Abstract
Background
Pompe disease (PD) is a rare autosomal recessive disorder caused by pathogenic variants in the GAA gene, resulting in deficient lysosomal acid α-glucosidase activity. Clinical manifestations range from classic infantile-onset (IOPD) to late-onset (LOPD) phenotypes. Understanding genotype–phenotype correlations in PD is essential for prognosis and individualized therapy.
Methods
This retrospective, single-center study included 26 Vietnamese pediatric patients diagnosed with PD. Clinical, biochemical, and genetic data were systematically collected. Genotype–phenotype correlation, CRIM status distribution, and survival outcomes were analyzed.
Results
Of 26 patients, 23 had IOPD and 3 had LOPD. CRIM-positive status was identified in 87.0% of IOPD and 33.3% of LOPD patients. The most frequent variants were c.1843G > A and c.1933G > C. Two previously unreported variants (c.2016del, c.1723T > C) were detected. Hypertrophic cardiomyopathy and hypotonia were universal among IOPD cases. Despite ERT administration in 52.9% of patients, overall mortality in the infantile group was 60.8%. Variant pathogenicity correlated with both CRIM status and clinical outcomes.
Conclusion
These findings underscore the clinical importance of comprehensive GAA genotyping and CRIM status determination in predicting disease course and guiding therapeutic decisions. Our results further emphasize the need for population-specific variant databases to inform newborn screening and precision treatment initiatives across Southeast Asia.
Keywords: Pompe disease, GAA mutations, cross-reactive immunologic material (CRIM), enzyme replacement therapy (ERT), genotype–phenotype correlation
PLAIN LANGUAGE SUMMARY
Pompe disease is a rare, inherited illness caused by changes in a gene called GAA. These GAA changes stop the body from making enough of an enzyme that breaks down sugar inside cells. The disease can start in babies less than one year of age or older children. Understanding how different gene changes lead to varied symptoms helps physicians predict how the disease progresses, which offers the right treatment. This study revealed 26 children in Vietnam with Pompe disease. Most were babies with serious heart and muscle problems. The study also found two new variants of the GAA gene. The study contributes to genetic databases for Southeast Asia to help find and treat the disease earlier.
HIGHLIGHTS
Deficiency of acid α-glucosidase (GAA) enzyme activity causes glycogen to build up in lysosomes, mainly affecting the heart and skeletal muscles.
Enzyme replacement therapy (ERT) using recombinant human GAA has significantly improved survival and motor function, especially in patients with early infantile-onset Pompe disease.
This study included both retrospective and prospective observations of 26 Vietnamese Pompe patients. A total of 54 GAA gene variants were identified, including two novel variants reported for the first time: c.2016del (a frameshift mutation in exon 14) and c.1723T > C (a missense mutation in exon 12).
The study analyzed the patients’ genotypes and clinical features, underscoring the clinical importance of comprehensive GAA genotyping and CRIM status determination in predicting disease course and guiding therapeutic decisions
These findings add to the limited genetic data on Pompe disease in Southeast Asia, particularly in Vietnam.
1. Introduction
Pompe disease (PD: glycogen storage disease type II; OMIM #232300) is a rare, autosomal recessive lysosomal storage disorder caused by pathogenic variants in the GAA gene, leading to deficient acid α-glucosidase (GAA) activity. This enzymatic deficiency results in lysosomal glycogen accumulation, predominantly affecting cardiac and skeletal muscles, and manifests as a continuum of phenotypes ranging from classic infantile-onset (IOPD) to late-onset Pompe disease (LOPD) [1]. The classic IOPD typically presents within the first few months of life with hypertrophic cardiomyopathy (HCM), generalized hypotonia, respiratory insufficiency, and rapidly progressive disease if left untreated. In contrast, LOPD manifests after infancy with slowly progressive proximal muscle weakness and variable extracardiac involvement [2].
Currently, functional assays of the GAA enzyme remain central to diagnosis. Dried blood spot (DBS) testing offers a convenient, cost-effective platform for large-scale screening. However, due to a relatively high false-positive rate, confirmatory testing is required. Combining DBS with GAA gene sequencing enhances diagnostic accuracy and helps differentiate true deficiency from pseudodeficiency alleles, such as c.1726G > A, which was more prevalent in Asian populations [3]. Several countries, including Taiwan, Japan, and the United States, have incorporated PD into their newborn screening (NBS) programs, highlighting regional differences in variant distribution [4].
Enzyme replacement therapy (ERT) with recombinant human GAA has markedly improved survival and motor outcomes, especially in IOPD. However, therapeutic efficacy is influenced by multiple factors, notably the cross-reactive immunologic material (CRIM) status. CRIM-negative patients, who produce no endogenous GAA protein, are at higher risk of developing neutralizing anti-drug antibodies that limit ERT effectiveness [5].
