AAV9-Mediated Gene Therapy for Infantile-Onset Pompe's Disease
Authors, Journal, Affiliations, Type, DOI
- Authors: Xiuwei Ma, Lu Zhuang, Wenhao Ma, Jun Li, et al.
- Journal: New England Journal of Medicine, 2025;392:2438–2446
- Affiliations: Chinese People's Liberation Army General Hospital, Beijing; Genecradle Therapeutics, Beijing; Beijing Children's Hospital, Capital Medical University
- Type: Brief report / first-in-human pilot trial
- DOI: https://doi.org/10.1056/NEJMoa2407766
- Funding: National Natural Science Foundation of China; National High Level Hospital Clinical Research Funding
- Trial Registry: ChiCTR2200063229
Overview
GC301, a recombinant AAV9 vector carrying codon-optimized human acid α-glucosidase (rAAV9-coGAA), was administered as a single intravenous infusion (1.2 × 10¹⁴ vg/kg) to four infants with infantile-onset Pompe disease. Three of four patients survived the 52-week observation period with improvement in cardiac outcomes (LVM index, LVEF) and psychomotor milestones; one patient died after parents declined invasive ventilation for respiratory failure. No anti-GAA antibodies were detected in any patient throughout the observation period — a key advantage over enzyme-replacement therapy. This is the first published human clinical data for systemic AAV9-GAA gene therapy in infantile-onset Pompe disease.
Keywords
Pompe disease, acid α-glucosidase (GAA), adeno-associated virus serotype 9 (AAV9), enzyme-replacement therapy, infantile-onset, lysosomal storage disorder, GC301, gene therapy, CRIM status, Hammersmith Infant Neurological Examination (HINE)
Key Takeaways
Study Design and Methods
- Eligible patients: infants <6 months with infantile-onset Pompe disease, cardiomegaly, hypotonia, and biallelic pathogenic GAA variants; modified Ross class IV (severe HF) excluded
- Sentinel design: Patient 1 enrolled first, monitored ≥4 weeks for ≥grade 3 drug-related AE before additional enrollment
- Dose: 1.2 × 10¹⁴ vector genomes/kg IV over 55–77 minutes, with prophylactic prednisolone
- All patients had received 1–5 prior rhGAA infusions; none received immune-tolerance induction
- Primary outcome: safety (adverse events, labs); secondary: efficacy (psychomotor and cardiac outcomes)
- Observation period: 52 weeks; long-term 5-year follow-up ongoing at 6-month intervals
Patient Characteristics
- 4 patients enrolled (Sept 2022 – March 2023)
- CRIM status: 3 patients predicted CRIM-positive by genotype; CRIM status uncertain for Patient 1
- Patient 2 was preterm (35+6 weeks), with septal hypertrophy noted at birth — disease onset likely in utero
Safety
- Most common AEs: Mild/moderate respiratory tract infections, hyperlipidemia, BNP elevation, diarrhea — all 4 patients
- Hyperlipidemia and BNP elevation: transient, asymptomatic, no treatment required
- Severe AEs (across 3 patients): Pneumonia (Patients 1, 2, 4); left ventricular dysfunction (Patient 1); bronchitis (Patient 2) — none attributed to GC301 by investigators
- Elevated AST/ALT in Patients 1, 2, 3 — no bilirubin or ALP abnormalities
- Patient 2 withdrew at week 21 due to respiratory failure; parents declined invasive ventilation; died 8 hours after cessation of care
Psychomotor Outcomes
- HINE scores increased to ≥62 in 3 patients within 4 weeks of injection
- Patients 3 and 4: HINE scores 74 and 75 respectively at 12 months of age — within normal range (reference >72 at 12 months for full-term infants)
- Patient 2: HINE improved from 45.5 at baseline to 53 at week 20 (6 months of age)
- Patients 1, 3, 4 attained the ability to sit without assistance, stand with assistance, and walk with assistance by week 52
- By February 2025 (22–29 months post-infusion), Patients 3 and 4 could stand and walk without assistance
- Mechanism: AAV9 crosses the blood–brain barrier (unlike rhGAA), potentially clearing CNS glycogen accumulation
Cardiac Outcomes
- LVM index (normal <64 g/m²):
- Patient 1: decreased from 157.15 → 67.83 at week 24 (LVM z-score 13.05 → 2.29); transiently worsened during pneumonia (week 30) but remained below baseline
