Engineered T cells for anti-cancer therapy

Abstract

Recent advances enabling efficient delivery of transgenes to human T cells have created opportunities to address obstacles that previously hindered the application of T cell therapy to cancer. Modification of T cells with transgenes encoding TCRs or chimeric antigen receptors allows tumor specificity to be conferred on functionally distinct T cell subsets, and incorporation of costimulatory molecules or cytokines can enable engineered T cells to bypass local and systemic tolerance mechanisms. Clinical studies of genetically modified T cell therapy for cancer have shown notable success; however, these trials demonstrate that tumor therapy with engineered high avidity tumor-reactive T cells may be accompanied by significant on-target toxicity, necessitating careful selection of target antigens and development of strategies to eliminate transferred cells.

Figure 1. Schema for adoptive immunotherapy with genetically modified T cells

T cells or defined T cell subsets with distinct functional attributes (green) can be isolated from patient blood and genetically modified to express a transgene ecoding a tumor-targeting TCR or CAR (yellow). The genetically modified T cells are then expanded in vitro and if necessary, enriched before infusion into the patient. Conditioning chemoradiotherapy may be administered to lymphodeplete the patient prior to T cell infusion to enhance persistence of the infused T cells. The figure depicts an example of a general strategy for engineering autologous T cells for adoptive immunotherapy.

Figure 2. Factors to consider in the design of tumor-targeting CARs

CAR specificity is governed by the selection of scFv but variation in target epitope density and location, scFv affinity, V_H/V_L orientation, spacer length and costimulatory signaling may influence the nature of synapse formation between the T cell and target cell, the functional efficacy of a given CAR, and effects on T cell gene transcription and epigenetic programming. Evaluation of the variables in CAR design remains empiric, and relies on in vitro assays and in vivo antitumor activity in animal models that are of uncertain value for predicting clinical efficacy.