Understanding genotype–phenotype correlations in PD is essential for prognosis and individualized therapy. Certain *GAA* variants were associated with CRIM-negative status and severe phenotypes, while others correlated with residual enzyme activity and milder presentations [6]. Recent studies expanded the mutational spectrum and underscored the value of integrating molecular and clinical data for comprehensive patient management [7].
Despite global progress, data from Southeast Asia, particularly Vietnam, remained limited. This study aimed to characterize the clinical and molecular features of PD in a Vietnamese pediatric cohort, focusing on GAA mutation profiles, CRIM status, and their association with clinical outcomes. Identifying population-specific variants and their phenotypic consequences may facilitate earlier diagnosis and inform context-appropriate treatment strategies in resource-constrained settings.
2. Materials and methods
2.1. Study design and patient recruitment
This was a retrospective and prospective observational study conducted at a single tertiary pediatric center – Children’s Hospital 1 in Ho Chi Minh City, Vietnam. Medical records of pediatric patients diagnosed with PD between January 2019 and December 2024 were reviewed. Inclusion criteria comprised confirmed deficiency of GAA activity and identification of pathogenic or likely pathogenic variants in the GAA gene. Patients were classified as having IOPD or LOPD based on age at symptom onset and presence of cardiac involvement. The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board of Children’s Hospital 1 (1608/QD-BVND1, date of approval: 16 September 2022).
2.2. Clinical and biochemical assessment
Data collected included demographic characteristics, age at symptom onset and diagnosis, family history, and clinical features such as HCM, generalized hypotonia, respiratory complications, and feeding difficulties. Biochemical investigations included serum levels of creatine kinase (CK), aspartate aminotransferase (AST), alanine aminotransferase (ALT), and GAA enzymatic activity. GAA enzyme activity was measured using fluorometric assays performed on DBS specimens.
2.3. Molecular genetic analysis
Genomic DNA was extracted from peripheral blood leukocytes. The coding regions and exon–intron boundaries of the GAA gene were amplified by polymerase chain reaction (PCR) and subjected to Sanger sequencing. Variants were interpreted according to ACMG guideline and Association for Molecular Pathology (AMP) criteria and cross-checked with publicly available databases [8]. Novel variants were further evaluated using in silico prediction tools.
2.4. CRIM status and enzyme replacement therapy (ERT)
CRIM status was classified as positive, negative, or unknown based on either direct protein testing or inference from genotype. Information regarding ERT – including initiation age, treatment duration, and patient outcomes (alive or deceased) – was recorded. For patients receiving ERT, clinical status and survival were evaluated at last follow-up.
2.5. Statistical analysis
Descriptive statistics were used to summarize clinical, biochemical, and genetic data. Categorical variables were displayed as frequency (%). Continuous variables were described under mean values and standard deviation (SD) for normally distributed variables and median (min-max) for non-normally distributed variables. Associations between genotype, CRIM status, and clinical outcomes were assessed.
3. Results
3.1. Patient demographics and clinical features
A total of 26 patients with confirmed PD were included in the study, comprising 23 with IOPD and 3 with LOPD (Table 1). Among the IOPD cohort, 56.5% were male, whereas 66.7% of the LOPD patients were female. The median age at disease onset for IOPD was 4.74 months (range: 1–17), with a median diagnostic age of 7.67 months (range: 2.3–32). LOPD patients presented at a median age of onset of 37.3 months and were diagnosed at a significantly later age of 152.3 months (range: 121–176).
Table 1.
Characteristics of 26 Pompe cases.
| Characteristics | IOPD | LOPD |
|---|---|---|
| Total | N = 23 | N = 3 |
| Gender | (n, %) | |
| Male | 13 (56.5) | 1 (33.3) |
| Female | 10 (43.5) | 2 (66.7) |
| Age at disease onset (months) (median, min–max) |
4.74 (1–17) | 37.33 (36–40) |
| Age at diagnosis (months) (median, min–max) |
7.67 (2.3–32) | 152.33 (121–176) |
| Family history of Pompe disease (yes) | 3 (13.0) | 2 (66.7) |
| Clinical manifestations | ||
| Hypertrophic cardiomyopathy | 23 (100) | 0 (0) |
| Hypotonia/ muscle weakness | 22 (95.6) | 3 (100) |
| Recurrent pneumonia | 19 (82.6) | 3 (100) |
| GAA enzyme activity (µM/hour) (median, min–max) |
0.47 (0.06–2.84) | 0.73 (0.47–0.95) |
| Biomarkers | ||
| AST (UI/L) | 326.6 (154.1–702.8) | 212.3 (181.5–233.5) |
| ALT (UI/L) | 164.6 (58.1–473.6) | 158.1 (105.7–219.9) |
| CK (UI/L) | 696.8 (294.5–1979.1) | 1327.4 (1001.3–1781.8) |
| CRIM status | ||
| CRIM-positive | 20 (87.0) | 1 (33.3) |
| CRIM-negative | 1 (4.3) | 0 (0) |
| CRIM-unknown | 2 (8.7) | 2 (66.7) |
| ERT (yes) | 15 (65.2) | 3 (100) |
| Death | 17 (60.8) | 0 (0) |
| With ERT | 9 (52.9) | 0 (0) |
| No ERT | 8 (47.1) | – |
Notes: Infantile-onset Pompe disease (IOPD): Individual diagnosis before age 12 months with hypertrophic cardiomyopathy. Late-onset Pompe disease (LOPD): Individuals with onset before age 12 months without hypertrophic cardiomyopathy. ERT: Enzyme replacement therapy; CRIM: cross reactive immunological material.