- Patient 4: LVM index decreased over observation period despite 3 intercurrent infections
- Patient 3: decreased initially, then increased with recurrent respiratory infections
- Patient 2: no notable improvement
- LVEF: Improved in Patients 1, 3, 4; remained above lower normal limit in all 4 (except transient decrease in Patient 1 with pneumonia)
- Wall thickness (LV posterior wall and IVS during diastole) remained constant in Patients 1, 2, 4
GAA Enzyme Activity and Vector Genome
- GAA enzyme activity (dried blood spots) increased after GC301 injection in all 4 patients
- After week 2: remained above baseline in Patients 1, 3, 4 throughout observation
- Patient 1: GAA activity decreased to near baseline during frequent respiratory infections — infection can impair gene therapy efficacy
- Vector DNA (whole blood): decreased progressively after injection; undetectable at end of study in 3 surviving patients (likely due to anti-AAV9 antibody clearance)
Host Immune Response (Critical Finding)
- Anti-GAA binding antibodies: NOT detected in any patient at any time point (all 4 patients; unlike ERT recipients who develop anti-rhGAA IgG regardless of CRIM status)
- GAA-specific cellular immune responses (IFN-γ ELISpot): not detected in any patient
- Pre-existing anti-AAV9 neutralising antibody titers: <1:20 to 1:262 (range consistent with prior IV AAV studies)
- Post-injection: AAV-binding IgG developed and persisted throughout observation in all 4 patients (expected)
- Clinical implication: Gene therapy-induced immune tolerance to the transgenic GAA product may avoid the major limitation of ERT (persistent anti-rhGAA antibodies reduce efficacy, especially in CRIM-negative patients)
Respiratory Infection as a Confounding Factor
- All 4 patients experienced pneumonia during follow-up
- Respiratory infections associated with increased LVM index/z-score in Patients 1, 2, 3 — suggesting infections worsen cardiac function
- GAA activity dropped during infections in Patient 1 — infections may reduce gene therapy efficacy
- Proposed mechanisms: airway involvement in Pompe disease; AAV9 does not transduce alveolar cells; hypothetical "immunity gap" after COVID-19 social-isolation measures
Limitations of the Document
- Very small sample size (n=4): No control arm; preliminary safety and efficacy data only
- GAA activity measured in dried blood spots only: Plasma/leukocyte assays would provide better insight into cross-correction of affected tissues
- CRIM status undetermined for Patient 1: CRIM status is an important prognostic factor; uncertainty limits interpretation of immune tolerance findings
- One patient death: Confounded by parental refusal of invasive ventilation; not a drug-related AE per investigator assessment, but outcome reflects real-world disease severity
- Respiratory infection confounding: Infections affected both cardiac function and GAA activity — difficult to disentangle treatment effect from infection effect
- No CNS imaging or cognitive testing: AAV9 CNS penetration is inferred from HINE improvement; direct evidence of CNS glycogen clearance not demonstrated
- 52-week follow-up insufficient: Long-term durability of effect, vector dilution in growing infants, and re-dosing limitations are unaddressed
Key Concepts Mentioned
- concepts/AAV-Gene-Delivery — AAV9 systemic delivery platform; IV route; immune responses; paediatric dilution challenge
- concepts/Pompe-Disease — clinical features, natural history, ERT limitations, CNS involvement
- CRIM status — cross-reactive immunologic material; CRIM-negative predicts worse outcomes with ERT
Key Entities Mentioned
- entities/Genecradle-Therapeutics — manufacturer and study collaborator for GC301
Wiki Pages Updated
wiki/sources/aav9-pompe-nejm-2025.md— created (this file)wiki/concepts/AAV-Gene-Delivery.md— updated with Pompe clinical data and IV immune tolerance findingwiki/concepts/Pompe-Disease.md— created (new concept page)wiki/sourceindex.md— new entry addedwiki/wikiindex.md— Pompe-Disease concept entry added; AAV-Gene-Delivery description updated