647. Repeated Infusion of AAV2-hAADC Into the Thalamus of Non-Human Primates: Effect of Anti-AAV2 Antibodies on Transgene Expression

AAV-based gene therapy has considerable potential as a therapeutic tool to treat many monogenic diseases by restoring deficient expression of key genes after a single injection of an AAV hosting a functional copy of the gene. Under some circumstances, these diseases may need more than a single administration of the AAV-based therapeutic agent. One of the main concerns related to repeated AAV delivery is inhibition of cell transduction at the second infusion site by neutralizing antibodies (NAb) generated after the first exposure.In the present study, we evaluated the immune response after repeated AAV2 delivery to the brain in AAV2 seronegative Rhesus macaques. We infused AAV2-hAADC into left and right thalamus, with a 91-day delay between vector infusions. We determined the levels of anti-AAV2 NAb in sera weekly to monitor antibody production until necropsy 130 days after the first infusion. To ensure high levels of NAb before second infusion, we boosted the NAb production with 2 intramuscular AAV2-VLP immunizations.AADC expression in the thalamus was analyzed in post-mortem tissue by immunohistochemistry and inhibition of transduction was observed in the second infused hemisphere. Immune markers such as GFAP, Iba1, MHCII, CD8 and CD20 were also analyzed, as well NeuN and Hu0026E, to investigate the extent of the immunologic activation after a second brain delivery. Surprisingly, these markers evinced a more dramatic immune response at the first injection site than at the second, resulting even in neuronal loss at the infusion site. CD20 staining revealed the abundant presence of B-lymphocytes within the infusion site. A20 antibody immunostaining revealed the presence of the AAV2 capsid in both hemispheres suggesting a possible trigger of the immune response.The present work demonstrates that very high levels of NAb generated after a primary injection can impair the transduction levels in the second injection site. Moreover, it shows that infiltration of B-lymphocytes can induce tissue damage under conditions of AAV2 capsid persistence. Further studies will be performed to determine the chronology of the events and then establish optimal design and timeline for serial multiple-administration of the AAV-based therapeutic agents. The data further suggest that use of serotypes, such as AAV2, that clear from tissues slowly may not be suitable for therapies in which readministration is being considered.