Clinical manifestations among IOPD patients included HCM in 100% of cases, hypotonia in 95.6%, and recurrent pneumonia in 82.6%. In contrast, LOPD patients presented with progressive muscle weakness and recurrent respiratory infections, without evidence of cardiomyopathy. ERT was initiated in 15 IOPD and all 3 LOPD patients. The overall mortality rate in the IOPD group was 60.8%, with 52.9% of deaths occurring despite ERT. No deaths were reported among LOPD patients. A family history suggestive of PD or related symptoms was reported in 5 cases, including two consanguineous siblings diagnosed with LOPD.
3.2. Distribution of GAA variants and CRIM status
A total of 54 GAA gene variants were identified in the 26 patients (Table 2). The most frequent variants were c.1843G > A (n = 10, 18.5%), c.1933G > C (n = 8, 14.8%), and c.2040 + 1G > T (n = 7, 13.0%), all of which were consistently associated with the infantile phenotype. Missense variants were most prevalent (n = 38, 70.4%), followed by splice-site (n = 9, 16.7%), frameshift (n = 4, 7.41%), and nonsense variants (n = 3, 5.6%). The predicted pathogenicity of these variants ranged from very severe to potentially mild, with less severe variants (n = 8, 14.8%) being more common overall.
Table 2.
Genotype and GAA aberration frequency in 26 Pompe cases.
| No. | GAA variants | Amino acid change | Type | Frequency (n = 54) |
Genomic location | Predicted severitya | Phenotype with null allelea | CRIM |
|---|---|---|---|---|---|---|---|---|
| 1 | c.1843G > A | p.Gly615Arg | Missense | 10 | Exon 13 | Potentially less severe | Classic infantile | Positive |
| 2 | c.1933G > C | p.Asp645His | Missense | 8 | Exon 14 | Potentially less severe | Classic infantile | Positive |
| 3 | c.2040 + 1G > T | – | Splice donor | 7 | Intron 14 | Very severe | Classic infantile | Unknown |
| 4 | c.1099T > C | p.Trp367Arg | Missense | 2 | Exon 7 | Potentially less severe | Classic infantile | Positive |
| 5 | c.1411_1414del | p.Glu471Profs*5 | Frameshift | 2 | Exon 9 | Very severe | Classic infantile | Negative |
| 6 | c.1444C > T | p.Pro482Ser | Missense | 2 | Exon 10 | – | – | – |
| 7 | c.1726G > A | p.Gly576Ser | Missense | 2 | Exon 12 | Presumably nonpathogenic | Unknown | Unknown |
| 8 | c.1822C > T | p.Arg608Ter | Nonsense | 2 | Exon 13 | Very severe | Classic infantile | Negative |
| 9 | c.2563G > C | p.Gly855Arg | Missense | 2 | Exon 18 | Potentially less severe | Classic infantile | Positive |
| 10 | c.2646 + 2T > G | – | Splice donor | 2 | Intron 18 | – | – | – |
| 11 | c.625T > C | p.Tyr209His | Missense | 2 | Exon 3 | – | – | – |
| 12 | c.796C > T | p.Pro266Ser | Missense | 1 | Exon 4 | Potentially mild | Classic infantile | Positive |
| 13 | c.1004G > A | p.Gly335Glu | Missense | 1 | Exon 6 | Potentially less severe | Unknown (disease-associated) | Positive |
| 14 | c.1057C > T | p.Gln353Ter | Nonsense | 1 | Exon 6 | Very severe | Unknown (disease-associated) | Negative |
| 15 | c.1723T > Cb | p.Tyr575His | Missense | 1 | Exon 12 | – | – | – |
| 16 | c.1927G > A | p.Gly643Arg | Missense | 1 | Exon 14 | Potentially less severe | Classic infantile | Unknown |
| 17 | c.2016delb | p.Asn673ThrfsTer23 | Frameshift | 1 | Exon 14 | – | – | – |
| 18 | c.2065G > A | p.Glu689Lys | Missense | 1 | Exon 15 | Nonpathogenic | Unknown | Positive |
| 19 | c.2104C > T | p.Arg702Cys | Missense | 1 | Exon 15 | Potentially less severe | Classic infantile | Positive |
| 20 | c.2173C > T | p.Arg725Trp | Missense | 1 | Exon 15 | Less severe | Childhood or adult | Positive |
| 21 | c.2723del | p.Gly908AlafsTer35 | Frameshift | 1 | Exon 19 | – | – | – |
| 22 | c.307T > C | p.Cys103Arg | Missense | 1 | Exon 2 | Potentially less severe | Unknown | Positive |
| 23 | c.752C > T | p.Ser251Leu | Missense | 1 | Exon 4 | Presumably nonpathogenic | Unknown | Positive |
| 24 | c.761C > T | p.Ser254Leu | Missense | 1 | Exon 4 | Presumably nonpathogenic | Unknown | Positive |
aPredicted severity and phenotype with the null allele were referred from the Pompe disease GAA variant online database (https://www.pompevariantdatabase.clmz.nl/pompe_mutations_list.php?orderby=aMut_ID1).
bNovel mutations. CRIM: cross reactive immunological material.
