AAV-2 Capsid-Specific CD8+ T Cells Limit the Duration of Gene Therapy in Humans and Cross-React with AAV-8 Capsid.

Efforts to establish an adeno-associated viral (AAV) vector-mediated gene therapy for the treatment of hemophilia B have been hindered by an immune response to the viral capsid antigen. Preclinical studies in small and large animal models of the disease showed long-term factor IX (F.IX) transgene expression and correction of the phenotype. However, in a recent phase I/II clinical trial in humans (Manno et al., Nat. Med. 2006), after hepatic gene transfer with an AAV-2 vector expressing human F.IX transgene, expression lasted for only a few weeks, declining to baseline concurrently with a peak in liver enzymes. We hypothesized that T cells directed towards AAV capsid antigens displayed by transduced hepatocytes were activated and these mediated destruction of the transduced hepatocytes, thereby causing loss of transgene expression and a transient transaminitis. Peripheral blood mononuclear cells isolated from AAV-infused subjects were stained with an AAV capsid-specific MHC class I pentamer either directly or after in vitro expansion. Two weeks after vector infusion 0.14% of circulating CD8+ T cells were capsid-specific on direct staining, and five weeks after infusion the capsid-specific population had expanded to 0.5% of the circulating CD8+ T cells, indicating proliferation of this T cell subset. By 20 weeks after vector infusion, the capsid-specific CD8+ T cell population had contracted to the level seen at 2 weeks. The expansion and contraction of this capsid-specific CD8+ T cell population paralleled the rise and fall of serum transaminases in the subject observed. Subsequent ex vivo studies of PBMC showed the presence of a readily expandable pool of capsid-specific CD8+ T cells up to 2.5 years post vector-infusion. Similarly, we were able to expand AAV-specific CD8+ T cells from peripheral blood of normal donors, suggesting the existence of a T cell memory pool. Expanded CD8+ T cells were functional as evidenced by specific lysis of HLA-matched target cells and by IFN-γsecretion in response to AAV epitopes. It has been argued that potentially harmful immune responses could be avoided by switching AAV serotypes, however, capsid protein sequences are highly conserved among different serotypes, as are some immunodominant epitopes that we identified. Indeed, we demonstrated that capsid-specific CD8+ T cells from AAV-infused hemophilic subjects functionally cross-react with AAV-8. Moreover, cells expanded from normal donors with AAV-2 vector capsids proliferated upon culture with AAV-8 capsids, demonstrating that both vectors could be processed appropriately in vitro to present the epitopic peptide to capsid-specific T cells. This suggests that AAV-2-specific memory CD8+ T cells normally present in humans likely would expand upon exposure to AAV-8 capsid epitopes. We conclude that the use of immunomodulatory therapy may be a better approach to achieving durable transgene expression in the setting of AAV-mediated gene therapy.