NY-ESO-1 Single Edited T Cells to Treat Multiple Myeloma without Inducing GvHD

Background. Gene transfer of T cell receptors (TCR) specific for tumor associated antigens is a promising approach of adoptive immunotherapy for cancer patients. However, T cells transduced with an exogenous TCR, have proved less effective than naturally occurring tumor specific T cells, with suboptimal clinical results. This could be due to the dilution of the tumor specific TCR that competes with endogenous TCRs for its expression on the T cell surface. Moreover, mispairing between endogenous and exogenous TCR a and b chains could generate new TCRs with unpredictable and potentially harmful specificities. To overcome these issues, our group developed a TCR gene editing procedure, based on the knockout of the endogenous TCR genes by transient exposure to a and b chain specific Zinc Finger Nucleases (ZFNs), followed by the introduction of tumor-specific TCR genes by lentiviral vectors (Provasi, Genovese et al, Nature Medicine 2102). The complete editing procedure requires multiple manipulation steps involving repeated cell activation cycles and four transduction procedures. Conversely, ‘single TCR editing’ (SE), based on the disruption of a single endogenous TCR chain, followed by transfer of the tumor specific TCR, generates redirected T cells fully devoid of their natural TCR repertoire, in a single round of cell activation. The aim of this project is to determine the efficacy and safety profile of SE T cells in vitro and in vivo.

Results. We validated the SE protocol exploiting an HLA-A2 restricted TCR specific for the cancer testis antigen NY-ESO-1_157-165 peptide. NY-ESO-1 is highly expressed by a large number of solid tumors and by the majority of monoclonal plasma cells of patients affected by high risk multiple myeloma, while its expression in healthy tissues is restricted to testicular germ cells. T cells harvested from healthy donors were activated with CD3/CD28 conjugated beads and cultured with low doses of IL7 and IL15. Two days after activation ZFNs were transiently delivered to T cells by mRNA transfection or by infection with adenoviral vectors, with similar efficacy. Sorted CD3^- T cells were transduced at day 8 with a bidirectional lentiviral vector encoding both the a and b chains of the NY-ESO-1-specific TCR. The SE strategy allowed the rapid production of high numbers of extremely fit tumor specific T cells, enriched for cells with an early differentiated phenotype (CD45RA+/CD62L+/CD95+, namely stem memory T cells). NY-ESO-1 TCR expression, investigated through dextramer specific binding, was significantly higher in SE than in T cells undergoing TCR gene transfer (TR). To test the efficacy of NY-ESO-1 redirected T cells we took advantage of the NY-ESO-1+HLA-A2+ U266 myeloma cell line. When tested against U266 cells in different assays (co-cultures, g-IFN and ⁵¹Cr release), all NY-ESO-1 redirected T cells showed a strikingly high killing potential. However, when we compared the alloreactive potential of the different NY-ESO-1 specific T cell populations in mixed lymphocyte reactions against completely mismatched allogeneic targets, we interestingly observed that the lysis of the allogeneic target by TR cells was significantly higher than that of SE T cells (p=0.05). This suggests that the residual endogenous polyclonal TCRs and possibly mispaired TCRs expressed on the cell surface of TR T cells can lead to off-target recognition, while SE T cells are devoid of such reactivity.

These results were validated in NSG mice, engrafted with U266 cells and subsequently infused with NY-ESO-1 specific T cells. All animals treated with tumor specific T cells were completely disease free at the time of sacrifice, demonstrating the powerful anti-tumor potential of the NY-ESO-1 TCR. However, a substantial proportion of mice administered with TR cells suffered from xenogeneic chronic graft versus host disease (GvHD), while all animals treated with SE T cells remained free of GvHD throughout the course of the study.

Discussion. The single gene editing protocol enables the rapid generation of clinically relevant doses of highly performing tumor specific T cells, fully devoid of their endogenous TCR repertoire, and thus particularly appealing for a future clinical translation. Donor-derived SE T cells, with a significantly reduced alloreactive potential, could be especially suitable to treat patients with minimal residual disease after allogeneic hematopoietic stem cell transplantation.