All nonsense and frameshift variants – such as c.1057C > T, c.1411_1414del, and c.1822C > T – were associated with CRIM-negative or unknown status, and correlated with more severe phenotypes. Conversely, almost all missense variants were CRIM-positive and more frequently observed in patients with better clinical responses to ERT.
Two novel variants, c.2016del (frameshift, exon 14) and c.1723T > C (missense, exon 12), were identified in compound heterozygosity with known pathogenic alleles in IOPD patients. Variant pathogenic validation was confirmed using in silico predictions and clinical correlation.
There were two pseudodeficiency variants, GAA c.2065G > A and c.1726G > A, and three variants of uncertainty, c.625T > C, c.752C > T, and c.761C > T, found in patients with clinical features of PD.
3.3. Individual case analysis
Among the first 23 cases (Cases #1–23; Table 3), patients exhibited marked GAA enzyme deficiency and carried compound heterozygous pathogenic/likely pathogenic GAA variants. Most were CRIM-positive, with the exception of Case #6, who harbored only a single pathogenic variant (c.1843G > A). All IOPD patients in this group presented with HCM, generalized hypotonia, and recurrent respiratory infections. GAA activity was profoundly reduced, with values typically <1 µM/h, some as low as 0.06 µM/h.
Table 3.
26 Cases diagnosed with Pompe disease.
| No. | GAA variant | Amino acid change | Allele | Pathogenicity by ACMG | CRIM status | Sex | GAA enzyme (µM/hour) | Age at diagnosis (months) |
Category | Phenotype |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | c.1933G > C | p.(Asp645His) | Het | P | Pos | F | 0.06 | 4.0 | IOPD | HCM Hypotonia Episodes of recurrent pneumonia |
| c.2040 + 1G > T | – | Het | P | |||||||
| 2 | c.1843G > A | p.(Gly615Arg) | Het | P | Pos | F | 0.14 | 3.5 | IOPD | HCM Hypotonia Respiratory failure and hypotonia |
| c.1933G > C | p.(Asp645His) | Het | P | |||||||
| 3 | c.1843G > A | p.(Gly615Arg) | Het | P | Pos | F | 0.13 | 5.0 | IOPD | HCM Hypotonia Severe pneumonia |
| c.1933G > C | p.(Asp645His) | Het | P | |||||||
| 4 | c.1933G > C | p.(Asp645His) | Het | P | Pos | M | 0.91 | 5.5 | IOPD | HCM Episodes of recurrent pneumonia |
| c.1843G > A | p.(Gly615Arg) | Het | P | |||||||
| 5 | c.2040 + 1G > T | – | Het | P | Pos | M | 2.84 | 6.0 | IOPD | HCM Hypotonia Episodes of recurrent pneumonia |
| c.2563G > C | p.(Gly855Arg) | Het | LP | |||||||
| 6a | c.1843G > A | p.(Gly615Arg) | Het | P | Pos | M | 0.59 | 4.0 | IOPD | HCM Hypotonia Pneumonia |
| 7 | c.1933G > C | p.(Asp645His) | Het | P | Pos | F | 0.49 | 4.0 | IOPD | HCM Hypotonia Pneumonia |
| c.2040 + 1G > T | – | Het | P | |||||||
| 8 | c.1933G > C | p.(Asp645His) | Het | P | Pos | M | 0.33 | 8.8 | IOPD | HCM Slanting of the eyes, not closing completely, short neck, and macroglossia Episodes of recurrent pneumonia |
| c.2173C > T | p.(Arg725Trp) | Het | P | |||||||
| 9 | c.1933G > C | p.(Asp645His) | Het | P | Pos | M | 0.16 | 3.77 | IOPD | HCM Hypotonia Respiratory failure |
| c.2040 + 1G > T | – | Het | P | |||||||
| 10 | c.1057C > T | p.Gln353Ter | Het | LP | Unk | M | 0.502 | 5.0 | IOPD | HCM Hypotonia Pneumonia |
| c.2646 + 2T > G | – | Het | P | |||||||
| 11 | c.2723del | p.Gly908AlafsTer35 | Het | LP | Neg | F | 0.43 | 4.9 | IOPD | HCM Hypotonia Pneumoni |
| c.2016delc | p.Asn673ThrfsTer23 | Het | LP | |||||||
| 12 | c.2040 + 1G > T | – | Het | P | Unk | F | 0.59 | 12.0 | IOPD | HCM Hypotonia and macroglossia Pneumonia |
