Duchenne muscular dystrophy (DMD) is an X-linked recessive disorder with monogenic mutations setting the stage for successful gene therapy treatment. We have completed a study that directly deals with the following key issues that can be directly adapted to a gene therapy clinical trial using rAAV considering the following criteria: 1) A regional vascular delivery approach that will protect the patient from widespread dissemination of virus; 2) an approach to potentially facilitate safe passage of the virus for efficient skeletal muscle transduction; 3) the use of viral doses to accommodate current limitations imposed by vector production methods; 4) and at the same time, achieve a clinically meaningful outcome by transducing multiple muscles in the lower limb to prolong ambulation.

The capacity of AAV1, AAV6 or AAV8 to cross the vascular endothelial barrier carrying a micro-dystrophin cDNA was compared under identical conditions with delivery through a catheter placed in the femoral artery of the mdx mouse. Transduction efficiency was assessed by immuno-staining using an antibody (Manex1a) that recognizes the N-terminus of micro-dystrophin. The degree of physiologic correction was assessed by measuring tetanic force and protection from eccentric contraction in the extensor digitorum longus muscle (EDL). The vascular delivery paradigm found successful in the mouse was carried to the non-human primate to test its potential translation to boys with DMD.

Regional vascular delivery resulted in transduction by rAAV8.micro-dystrophin reaching 94.5 ± 0.9 (1 month), 91.3 ± 3.1 (2 months), and 89.6 ± 1.6% (3 months). rAAV6.micro-dystrophin treated animals demonstrated 87.7 ± 6.8 (1 month), 78.9 ± 7.4 (2 months), and 81.2 ± 6.2% (3 months) transduction. In striking contrast, rAAV1 demonstrated very low transduction efficiency [0.9 ± 0.3 (1 month), 2.1 ± 0.8 (2 months), and 2.1 ± 0.7% (3 months)] by vascular delivery. Micro-dystrophin delivered by rAAV8 and rAAV6 through the femoral artery significantly improved tetanic force and protected against eccentric contraction. Mouse studies translated to the hindlimb of cynamologous macaques using a similar vascular delivery paradigm. rAAV8 carrying eGFP in doses proportional to the mouse (5 × 10¹² vg/kg in mouse vs 2 × 10¹² vg/kg in monkey) demonstrated widespread gene expression [medial gastrocnemius – 63.8 ± 4.9%, lateral gastrocnemius – 66.0 ± 4.5%, EDL – 80.2 ± 3.1%, soleus – 86.4 ± 1.9%, TA – 72.2 ± 4.0%.

These studies demonstrate regional vascular gene delivery with AAV serotype(s) in mouse and non-human primate at doses, pressures and volumes applicable for clinical trials in children with DMD.