Glycogen storage disease type II (GSDII), also termed acid maltase deficiency (AMD) or Pompe disease, is an inherited disorder of glycogen metabolism resulting from defects in activity of the lysosomal hydrolase acid α-glucosidase in all tissues of affected individuals. The disorder is transmitted as an autosomal recessive trait.

The clinical presentation of GSDII encompasses a range of phenotypes, all of which include varying degrees of myopathy but differ with respect to age of onset, extent of organ involvement, and rate of progression to death. The most severe is the classic infantile-onset disease, described by Pompe and delineated prior to discovery of the deficiency of acid maltase, with prominent cardiomegaly, hypotonia, hepatomegaly, and death due to cardiorespiratory failure, usually before 2 years of age. At the other extreme is a slowly progressive proximal myopathic adult-onset disease with onset as late as the second to sixth decade and involvement essentially only of skeletal muscle. Between these two extremes there is a heterogeneous group, variously termed childhood, juvenile, or muscular variant, generally with onset after early infancy, a predominance of skeletal muscle involvement, usually without cardiac involvement, and a more slowly progressive course as compared with classic infantile-onset Pompe disease. Progressive proximal muscle weakness including major impairment of respiratory function dominates the picture and death results usually from respiratory failure.

The enzyme deficiency results in intralysosomal accumulation of glycogen of normal structure in numerous tissues. The accumulation is most marked in cardiac and skeletal muscle and in hepatic tissues of infants with the generalized disorder. In late-onset GSDII, intralysosomal accumulation of glycogen is virtually limited to skeletal muscle and is of lesser magnitude. Electromyographic abnormalities suggestive of the diagnosis include pseudomyotonic discharges and irritability, but in juvenile- and adult-onset patients, the abnormalities can vary in different muscles. CAT scans can reveal the site(s) of affected muscles. Most patients have elevated blood plasma levels of creatine kinase (CK) and elevations in hepatic enzymes, particularly in adult-onset patients, can be found.

The clinical diagnosis is confirmed by virtual absence (infantile-onset) or markedly reduced activity (late-onset) of acid α-glucosidase in muscle biopsies and cultured fibroblasts. Purified lymphocytes and lymphoid cell lines also exhibit the enzyme abnormalities, but have been less extensively used for diagnosis, and misdiagnosis may occur with imperfectly fractionated peripheral blood lymphocytes. Assay of unfractionated leukocytes is not reliable in most laboratories unless combined with use of antibodies. In general, the amount of residual enzyme activity correlates inversely with the severity of disease.

Prenatal diagnosis has been made by determination of acid α-glucosidase deficiency in cultured amniotic cells and chorionic villus biopsies. DNA analysis is of supportive diagnostic value and provides a means for definitive carrier detection.

Both the human cDNA and structural gene have been isolated, characterized, and used to analyze mutations. The cDNA for human acid α-glucosidase is over 3.6 kb, with 2859 nucleotides (nt) of coding sequence, a 3′ untranslated region of 550 bp, and a 5′ untranslated region of at least 218 bp. The coding region predicts 952 amino acids in a nonglycosylated protein of 105.37 kDa, with an amino-terminal signal sequence for synthesis within the ER. The enzyme is extensively modified posttranslationally by glycosylation within the ER at all seven predicted sites, subsequent remodeling of the asparagine-linked carbohydrate and phosphorylation of mannose residues, providing the mannose-6-phosphate recognition marker for targeting to lysosomes. The enzyme is additionally modified by both amino- and C-terminal proteolytic cleavage, primarily within lysosomes. The structural gene contains 20 exons in approximately 20 kb of genomic DNA and has been localized to human chromosome 17q25 and designated GAA.

The disorder is genetically heterogeneous, with missense, nonsense, and splice-site mutations, partial deletions, and insertions described in the relatively small number of patients analyzed. A few mutations have been described that are common in particular ethnic groups, including Chinese and African-Americans. Additionally a single mutation is responsible for adult-onset disease in over half of Caucasian adult-onset patients.

Definitive therapy is not currently available, although supportive measures, particularly of respiratory function, can have a significant impact on patients with the late-onset disease. Dietary therapy has been used in late-onset patients. Trials of enzyme replacement therapy have begun. The first published results of a phase 2 trial with recombinant human α-glucosidase from milk of transgenic rabbits have shown some therapeutic effects. Possibilities for gene therapy are under investigation.

There are several naturally occurring animal models of the infantile- and late-onset disease. Knockout mice have now also been developed. Ameliorative effects of enzyme therapy have been described in knockout mice and in a quail model.

There is a second syndrome associated with intravacuolar accumulation of glycogen but normal acid α-glucosidase activity, characterized by prominent cardiac abnormalities, involvement of skeletal muscle, and variable mental retardation. This syndrome may encompass more than one mode of inheritance. The X-linked inherited Danon disease is caused by a primary deficiency of the lysosomal membrane protein LAMP-2.