Abstract

Immunotherapy using peripheral blood T-cells engineered with T-cell receptor (TCR) or chimeric antigen receptor genes is a promising approach for the treatment of malignant diseases, and has demonstrated clinical efficacy capable of curing late stage cancer patients. Unfortunately the complete response rate remains low, and the majority of patients respond transiently and then relapse. The massive ex vivo expansion of autologous cells required to generate a therapeutic bolus may exhaust the replicative capacity of the infused cell product. An approach using cells with greater regenerative capacity is an attractive solution to this problem. We hypothesize that gene transfer of an NY-ESO-1 cancer/testes antigen specific TCR into human hematopoietic stem cells (HSCs) is capable of generating a continuous supply of effector T-cells capable of killing cancer in vivo.

To evaluate this approach, we utilize a humanized mouse model where peripheral blood stem cells (PBSC) enriched for the stem and progenitor marker CD34+ are transduced with a lentivirus encoding codon optimized NY-ESO-1 TCR and HSV-sr39TK transgenes, then adoptively transferred to preconditioned (100 cGy TBI) NSG-HLA-A2.1 neonatal mice. Development of NY-ESO-1 TCR bearing effector T-cells was detected in the peripheral blood of mice as early as 2 months post-transplant, and persisted for at least 5 months post-transplant. Ex vivo assay of T-cells developed from engineered HSCs showed robust release of interferon-γ when cocultured with NY-ESO-1 antigen and HLA-A2.1 matched tumor cells but not when HLA was mismatched, indicating HLA restricted antigen recognition. Furthermore cytotoxicity assays showed that engineered T-cells were capable of specifically killing tumor cells when antigen and HLA matched.

The inclusion of the PET imaging/suicide gene HSV-sr39TK allows both the non-invasive tracking of progeny derived from gene modified HSCs, and their ablation in the event of on-target / off-cancer reactivity or hematopoietic dysplasia due to insertional mutagenesis of gene modified cells. [18F]-FHBG PET imaging of TCR/TK engineered humanized mice allowed the detection of gene modified cells in the marrow compartments of mice, namely the long bones of the legs and arms. Importantly, strong thymus signal was also observed indicating robust thymic population/thymopoiesis of gene modified cells. No specific signal was detected in mock transduced HSC transplanted humanized mice, with identical biodistribution of signal as observed in non-transplanted NSG-A2.1 mice.

To examine the suicide gene function of the HSV-sr39TK cassette, we treated engineered humanized mice with vehicle or ganciclovir, the prodrug for suicide gene function of HSV-sr39TK. Ganciclovir treatment resulted in ablation of hematopoietic compartment specific PET signal, while vehicle did not. Mice were subsequently euthanized, and the presence of gene modified cells in organ compartments was examined by digital qPCR for the lentiviral psi element. We detected an order of magnitude decrease in the amount of gene modified cells in ganciclovir treated animals compared with controls.

Our studies demonstrate the feasibility of using TCR engineered HSCs for immunotherapy, as functional, HLA-restricted NY-ESO-1 T-cells developed in vivo from transplanted HSCs. In vivo imaging of gene modified cells supports visual resolution of gene modified HSC progeny at a level not previously described, and validation of the suicide gene function of HSV-sr39TK is of particular importance to gene therapy studies. Further work is focused on examining the immunological memory phenotype of TCR modified cells that develop from TCR modified HSCs, the effect of in vivo activation of these cells by dendritic cell pulse, and the ability of NY-ESO-1 humanized mice to mount a persistent anti-tumor response.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.