in the gnomAD database is approximately 2 fold of the estimated maximal expected allele frequency for a pathogenic variant in GALC causing Krabbe Disease phenotype (0.0022), strongly suggesting that the variant is a benign polymorphism found primarily in populations of Non-Finnish European origin. The variant, c.334A>G has been reported in the literature in multiple individuals affected with adult onset cases of Krabbe Disease (Globoid Cell Leukodystrophy (GCL), Debs _2013, Luzi _1996 ). In one French Canadian family, the variant was observed in a compound heterozygous genotype with co-segregation across two generations in 5 patients who were diagnosed with adult onset of Krabbe disease (Shao _2016). In newborn screening done for Krabbe disease at New York State, the infants carrying this variant were classified diagnostically in moderate, low, and no-risk categories, but none was classified as being at high risk (Orsini _2016). These data indicate that the variant is likely to be associated with disease with adult onset of Krabbe disease. Recently, this variant has also been reported in the literature in two homozygous individuals diagnosed with Krabbe disease (Alfares_2018, Nashabat_2019). One of these patients had infantile onset of the disease (Nashabat_2019). However, both reports do not unequivocally confirm the biochemical and clinical features of this disease either due to non-reporting of testing information in the patient reported by Alfares_2018 OR normal CSF protein levels, equivocal MRI measurements, conflicting measurements of enzyme activity in blood (normal activity) and fibroblasts (moderate reduction) and no information regarding psychosine levels in the patient reported by Nashabat_2019. At least one publication reports experimental evidence evaluating an impact on protein function. The most pronounced variant effect results in <30% of normal activity (Saavedra-Martiz_2016). In the same study, the authors also reported the presence of this variant (reported as p.T96A) in a compound heterozygous patient (among confirmed cases of Krabbe's disease) with two other severe mutations, namely p.R53X and p.R204X. This implies that p.T96A is in cis with one of these two severe mutations in this specific patient with Krabbe disease, supporting a non-pathogenic outcome for this variant. Furthermore, the authors conclude that the pathogenicity of this variant in late-onset Krabbe disease is related to its presence with a haplotype in cis with another variant, p.I546T. Fifteen ClinVar submitters (evaluation after 2014) cite this variant as uncertain significance (n=10), likely pathogenic (n=4) or likely benign (n=2). Based on the evidence outlined above, to reflect the variable expressivity and variable penetrance that could be associated with this variant in different genetic backgrounds, it was classified as a VUS-possibly pathogenic

This sequence change replaces threonine, which is neutral and polar, with alanine, which is neutral and non-polar, at codon 112 of the GALC protein (p.Thr112Ala). This variant is present in population databases (rs147313927, gnomAD 0.4%), and has an allele count higher than expected for a pathogenic variant. This variant has been reported in combination with other GALC variants in individuals affected with late-onset Krabbe disease with evidence of disease co-segregation in one family (PMID: 23197103, 26915362). This variant is also one of the most common variants that is detected in newborns who are referred for low galactocerebrosidase activity on newborn screening, but interestingly, none of the newborns were categorized as high risk for Krabbe disease (PMID: 26795590). One study has suggested that the p.Thr112Ala variant may become a disease-associated allele only when it co-occurs on the same chromosome (in cis) with the p.Ile562Thr polymorphism (PMID: 27638593). However, the evidence is not sufficient to support a pathogenic haplotype at this time. It has also been observed to segregate with disease in related individuals. ClinVar contains an entry for this variant (Variation ID: 92503). 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) has been performed at Invitae for this missense variant, however the output from this modeling did not meet the statistical confidence thresholds required to predict the impact of this variant on GALC protein function. Experimental studies have shown that this missense change alone causes a mild reduction of GALC enzyme activity in vitro and the effect is exacerbated in the presence of other GALC variants in cis (PMID: 27638593). In summary, while this variant has been reported in individuals with Krabbe disease in the literature, there are multiple homozygous individuals observed in population databases. In addition, clinical and experimental data suggest that co-occurrence in cis with other GALC variant(s) may be required for this variant to be disease-causing (PMID: 27638593). The available evidence is currently insufficient to determine the role of this variant in disease. Therefore, it has been classified as a Variant of Uncertain Significance.

This sequence change is predicted to replace threonine with alanine at codon 112 of the GALC protein (p.(Thr112Ala), also known as p.Thr96Ala). The threonine residue is evolutionarily conserved (100 vertebrates, UCSC), and is located in the Glycosyl hydrolase family 59 domain in a triosephosphate isomerase barrel (PMID: 21876145). There is a small physicochemical difference between threonine and alanine. The variant is present in a large population cohort at a frequency of 0.25% (rs147313927, 709/280,348 alleles, 4 homozygotes in gnomAD v2.1), with an allele frequency of 0.4% in the European (non-Finnish) population (BS1; gnomAD v2.1). The variant has been identified in the homozygous state and compound heterozygous with a second allele in at least three individuals with Krabbe disease, and reported to segregate with adult-onset predominant cerebellar ataxia with mild spasticity in five affected individuals (PP1_Strong, PM3; PMID: 8687180, 23197103, 26915362, 31053700). The variant has been reported with reduced enzyme activity in patient cells and in in vitro assays. However, there is in vitro and patient enzyme activity evidence that occurrence of the pseudoefficiency variant p.Ile562Thr is required in cis with the variant to produce disease causing GALC deficiency (PMID: 8687180, 26795590, 27638593). Multiple lines of computational evidence predict a deleterious effect for the missense substitution (PP3; 5/6 algorithms). Based on the classification scheme RMH ACMG Guidelines v1.2.1, this variant is classified as a VARIANT OF UNCERTAIN SIGNIFICANCE. Following criteria are met: PP1_Strong, PM3, BS1, PP3.

