506. In Vivo Transduction of T Cells: The Future of Immunotherapy?

Genetically engineered T cells carrying antigen-specific chimeric antigen receptor (CAR) or T cell receptor (TCR) genes are highly effective in immunotherapy of leukemia and other solid cancers. At present, the modification of T cells requires an ex vivo gene transfer using viral or non-viral vector systems and the subsequent expansion of the engineered T cells to yield therapeutic numbers. However, extended periods of cell culture can lead to an alteration of the T cell phenotype due to antibody and cytokine-driven activation and may negatively influence T cell functionality. To counteract these drawbacks, efforts were undertaken to engineer naive or memory T cells in vitro. However, these T cell subsets are already naturally available in vivo and could be used as a target for genetic engineering if it would be feasible to establish protocols that allow in vivo transduction of a sufficient number of these T cells to achieve tumor rejection or to control infectious diseases. We developed targetable, injectable gamma-retrovirus vectors capable of transducing specifically CD8+ and CD4+ T cells in vivo. We employed the measles virus (MV) envelope glycoproteins hemagglutinin (H) and fusion to generate retroviral vectors (RV), specific for either murine CD8α or murine CD4. To do so, we added single-chain antibody fragment (scFv) coding sequences derived from CD8α or CD4 hybridomas to the carboxy-terminus of the H-protein encoding sequence. We generated CD8α (MVm8) and CD4 (MVm4) targeted RV carrying fluorescence marker or TCR genes. When applying both targeted RV simultaneously in in vitro transduction experiments MVm8 and MVm4 exclusively transduced CD8+ or CD4+, respectively, both in T cell lines and primary mouse T cells. By intravenous injection of MVm8 RV, encoding an ovalbumin- (OVA) reactive TCR (OT-I TCR) linked to luciferase, into Rag2−/− mice previously repopulated with CD8 T cells, we demonstrate that MVm8 RV mediates selective in vivo transduction of CD8+ T cells. The in vivo TCR engineered T cells showed antigen-specific functionality by homing towards sites of antigenic stimulation and persisted long-term, suggesting the formation of memory T cells. Upon in vivo T cell engineering, mice were challenged by OVA-expressing Listeria monocytogenes in an infection model. Notably, protective immunity could be demonstrated and was dependent on the number of in vivo OT-I TCR engineered T cells. In sum, we show that: (i) gamma-retroviruses in combination with a modified MV envelope can be employed as platform for the development of targeting vectors, (ii) these vectors are able to transduce subsets of T cells in vivo as shown for murine CD8+ and CD4+ T cells, and (iii) in vivo engineered T cells are functional and able to provide protective immunity. Currently, we are investigating the potential of in vivo engineered T cells to reject experimentally induced cancer in a mouse model.