| c.2646 + 2T > G | – | Het | P | |||||||
| 13 | c.1004G > A | p.(Gly335Glu) | Het | P | Pos | M | 0.69 | 4.5 | IOPD | HCM Hypotonia |
| c.2065G > A | p.(Glu689Lys) | Het | Benign | |||||||
| 14 | c.625T > C | p.(Tyr209His) | Het | VUS | Pos | M | 0.62 | 17.0 | IOPD | HCM Severe pneumonia Motor development delay, reduction of muscle strength, trouble eating and swallowing |
| c.1099T > C | p.(Trp367Arg) | Het | P | |||||||
| 15b | c.1843G > A | p.(Gly615Arg) | Het | P | Pos | M | 1.00 | 32.0 | IOPD | HCM Hypotonia Sepsis |
| c.625T > C | p.(Tyr209His) | Het | VUS | |||||||
| 16 | c.2040 + 1G > T | – | Het | P | Pos | F | 0.14 | 7.0 | IOPD | HCM Hypotonia Episodes of recurrent pneumonia |
| c.2104C > T | p.(Arg702Cys) | Het | P | |||||||
| 17 | c.1411_1414del | p.(Glu471Profs*5) | Het | P | Pos | M | 0,13 | 14.0 | IOPD | HCM Hypotonia |
| c.1843G > A | p.(Gly615Arg) | Het | P | |||||||
| 18 | c.796C > T | p.Pro266Ser | Het | P | Pos | F | 0.46 | 11.87 | IOPD | HCM Hypotonia Episodes of recurrent pneumonia |
| c.1843G > A | p.(Gly615Arg) | Het | P | |||||||
| 19 | c.1843G > A | p.(Gly615Arg) | Het | P | Pos | M | 0.1 | 2.3 | IOPD | HCM Hypotonia Episodes of recurrent pneumonia |
| c.1933G > C | p.(Asp645His) | Het | P | |||||||
| 20 | c.307T > C | p.Cys103Arg | Het | P | Pos | M | 0.1 | 2.8 | IOPD | HCM Hypotonia |
| c.1927G > A | p.Gly643Arg | Het | P | |||||||
| 21 | c.2040 + 1G > T | – | Het | P | Pos | F | 0.16 | 5.6 | IOPD | HCM Hypotonia Episodes of recurrent pneumonia |
| c.1099T > C | p.(Trp367Arg) | Het | P | |||||||
| 22 | c.1843G > A | p.(Gly615Arg) | Het | P | Pos | F | 0.13 | 6.2 | IOPD | HCM Hypotonia Episodes of recurrent pneumonia |
| c.2563G > C | p.(Gly855Arg) | Het | LP | |||||||
| 23 | c.1411_1414del | p.(Glu471Profs*5) | Homo | P | Pos | M | 0.21 | 6.3 | IOPD | HCM Hypotonia Episodes of recurrent pneumonia |
| c.761C > T | p.(Ser254Leu) | Homo | VUS | |||||||
| c.752C > T | p.(Ser251Leu) | Homo | VUS | |||||||
| 24 | c.1843G > A | p.(Gly615Arg) | Het | P | Pos | F | 0.47 | 176.0 | LOPD | Progressive muscle weakness (limbs), absent tendon reflexes, respiratory failure No report of HCM |
| c.1723T > Cc | p.(Tyr575His) | Het | LP | |||||||
| 25 | c.1822C > T | p.(Arg608Ter) | Het | P | Unk | F | 0.95 | 160.0 | LOPD | Muscle weakness at three years old. The patient frequently had a fall and found it hard to climb up-stair No report of HCM Case #25 and #26 were siblings |
| c.1444C > T | p.(Pro482Ser) | Het | P | |||||||
| c.1726G > A | p.(Gly576Ser) | Het | Benign | |||||||
| 26 | c.1822C > T | p.(Arg608Ter) | Het | P | Unk | M | 0.77 | 121.0 | LOPD | Muscle weakness at three years old. The patient frequently had a fall and found it hard to climb up-stair No report of HCM Case #25 and #26 were siblings |
| c.1444C > T | p.(Pro482Ser) | Het | P | |||||||
| c.1726G > A | p.(Gly576Ser) | Het | Benign |
Note: Het: heterogzygous; Homo: homozygous; Pos: positive; Neg: negative; Unk: unknown; P: pathogenic; LP: likely pathogenic; VUS: variant of uncertain significance; IOPD: infantile-onset Pompe disease; LOPD: late-onset Pompe disease; M/F: male/female; HCM: hypertrophic cardiomyopathy; CRIM: cross reactive immunological material; ACMG: American College of Medical Genetics and Genomics guidelines.
aCase #6: Only one variant was detected together with the deficiency of GAA enzyme.
bCase #15: Genetic aberrations were defined based on parental genetic analyses.
cCase #11 and #24 carry novel mutations.
