Gene therapies, antisense oligonucleotides (ASOs), RNA editing technologies, and other precision medicine approaches are rapidly transforming the therapeutic landscape for STXBP1 encephalopathy. Currently, around 16 candidate therapies are in development across five strategies: small molecule, including drug repurposing, gene transfer, CRISPR-based activation, RNA therapy, and readthrough compounds (Goss et al., 2024).

As the science expands, so does the number of precision therapies entering the pipeline — and each new mechanism necessitates a mouse model capable of testing it. The availability of the right validated model, not just the selection of one, is what determines whether a therapeutic program can move forward.

To support this evolving field, the STXBP1 Foundation led the development of a new mouse model carrying the patient-specific STXBP1 R122X mutation, which InnoSer characterized. This model gives developers of stop-codon readthrough compounds, mutation-specific ASOs, CRISPR base editors, and NMD inhibitors — therapeutic modalities that cannot be adequately tested in the gold-standard Stxbp1 HET KO model — a validated tool to directly test their approach against a genuine patient variant.

Characterization data will be first presented at the STXBP1 Foundation’s STXBP1 Summit+ (July 16–19, 2026, Philadelphia) by InnoSer study director Thomas Vogels. A manuscript detailing the full results is expected to be published in 2027. Results can also be made available upon request.

R122X mouse model recapitulates STXBP1 haploinsufficiency via introduction of patient-specific nonsense mutation

The R122X model is a STXBP1 mouse model carrying a direct patient variant: a premature stop codon mutation identified in a subset of STXBP1 patients, all presenting with severe early-onset developmental and epileptic encephalopathy, characterized by epileptic spasms, tonic and focal seizures, and profound neurodevelopmental impairment (Xian et al., 2022, Brain). This established nonsense and protein-truncating variants as among the most clinically severe mutation classes in STXBP1 encephalopathy. Premature stop codons result in incomplete protein production and reduced levels of functional STXBP1, ultimately leading to disease through haploinsufficiency.

The mutant allele was generated using CRISPR/Cas9-mediated genome editing to introduce a single-nucleotide substitution (C>T transition) within exon 6 of the Stxbp1 gene, resulting in a premature stop codon (R122X). Founder animals were bred on a 129S1/SvImJ genetic background. The resulting Stxbp1 R122X/+ mice carry a stable heterozygous loss-of-function allele producing a truncated transcript subject to nonsense-mediated decay (NMD) — modelling patient-relevant molecular pathology rather than complete gene deletion. Consistent with haploinsufficiency, a ~50% reduction in Stxbp1 expression was confirmed at both the mRNA and protein level.

To further characterize the model, we assessed a broad panel of readouts, including body weight, qPCR-based confirmation of genotype, behavioral phenotyping (open-field activity, fear conditioning, novel object recognition, and pole test for motor function), EEG monitoring for spike-wave discharge activity, and repetitive nerve stimulation to assess neuromuscular junction transmission. Full characterization data will be made available upon request for interested parties, with broader publication to follow at a later stage.

Importantly, the presence of this premature stop codon opens the door to variant-targeted therapeutic strategies that cannot be addressed by the gold-standard Stxbp1+/− model. As discussed above, Goss et al. (2024) has identified the following modalities already in development for patients carrying nonsense mutations, both directly testable in the R122X model:

• Premature stop-codon readthrough compounds
• Mutation-specific ASOs

Beyond these, the model’s defined genomic sequence — a genuine C>T nonsense mutation rather than a full gene deletion — makes it a mechanistically necessary tool for evaluating other nonsense-mutation-targeted approaches as they mature, including:

• CRISPR base editing
• RNA-editing therapies
• Nonsense-mediated decay (NMD) inhibitors

Modelling Stxbp1 haploinsufficiency via the knockout model of STXBP1: Stxbp1+/− mouse model

For many therapeutic programs, the Stxbp1+/− (heterozygous knockout) model — originally described by Kovacevic et al. (2018) — remains the gold standard for preclinical efficacy testing.

This model reproduces the underlying haploinsufficiency mechanism observed in patients: one functional allele remains intact and continues to produce normal STXBP1 protein, but at insufficient levels to maintain normal function. Because it represents the core disease mechanism rather than a specific mutation, it remains highly relevant to the majority of therapeutic programs under development, regardless of the patient’s underlying variant.

Applications include:

• • • Assessing the efficacy of therapies designed to restore STXBP1 expression or function, including AAV-mediated gene replacement, CRISPRa-mediated transcriptional upregulation of the intact allele, and ASO-driven upregulation of the intact allele
    • Evaluating broad-spectrum small-molecule anti-seizure medications acting downstream of STXBP1 deficiency
    • Evaluating therapeutic effects on key features of STXBP1 encephalopathy, including epileptiform activity (e.g., spike-wave discharges detected via wireless EEG), cognitive impairment, and altered neuronal function
    • Characterizing dose-response and durability of therapeutic effect longitudinally

Because the model represents the core disease mechanism rather than a specific mutation, it remains highly relevant for the majority of therapeutic programs under development.

The R122X mouse model: a case for mutation-specific regulatory data

As precision therapies increasingly target specific STXBP1 mutations rather than the broader disease mechanism, the regulatory landscape is evolving to match.

In February 2026, the FDA introduced draft guidance on a “Plausible Mechanism Framework,” intended to support approval of individualized therapies — including genome-editing and RNA-based treatments — for ultra-rare, mutation-specific patient populations. Under this framework, approval for one mutation subtype may extend to additional variants within the same gene without requiring a separate full clinical trial for each, provided sponsors can demonstrate that the same mechanism of action applies, supported by mechanistic and preclinical evidence in a model carrying the relevant mutation.

For STXBP1 gene-editing and RNA-based therapeutic candidates, the R122X mouse model could provide this kind of mutation-specific preclinical evidence base, with data that may become directly relevant as sponsors navigate this emerging pathway. Mutation-specific evidence alone, however, rarely tells the whole translational story: in many development programs, pairing R122X with the Stxbp1+/− model can offer a more complete package, supporting efficacy claims across both the mutation-specific and broader disease mechanisms these models each address.

In collaboration with the STXBP1 Foundation, InnoSer breeds and maintains the R122X colony on-site, supporting fast study start-up times for developers ready to begin preclinical testing. InnoSer offers both models, supported by EEG capabilities, behavioral phenotyping, biomarker analysis, and customized study designs.

Interested in discussing your STXBP1 program? Contact our scientific team to determine which model — or combination of models — best aligns with your therapeutic strategy.