Facioscapulohumeral muscular dystrophy (FSHD) is an autosomal dominant muscle disease affecting over a million people worldwide. Typical FSHD onset begins in late childhood or early adulthood with potentially years- or decades-long progression that could lead to wheelchair dependence and respiratory impacts. FSHD is caused by de-repression of the DUX4 gene in skeletal muscle, and DUX4 silencing is therefore the most direct route to FSHD therapy. There are currently no approved disease-modifying treatments for FSHD. We previously published pre-clinical efficacy and safety data supporting DUX4-targeted RNAi-based gene therapy for FSHD using an engineered miRNA, called mi405, delivered by first-generation AAV vectors. AAV6- and AAV9-delivered mi405 rescued molecular, histopathological, and functional outcomes in TIC-DUX4 mice, an FSHD mouse model of disease, but required high doses (3E14 vg/kg) to sufficiently transduce muscle in vivo. To improve biodistribution, efficacy, and safety, while reducing doses required for broad muscle transduction, we evaluated AAV-SLB101, a next-generation AAV capsid engineered for improved muscle tropism and liver de-targeting.
In this study, we performed a dose finding protocol using the FSHD TIC-DUX4 mouse model, and assessed DUX4-associated mouse biomarkers and behavioral assays. We determined that AAV-SLB101-delivered mi405 completely protected TIC-DUX4 functional, histopathological, and molecular deficits at doses ~1 log lower than first generation vectors, thereby establishing a potential minimal effective dose for human trials in the E13 vg/kg range. Our data support that combining AAV-SLB101 with the highly efficacious mi405 product will enable lower and potentially safer AAV doses while maintaining or improving the therapeutic efficacy achieved using high dose first generation vectors.