Inducible Enhancement of T Cell Function and Anti-tumor Activity after Adoptive Transfer

Main Text

The transfer of tumor-reactive T cells has shown clinical efficacy in a variety of malignancies—notably in the treatment of B cell malignancies—an indication likely to soon receive Food and Drug Administration (FDA) approval as the first ever T cell therapeutic product.1, 2, 3 However, improved methods are needed to achieve efficacy in a greater number of cancer patients. In this issue of Molecular Therapy, Foster et al.⁴ demonstrate a novel approach through the use of a ligand-inducible genetic element that offers the possibility of broadly improving the efficacy of adoptive cell therapy (ACT).

The adoptive transfer of tumor-reactive T cells can be a powerful therapeutic option for patients with metastatic disease resistant to other treatments.1, 2 Early ACT studies used endogenous non-genetically modified T cells and, therefore, relied on the inherent tumor reactivity encoded by the native T cell receptor (TCR) genes. Today, the transfer of ex vivo-expanded tumor-infiltrating lymphocytes into lympho-depleted patients with metastatic melanoma, plus the administration of high dose interleukin-2 (IL-2), can mediate complete responses in roughly 1 in 4 patients.⁵ Other endogenously derived T cell populations, such as tetramer-expanded tumor-reactive T cells, have shown promise.⁶ However, for many patients, it is difficult to isolate and expand tumor-reactive T cells. To overcome this barrier, T cell tumor reactivity can be genetically engineered with antigen receptors (such as TCRs or chimeric antigen receptors [CARs]).1, 2, 3

While the infusion of engineered T cells has demonstrated great potential in cancer therapy, these modified cells have not gone through normal tolerance mechanisms, such as thymic education. Therefore, despite allowing reactivity against tumor antigens, these artificially generated T cells may react against antigens found on healthy tissues.1, 2, 3 In the case of CD19-reactive CARs, CD19 is expressed on healthy B cells and, upon transfer of CD19-reactive CAR T cells, these healthy B cells are destroyed. As B cells are not essential for life when appropriate supportive therapy is provided, patients on balance derive great overall benefit from the effective treatment of their CD19⁺ malignancies. In contrast, other CAR-modified T cells have been administered to patients, including those targeting carbonic anhydrase IX, ERBB2 (HER2), and mesothelin. While some of these trials have demonstrated evidence of anti-tumor activity, they have been limited by toxicities associated with CAR T cell recognition of non-target cells, and, in fact, patients have died as a result of these therapies.1, 2, 7

With current clinical strategies, donor T cells are permanently genetically modified, and there is no method to easily turn off or modify their activity. One way to limit toxicity is to deliver T cells whose antigen specificity has been modified in a transient fashion; however, such T cells may not be capable of mediating deep durable responses.⁸ Another strategy is to allow removal of donor T cells with a drug-inducible suicide gene or a cell-expressed ligand that can be targeted with a depleting antibody.2, 7, 9 Approaches are also being developed in which the CAR activity depends on a small molecule drug for function or where CAR activity depends on co-expression of two tumor antigens or non-expression of a non-tumor antigen.2, 7, 9 Selective and conditional antigen targeting can also be achieved with the use of a bi-specific antibody to bring together lymphocyte and tumor cells¹⁰ or the use of a biotin-binding immune receptor in conjunction with a biotinylated molecule targeting a tumor antigen.¹¹

Another hurdle with adoptive cellular therapy is the induction of T cell dysfunction through tumor-induced suppressive pathways.12, 13 In metastatic melanoma, T cells with tumor reactivity reside within the tumor, and these T cells have evidence of dysfunction and express PD-1.14, 15 The extent to which such dysfunction develops in transferred tumor-reactive T cells in human cancer is not known, but such dysfunction has been observed with human CAR-modified T cells infused into mice.¹⁶ Methods to reverse T cell dysfunction will likely be critical in the development of more effective T cell therapies.

In the new study appearing in this issue, Foster et al.⁴ demonstrate a powerful approach to augmenting T cell reactivity after adoptive transfer using a ligand-inducible genetic element. With the addition of the small molecule ligand rimiducid (Rim; formerly AP1903), the authors show that dimerization induction of a cell membrane-targeted signaling molecule, MyD88/CD40 (iMC), leads to signaling through truncated MyD88 and CD40 genetic elements and greatly augments T cell functional activity. It is notable that Rim has been previously used safely in patients in conjunction with an inducible caspase 9 switch that allowed elimination of donor T cells so as to halt graft-versus-host disease following stem cell transplantation for relapsed acute leukemia.¹⁷ In the current study, using a mouse model, the authors show that, following adoptive transfer of T cells expressing this inducible molecular switch, administration of Rim augmented anti-tumor immunity. While more work needs to be done to establish how signaling of the MyD88 and CD40 genetic elements interact with signaling induced by TCR and CAR engagement in terms of cellular activation and function, this strategy represents a potentially powerful system whereby T cells with weak tumor reactivity and limited off-target toxicity may have their activity transiently increased to more effectively target the tumor. Alternatively, the ability to boost cellular function with administration of a drug may facilitate overcoming T cell dysfunction induced by an immunosuppressive tumor environment.

For certain patients with few or no further therapeutic options, the transfer of tumor-reactive T cells has yielded durable anti-tumor response, though treatment failure and toxicity remain significant hurdles. The development of new strategies that allow inducible enhancement of T cell activity after transfer into a patient will be critical for improved clinical outcomes and reduced toxicities. Foster et al.⁴ demonstrate an exciting exploration of this principle and point the way toward combinations of other genetic elements and inducible signaling molecules that may also enhance or regulate T cell function after adoptive transfer.