The NM_000152.5:c.271G>A missense variant in GAA (p.Asp91Asn) has a highest population minor allele frequency in gnomAD v2.1.1 of 0.03291 (4183/127102 alleles; 70 homozygotes) in the European non-Finnish population, which is higher than the ClinGen LSD VCEP’s threshold for BA1 (>0.01), and therefore meets this criterion (BA1). The computational predictor REVEL gives a score of 0.396 which is below the threshold of 0.5, evidence that does not predict a damaging effect on GAA function. In addition, the computational splice predictors SpliceAI and varSEAK predict that the variant has no impact on splicing (BP4). This variant is frequently described as a “pseudodeficiency” variant in the literature. It has been identified as the amino acid substitution responsible for the GAA*2 alloenzyme (also known as GAA2) (PMID: 2203258). GAA*2 was identified by its faster migration on starch gel electrophoresis compared to the GAA*1 alloenzyme (PMID: 2688539). In addition, compared to GAA*1, GAA*2 has a 10 fold higher Km for glycogen, but similar Km for maltose and 4-MUG (PMID: 2688539). No differences between the processing of GAA*2 and GAA*1 were noted on Western blot of protein from fibroblasts homozygous for either alloenzyme (PMID: 2688539). Furthermore, comparable amounts of glycogen were found in the lysosomes of the GAA*1 and GAA*2 homozygous cell lines, and there was no evidence of abnormal lysosomal glycogen storage in GAA*2 fibroblasts on electron microscopy (PMID: 2688539). When expressed in SV40-transformed GAA-deficient fibroblasts, c.271G>A (p.Asp91Asn) has similar GAA activity to wild type when 4-MUG is used as the substrate (PMID: 2203258). Studies in cells from patients show a difference in GAA activity depending on the substrate used. In cells from individuals who are homozygous, or compound heterozygous for the variant and a known pathogenic GAA variant, false positives arose when glycogen was used as the substrate in the leukocyte assay, while the expected result was obtained when 4-MUG was used as the substrate in either the leukocyte or fibroblast assays (PMID: 19387865, 33162552). Evidence from the crystal structures of native and recombinant GAA indicate that Asp91 is involved in a sugar binding site, which could explain why this variant binds carbohydrate less effectively (PMID: 29061980; https://www.biorxiv.org/content/10.1101/212837v1 ). In addition, while the computational splice site predictors SpliceAI and varSEAK predict no impact on splicing, a mini-gene assay indicated that the variant results in exon 2 exclusion in about 30% of transcripts (PMID: 31301153). As an explanation for how the biochemical features of this variant result in no clinical features, substrate concentrations are usually far below that required for saturation of the enzyme. A reduction in enzyme activity brought about by decrease in Vmax, or increase in Km, would be compensated for by an increase in substrate concentration, restoring the normal turnover rate and establishing an equilibrium substrate concentration (PMID: 2688539). In patients with Pompe disease, c.271G>A has been found in cis with other variants that have been classified as pathogenic by the ClinGen LSD VCEP, p.Cys103Gly (PMID: 16838077, 33073003) and c.877G>A (p.Gly293Arg) (PMID: 27189384).There is a ClinVar entry for this variant (Variation ID: 4020) with eight submitters classifying the variant as benign, two as likely benign, and two as “other”. In summary, based on the evidence available to date, this variant meets the criteria to be classified as benign for Pompe disease. However, it is important to note that GAA activity may be in the patient range in individuals who are homozygous for the variant, or compound heterozygous for the variant and a pathogenic GAA variant, if glycogen is used as a substrate. Careful evaluation and use of 4-MUG as the substrate is necessary to provide accurate results. Therefore, this variant is classified as “other” to ensure that an explanation for deficient enzyme activity in the context of a clinically benign variant is highlighted. GAA-specific ACMG-AMP criteria applied, as specified by the ClinGen LSD VCEP: BA1.