Parental testing was conducted in Case #18 postmortem to determine the underlying GAA genotype. Notably, Case #23 carried three homozygous GAA variants: c.1411_1414del (pathogenic) and c.761C > T, c.752C > T (classified as variants of uncertain significance, VUS).
In Cases #24–26, all classified as LOPD, patients had enzyme deficiency and compound heterozygous GAA variants. Presenting features included muscle weakness without cardiomyopathy. In addition to pathogenic variants, benign pseudodeficiency alleles (c.2065G > A, c.1726G > A) and VUS (c.625T > C) were identified. Notably, Cases #25 and #26 were siblings with a shared clinical course and identical genotypes, including two pathogenic variants (c.1822C > T and c.1444C > T) and one benign allele (c.1726G > A).
Patient #24, despite carrying a novel variant (c.1723T > C), was classified as LOPD due to age at diagnosis (176 months) and absence of cardiomyopathy. However, she exhibited progressive muscle weakness and respiratory compromise consistent with PD.
4. Discussion
4.1. Genotype – phenotype correlation
In our study, the three most prevalent variants were c.1843G > A (18.5%), c.1933G > C (14.8%), and c.2040 + 1G > T (13.0%). The c.1843G > A and c.1933G > C missense mutations, located in exons 13 and 14, respectively, were predicted to be less severe and were predominantly associated with the infantile phenotype and CRIM-positive status. These findings aligned with previous studies indicating partial residual enzyme activity and moderate clinical severity [1,6]. In contrast, the splice donor variant c.2040 + 1G > T, though relatively frequent, was linked to more severe clinical outcomes and CRIM-unknown status, likely due to aberrant splicing [9].
Missense variants tended to cluster in regions essential for enzymatic function or protein folding. Structural modeling and biochemical analyses showed that substitutions such as GAA c.1843G > A (p.Gly615Arg) and c.1933G > C (p.Asp645His) disrupt critical interactions, destabilizing the protein structure and impairing lysosomal trafficking [10]. These variants were typically associated with very low residual activity and the classic infantile phenotype in homozygous or compound heterozygous states.
Splice-site variants like c.2040 + 1G > T might cause exon skipping or activation of cryptic splice sites, often leading to absent or nonfunctional protein products [5]. While low-level functional protein might occasionally be produced, these variants were generally linked with severe clinical presentation, particularly when combined with null alleles.
Truncating mutations – including c.1411_1414del (frameshift), c.1822C > T, and c.1057C > T (nonsense) – were exclusively associated with CRIM-negative or unknown status and correlated with poor outcomes. These variants abolished protein synthesis, contributing to the development of anti-ERT immune responses and limited treatment efficacy [2,9].
Two novel variants were identified: c.2016del (frameshift) and c.1723T > C (missense), both observed in compound heterozygosity with known pathogenic alleles. Their functional relevance was supported by clinical severity, biochemical findings and absence from global databases, suggesting possible founder effects in the Vietnamese population. Notably, c.1723T > C was found in a patient clinically classified as LOPD, indicating that phenotypic variability may be influenced by modifier alleles or non-genetic factors such as age at therapy initiation.
Several variants of uncertain or benign classification (e.g., c.625T > C, c.1726G > A, c.752C > T, c.761C > T) were detected in patients with Pompe phenotypes. These might act as disease modifiers, although their exact contributions remain unclear. Further functional and segregation analyses were warranted [7].
Genotypic differences across populations were well-documented. The Asian population was proportionately higher than the other populations in terms of P/LP GAA variant carriers, especially variants related to IOPD [11] and pseudodeficiency alleles [4,12]. Evidence showed GAA c.1935C > A (p.Asp645Glu) not only as common variants in Chinese and Taiwan populations [4,10], but also in South Asian [13]. A case series in Hong Kong also revealed the c.1935C > A as three most common mutations related to PD cases [14]. In 2019, the 12-case reports in Thailand showed c.1935C > A (p.Asp645Glu) and c.1933G > C (p.Asp645His) as the most prevalent variants leading to the hot spots of aberrations in exon 14 and exon 5 [15]. This study, including 26 Vietnamese PD cases, observed without GAA c.1935C > A; instead, c.1843G > A and c.1933G > C were two top dominant variants. As a result, exon 13 and exon 14 were locations with hotspot side of GAA mutations. Adding to our supportive evidence, according to the multinational registry database, GAA c.1843G > A was shown as the five most reported variants in the Asia-Pacific region, especially in infantile cases [16]. Nevertheless, except for two IOPD cases in Thai, GAA c.1933G > C (p.Asp645His) was rarely reported, with a single case in Taiwan in 1995 [17]. Based on this study and previous evidence, we reported a higher number of IOPD cases compared to LOPD, and variant c.1933G > C was largely emerged in our PD cases, which might reflect the specific profile of GAA in the Vietnamese cases.