The p.Thr112Ala (NM_000153.3 c.334A>G, also referred to as p.Thr96Ala) variant in GALC has been reported in 4 compound heterozygous individuals with clinical features of late-onset Krabbe disease (Luzi 1996 PMID: 8687180, Debs 2013 PMID: 23197103, and Shao 2016 PMID: 26915362), and segregated in 3 affected siblings (Shao 2016 PMID: 26915362). This variant is also the most common referral for followup testing for Krabbe disease by newborn screening programs in the heterozygous, homozygous, and compound heterozygous states (Orsini 2016 PMID: 26795590). Upon further testing, these newborns were all classified into the moderate-, low-, or no-risk categories. This variant has been identified in 0.4% (508/126254) of European chromosomes and 4 homozygotes by the Genome Aggregation Database (gnomAD, http://gnomad.broadinstitute.org/; dbSNP rs147313927). This variant has also been reported in ClinVar (Variation ID 92503). In vitro functional studies provide evidence that the p.Thr112Ala variant leads to reduced enzymatic activity, though not at levels as low as variants causing infantile-onset disease (Saavedra-Matiz 2016 PMID: 23509109, Orsini 2016 PMID: 26795590). Computational prediction tools and conservation analysis suggest that the p.Thr112Ala variant may impact the protein, though this information is not predictive enough to determine pathogenicity. In summary, while there is some suspicion for a pathogenic role, the clinical significance of this variant is uncertain and although at least one individual affected during childhood has been reported, has typically been associated with late-onset disease. ACMG/AMP Criteria applied: BS1, PS3_Moderate, PM3, PP1, PP3.

A heterozygous variant in exon 4 of the GALC gene that results in the amino acid substitution of Alanine for Threonine at codon 112 was detected. The observed variant c.334A>G (p.Thr112Ala) has not been reported in the 1000 genomes database and has MAF of 0.2526% in gnomAD database. The in silico prediction is damaging by MutationTaster and DANN. In summary, the variant meets our criteria to be classified as likely pathogenic.

GALC: PS3:Moderate, PP3

The GALC p.T112A variant was identified in 106 of 1306 proband chromosomes (frequency: 0.08) from individuals with late-onset Krabbe disease (Orsini_2016_PMID:26795590; Debs_2013_PMID:23197103). This variant was also found to segregate with disease in one family with late-onset Krabbe disease with predominant cerebellar ataxia in five affected individuals, all of whom were compound heterozygous for the GALC p.T112A and p.E198K variants (Shao_2016_PMID:26915362). In many cases, the p.T112A variant is also found in cis with another GALC variant, such as the p.I1546T variant (Luiz_1996_PMID:8687180; Orsini_2016_PMID:26795590). The variant was identified in dbSNP (ID: rs147313927) and ClinVar (classified as uncertain significance by GeneDx, EGL Genetic Diagnostics, Invitae and Mayo Clinic, as likely pathogenic by Laboratory for Molecular Medicine and as likely benign by Mendelics). The variant was identified in control databases in 709 of 280348 chromosomes (4 homozygous) at a frequency of 0.002529 (Genome Aggregation Database March 6, 2019, v2.1.1). The variant was observed in the following populations: European (non-Finnish) in 516 of 128292 chromosomes (freq: 0.004022), European (Finnish) in 65 of 25000 chromosomes (freq: 0.0026), South Asian in 60 of 30568 chromosomes (freq: 0.001963), Other in 14 of 7138 chromosomes (freq: 0.001961), Latino in 34 of 35290 chromosomes (freq: 0.000963), African in 18 of 24176 chromosomes (freq: 0.000745), Ashkenazi Jewish in 1 of 10352 chromosomes (freq: 0.000097), and East Asian in 1 of 19532 chromosomes (freq: 0.000051). The p.T112 residue is conserved in mammals and computational analyses (PolyPhen-2, SIFT, AlignGVGD, BLOSUM, MutationTaster) provide inconsistent predictions regarding the impact to the protein; this information is not very predictive of pathogenicity. The variant occurs outside of the splicing consensus sequence and in silico or computational prediction software programs (SpliceSiteFinder, MaxEntScan, NNSPLICE, GeneSplicer) do not predict a difference in splicing. A functional GALC expression study conducted on this variant using COS1 cells found reduced GALC activity compared to the wild type, with GALC activity even more significantly reduced when another GALC variant is present with the p.T112A variant (Saavedra-Matiz_2016_PMID:27638593). In summary, this variant is considered a hypomorphic allele which, in combination with other variants in cis or trans, may result in reduced activity. Based on the above information the clinical significance of this variant cannot be determined with certainty at this time although we would lean towards a more pathogenic role for this variant. This variant is classified as likely pathogenic.