Preclinical model for evaluating human TCRs against chimeric syngeneic tumors

Background: The adoptive cell transfer (ACT) of T cell receptor (TCR)-engineered T cells targeting the HLA-A2-restricted epitope NY-ESO-1_157-165 (A2/NY) has yielded important clinical responses against several cancers. A variety of approaches are being taken to augment tumor control by ACT including TCR affinity-optimization and T-cell coengineering strategies to address the suppressive tumor microenvironment (TME). Most TCRs of clinical interest are evaluated in immunocompromised mice to enable human T-cell engraftment and do not recapitulate the dynamic interplay that occurs with endogenous immunity in a treated patient. A variety of humanized mouse models have been described but they have limitations in immune reconstitution and are technically challenging to implement. Here, we have developed a chimeric syngeneic tumor model in which A2Kb transgenic C57BL/6 mice are engrafted with B16 expressing A2Kb:NY as a single chain trimer (SCT) and treated by ACT with murine T cells expressing A2/NY TCRs comprising human variable fused to mouse constant regions.

Methods: We compared the function of a supraphysiological affinity A2/NY TCR (wtc51m), a computationally designed TCR in an optimal affinity range (DMβ), and a near non-binding TCR (V49I), engineered in both primary human and murine T cells by lentiviral and retroviral transduction, respectively. We evaluated a variety of strategies to stably express A2Kb:NY on the surface of mouse tumor cell lines including B16 melanoma, ultimately achieving success with an SCT comprising human β2m fused by GS linkers to both the NY-peptide and to α1 of the HLA complex. ACT studies were performed in B16-A2Kb:NY tumor-bearing, non-preconditioned immune-competent HLA-A*0201/H-2Kb (A2Kb) transgenic C57BL/6 mice and tumors characterized post-transfer.

Results: We observed significantly improved function of DMβ-T cells as well as superior infiltration and tumor control upon ACT as compared to the control TCR-T cells. Moreover, with our chimeric syngeneic tumor model, we were able to track dynamic and favorable changes in the TME upon DMβ-T cell transfer.

Conclusions: We have developed a robust, simple, and inexpensive preclinical strategy for evaluating human TCRs in the context of a fully competent murine immune system that can aid in the development of coengineered TCR-T cells and combination treatments translated to the clinic.