In this study, two benign variants of GAA c.2065G > A and c.1726G > A were identified in three cases (case #13, #25, and #26), considered as pseudodeficiency alleles. Pseudodeficiency alleles were highly observed in the Asian population, which could lead to the lowering of GAA enzyme level but not be consequent in PD manifestation. Nevertheless, these variants could interfere with the other GAA pathogenic variants to induce a greater enzyme reduction in carriers [12,18] that suggested the advantage of adding genetic sequencing in NBS and diagnosis process for PD [19].
Four variants (c.1444C > T, c.1057C > T, c.1723T > C, c.2016del) were not previously reported in ClinVar but were classified as pathogenic or likely pathogenic per American College of Medical Genetics (ACMG) guidelines. The c.1444C > T variant, reported in Chinese LOPD patients [20,21], was also observed in our two LOPD cases (#25, #26), both siblings with mild phenotypes. In 2018, the GAA c.1057C > T was first identified in one Chinese LOPD patient as a novel mutation that came with one heterozygous variant of unknown effect (GAA c.1201C > A) [16,22]. Our case #10 carried simultaneously GAA c.1057C > T and one P variant (c.2646 + 2T > G) that might explain the manifestation of IOPD. The remaining two variants, GAA c.2016del (case #11, IOPD) and c.1723T > C (case #24, LOPD), appeared to be novel and were being reported for the first time in our study, with no prior documentation.
In our cohort, enzyme analysis was complemented by genetic testing to confirm diagnoses. In Case #6, a single pathogenic variant and low enzyme activity established the diagnosis. In Case #15, postmortem parental testing confirmed the genetic cause of disease. These examples highlighted the diagnostic value of combining biochemical and molecular analyses, particularly in the absence of NBS in Vietnam.
4.2. CRIM status and therapeutic implications
CRIM status would be a key predictor of immune response to ERT. In our study, CRIM-positive patients predominantly carried missense variants and demonstrated improved outcomes with early ERT. In contrast, CRIM-negative or unknown-status patients more frequently harbored null variants and experienced poorer outcomes. These findings highlighted the importance of early CRIM determination to guide immunomodulation strategies [5].
Our results showed that frameshift and nonsense variants were associated with CRIM-negative status and were frequently observed in patients with early mortality or poor response to ERT, which aligned with prior data regarding heightened immunologic risk of these variants. It highlighted the value of early molecular diagnosis in predicting disease severity and optimizing treatment strategies. Integration of CRIM testing into early diagnostic algorithms may improve ERT efficacy and reduce complications. This study reinforced the need to include regional genetic data in global databases to enhance CRIM prediction accuracy and optimize personalized care. Limited Southeast Asian data currently hindered global efforts to tailor treatment strategies. Expanding these datasets would be crucial for equitable care across populations.
4.3. Strengths and limitations
This study provided the first comprehensive overview of PD in southern Vietnam, describing clinical and molecular characteristics in 26 pediatric patients. Our findings would suggest potential cost-effective variant panels for regional NBS. Early IOPD identification, combined with CRIM testing and prompt ERT, may improve long-term outcomes.
However, limitations included the relatively small cohort size and lack of functional validation for novel variants. Variant pathogenicity was inferred using ACMG criteria, enzymatic data, and phenotypic correlation. Additional in vitro studies and larger cohorts were encouraged to confirm these associations.
5. Conclusion
PD is a life-threatening disorder with high recurrence risk due to autosomal recessive inheritance. Early diagnosis and treatment significantly influence prognosis. Our study outlined the genetic landscape of PD in Vietnamese children, emphasizing the importance of combining enzymatic and molecular diagnostics in the absence of NBS. Future large-scale studies should aim to functionally validate novel variants, explore modifier genes, and assess long-term ERT outcomes. Integrated clinical, genetic, and immunological data would be essential to advance precision medicine in PD.
Funding Statement
This paper was not funded.
Ethical declaration
The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board of Children’s Hospital 1 (1608/QD-BVND1, date of approval: 16 September 2022).
Consent form
Informed consent was obtained from all subjects involved in the study.
Disclosure statement
The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.
Data availability statement
Data sharing is not applicable to this article as no new data were created or analyzed in this study.
