AAV9 gene replacement (FRF-001) for FOXG1 c.217C>T nonsense variant: promise and dosage sensitivity concerns
CONCLUSION
AAV9-mediated FOXG1 gene replacement therapy (FRF-001, using scAAV9 with hSyn1 promoter) is the most clinically advanced therapeutic approach for this early nonsense variant. FRF-001 has received FDA Fast Track, Orphan Drug, and Rare Pediatric Disease designations, with clinical trials expected in 2026. Preclinical data in Q84P knock-in mice show rescue of locomotor and behavioral phenotypes following postnatal intracerebroventricular delivery. Because c.217C>T creates a premature stop at codon 73 — well before the forkhead DNA-binding domain — the truncated protein is entirely nonfunctional, making exogenous gene delivery the most straightforward rescue strategy for this variant.
EVIDENCE
Behavioral efficacy study (bioRxiv, April 2025): ICV injection of scAAV9.hSyn1-opthFOXG1 at P6 in Q84P mice rescued open-field activity, running wheel locomotion, and multiple SmartCube behavioral endpoints. A separate 2024 study demonstrated that postnatal AAV-FOXG1 injection in neonatal Foxg1-heterozygous mice rescued callosal axon connectivity and substantially recovered the corpus callosum. However, a follow-up safety study (bioRxiv, October 2025) revealed that early P2 treatment exacerbated certain motor phenotypes in female but not male mice, indicating sex-specific dose-response effects. CRISPRa-based upregulation of endogenous FOXG1 from the wild-type allele has also shown proof-of-concept in patient-derived cells (NAR Molecular Medicine, 2025), achieving 1.68–2.35-fold upregulation with no detectable off-target effects. PLP1-targeting ASOs have been shown to normalize grip strength deficits in Foxg1-mutant mice by correcting downstream PLP1 overexpression (PMID: 39409184).
For early nonsense variants like p.Gln73Ter, gene replacement is variant-agnostic and bypasses the need for allele-specific correction. CRISPRa upregulation of the wild-type allele (NAR Molecular Medicine, 2025) offers a potentially safer alternative by preserving endogenous regulatory control, but remains in cell-based stages. The FOXG1 Research Foundation has been exploring suppressor tRNA partnerships for nonsense-specific readthrough, though the UAG context at this position has intermediate readthrough efficiency.
LIMITATIONS
The central challenge for FOXG1 gene therapy is extreme dosage sensitivity: both haploinsufficiency and overexpression of FOXG1 cause disease (duplications are associated with autism-like phenotypes and immune dysregulation). The therapeutic window between insufficient and excessive FOXG1 expression is narrow and may vary by brain region and developmental stage. The October 2025 preclinical finding of sex-specific adverse effects at early treatment timepoints adds significant complexity to clinical dosing. Not all behavioral phenotypes were rescued in preclinical models (contextual fear conditioning was not improved). Brain delivery and biodistribution remain challenging — achieving uniform transduction across cortical and subcortical regions with a single ICV injection is not guaranteed. The small patient population (~3 per 100,000 births) necessitates international collaboration for adequately powered trials. For this specific nonsense variant, suppressor tRNA technology is an exciting mutation-class-specific alternative under early exploration by the FOXG1 Research Foundation, but no FOXG1-specific readthrough data exist yet.