TCR Gene Editing Results in Effective Immunotherapy of Leukemia without the Development of GvHD

Abstract 667

Transfer of high-avidity T-cell receptor (TCR) genes isolated from rare tumor-specific lymphocytes into polyclonal CD8+ T cells is an attractive strategy for targeted cancer immunotherapy. However, the successful implementation of this approach is limited by technical and safety issues, including inefficient gene transfer, unstable transgene expression, exhaustion of gene-modified cells and most importantly, the unpredictable results of mispairing between the endogenous and exogenous TCR chains. Indeed, co-expression of the endogenous and exogenous TCR in the same cell not only reduces cell-surface expression of the introduced tumor-specific TCR, but also drives the potential for these gene-modified T cells to acquire autoreactive specificities. Such TCR mispairing has been shown to result in autoreactive T cells in animal models [Bendle et al., Nat Med. 2010]. To partially overcome these limitations, we have developed a novel strategy based on zinc finger nucleases (ZFNs) that permits editing of T cell specificity at the DNA level, combining the disruption of the endogenous TCR chain genes with the transfer of a tumor-specific TCR. Two sets of ZFNs were designed targeting the constant regions of the α (TRAC-ZFN) and β (TRBC-ZFN) TCR chain genes, respectively. We transiently delivered these ZFNs into primary T lymphocytes activated with anti-CD3 and anti-CD28 antibody-conjugated beads and cultured with low doses of IL-7 and IL-15, to promote the survival and expansion of the ZFN-modified cells. ZFN delivery into activated T lymphocytes abrogated expression of the CD3/TCR complex on the cell surface (% of CD3^neg cells with TRAC-ZFN: 34%±11 and with TRBC-ZFN: 30%±9). No phenotypic differences were observed in CD3^pos and CD3^neg lymphocytes, which displayed a similar CD4/CD8 ratio while displaying an early T-cell differentiation phenotype, as evidenced by high expression of CD62L, CD27, CD28 and IL-7Rα markers. Sorted CD3^neg cells proved stable in culture (demonstrating that ZFN exposure was well tolerated), and did not respond to TCR-dependent stimulation with the mitogen PHA, as expected for cells carrying a disrupted TCR gene. CD3^neg cells were efficiently transduced with a lentiviral vector encoding a tumor-specific exogenous TCR chain, resulting in the restoration of cell surface translocation of CD3, thus facilitating the selective expansion of TCR-transduced cells by polyclonal stimulation. To demonstrate the antitumor activity of these modified cells we selected an HLA-A2 restricted, codon-optimized cysteine-modified TCR specific for the Wilms' tumor antigen 1 (WT1). To achieve complete editing of T cell specificity, we established a protocol that sequentially disrupted the endogenous TCR α and β chains with high efficiency (averages: 36% and 18%), followed by lentiviral transfer of the tumor-specific TCR α and β chains (average efficiencies: 65% and 25%). This procedure resulted in a population of TCR-edited lymphocytes encoding only the tumor-specific TCR that, in the absence of competition from the endogenous receptor, was expressed at high physiological levels. Accordingly, TCR-edited lymphocytes were superior to conventional TCR-transferred cells in promoting specific recognition of WT1-expressing targets, including primary leukemias, and most importantly, were devoid of residual endogenous TCR reactivity including alloreactivity. Finally, in a humanized GvHD model, we showed that, at a variance with 100% of mice infused with unmanipulated T cells, and 80% of mice receiving TCR-transferred lymphocytes developing lethal GvHD, no GvHD was observed upon infusion of matched TCR-edited cells, despite robust T cell engraftment rates across all groups. These data demonstrate that the successful genetic re-programming of T cell specificity in primary lymphocytes results in a functionally superior target specific killing activity and thus has the potential to greatly improve the safety and therapeutic benefit of cancer immunotherapy, without triggering its potentially negative effects. (Provasi and Genovese: equal contribution).