Advances and challenges in human AAV-mediated gene transfer: Immunological insights from a mouse model of human glycosylation

Successful gene transfer for monogenic human disease can potentially provide a singularly administered, lifelong cure. Yet concerns remain over the safety and efficacy of gene transfer. Adeno-associated virus (AAV) is a commonly used gene transfer vector that is predominantly non-integrating, can transduce and persist in non-dividing cells, and is relatively non-inflammatory. AAV\u27s have seen extensive pre-clinical success in animal models of Hemophilia B, with recent efficacy in the clinic. In both muscle-directed and liver-directed gene transfer of Factor IX (F.IX), multi-year expression of F.IX from AAV was observed in mice, dogs, and non-human primates. However, muscle-directed transfer to Hemophilia B human subjects resulted in sub therapeutic circulating F.IX, and liver-directed transfer led to transient therapeutic F.IX plasma levels that were eliminated by a hepatocyte-clearing CD8 T cell response directed against the AAV capsid. Thus the challenge of maintaining long-term, clinically meaningful levels of F.IX from an AAV vector in human subjects remains. The second chapter of this dissertation details a follow up study on the initial muscle-directed, AAV-F.IX trial. We now show 10-year F.IX expression in the muscle of a trial subject. This is the longest expression yet demonstrated in humans from a parenterally administered gene therapy vector. While therapeutic levels were never achieved in this trial, the persistence of gene expression over a decade after vector administration is an important finding for the field of gene transfer. Interestingly, AAV-F.IX delivery to the liver did result in efficacious levels of F.IX, but in the first human trial in liver, levels fell to baseline by two months post- delivery. The CD8 T cell response directed against the AAV capsid in these patients was not predicted in any pre-clinical animal studies. The third chapter of this dissertation investigates a uniquely human glycosylation mutation in the Cmah gene that potentially rendered humans more immunologically reactive. When modeled in mice, this mutation leads to enhanced T cell proliferation and activation in vitro, and to more robust T cell responses to viral challenges in vivo. The goal of these investigations is to highlight the long-term potential of AAV-mediated gene transfer, while attempting to delineate the uniquely human immune mechanisms that that influence duration of expression and that were not predicted by extensive studies in other species