References
- 1.Rajamani G, Pillai NR, Stafki SA, et al. Predicting the phenotype of Pompe disease from features of GAA variants. Eur J Hum Genet. 2025;33(5):688–691. doi: 10.1038/s41431-024-01771-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Martinez-Montoya V, Sánchez-Sánchez LM, Sandoval-Pacheco R, et al. Mutational spectrum and genotype-phenotype correlation in Mexican patients with infantile-onset and late-onset Pompe disease. Mol Genet Genomic Med. 2024;12(7):e2480. doi: 10.1002/mgg3.2480 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Tarnopolsky M, Katzberg H, Petrof BJ, et al. Pompe disease: diagnosis and management. evidence-based guidelines from a Canadian expert panel. Can J Neurol Sci. 2016;43(4):472–485. doi: 10.1017/cjn.2016.37 [DOI] [PubMed] [Google Scholar]
- 4.Momosaki K, Kido J, Yoshida S, et al. Newborn screening for Pompe disease in Japan: report and literature review of mutations in the GAA gene in Japanese and Asian patients. J Hum Genet. 2019;64(8):741–755. doi: 10.1038/s10038-019-0603-7 [DOI] [PubMed] [Google Scholar]
- 5.Bali DS, Goldstein JL, Banugaria SG, et al. Predicting cross-reactive immunological material (CRIM) status in Pompe disease using GAA mutations: Lessons learned from 10 years of clinical laboratory testing experience. Am J Med Genet C Semin Med Genet. 2012;160C(1):40–49. doi: 10.1038/gim.2011.2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Moschetti M, Lo Curto A, Giacomarra M, et al. Mutation spectrum of GAA gene in Pompe disease: current knowledge and results of an Italian study. Int J Mol Sci. 2024;25(17):9139. doi: 10.3390/ijms25179139 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Najac C, van der Beek NAME, Boer VO, et al. Brain glycogen build-up measured by magnetic resonance spectroscopy in classic infantile Pompe disease. Brain Commun. 2024;6(5):fcae303. doi: 10.1093/braincomms/fcae303 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Richards S, Aziz N, Bale S, et al. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation. Genet Med. 2015;17(5):405–424. doi: 10.1038/gim.2015.30 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Niño MY, In ‘t Groen SLM, Bergsma AJ, et al. Extension of the Pompe mutation database by linking disease-associated variants to clinical severity. Hum Mutat. 2019;40(11):1954–1967. doi: 10.1002/humu.23854 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Moreland RJ, Jin X, Zhang XK, et al. Lysosomal acid α-glucosidase contains highly conserved catalytic and structural domains that are targets of common mutations. Hum Mutat. 2019;40(3):455–468. doi: 10.1002/humu.23703 [DOI] [Google Scholar]
- 11.Park KS. Carrier frequency and predicted genetic prevalence of Pompe disease based on a general population database. Mol Genet Metab Rep. 2021;27:100734. doi: 10.1016/j.ymgmr.2021.100734 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Tang H, Feuchtbaum L, Sciortino S, et al. The first year experience of newborn screening for Pompe disease in California. Int J Neonatal Screen. 2020;6(1):9. doi: 10.3390/ijns6010009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Labella B, Cotti Piccinelli S, Risi B, et al. A comprehensive update on late-onset Pompe disease. Biomolecules. 2023;13(9):1279. doi: 10.3390/biom13091279 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Chu YP, Sheng B, Lau KK, et al. Clinical manifestation of late onset Pompe disease patients in Hong Kong. Neuromuscul Disord. 2016;26(12):873–879. doi: 10.1016/j.nmd.2016.09.004 [DOI] [PubMed] [Google Scholar]
- 15.Ngiwsara L, Wattanasirichaigoon D, Tim-Aroon T, et al. Clinical course, mutations and its functional characteristics of infantile-onset Pompe disease in Thailand. BMC Med Genet. 2019;20(1):156. doi: 10.1186/s12881-019-0878-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Reuser AJJ, van der Ploeg AT, Chien Y-H, et al. GAA variants and phenotypes among 1,079 patients with Pompe disease: data from the Pompe Registry. Hum Mutat. 2019;40(11):2146–2164. doi: 10.1002/humu.23878 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Lin CY, Shieh JJ.. Identification of a de novo point mutation resulting in infantile form of Pompe’s disease. Biochem Biophys Res Commun. 1995;208(2):886–893. doi: 10.1006/bbrc.1995.1418 [DOI] [PubMed] [Google Scholar]
- 18.Sawada T, Kido J, Nakamura K.. Newborn screening for Pompe disease. Int J Neonatal Screen. 2020;6(2):31. doi: 10.3390/ijns6020031 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Gal A, Grosz Z, Borsos B, et al. Correlation of GAA genotype and acid-α-glucosidase enzyme activity in Hungarian patients with Pompe disease. Life. 2021;11(6):507. doi: 10.3390/life11060507 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Yue D, Jiao K, Xia X, et al. Diagnostic delay in late-onset Pompe disease among Chinese patients: a retrospective study. JIMD Rep. 2024;65(1):39–46. doi: 10.1002/jmd2.12404 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Jiao K, Zhu B, Chang X, et al. High-risk screening for late-onset Pompe disease in China: an expanded multicenter study. J Inherit Metab Dis. 2025;48(1):e12793. doi: 10.1002/jimd.12793 [DOI] [PubMed] [Google Scholar]
- 22.Liu HX, Pu CQ, Shi Q, et al. Identification of seven novel mutations in the acid alpha-glucosidase gene in five Chinese patients with late-onset Pompe disease. Chin Med J (Engl). 2018;131(4):448–453. doi: 10.4103/0366-6999.225056 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Data sharing is not applicable to this article as no new data were created or analyzed in this study.