Tumor-specific allogeneic cells for cancer therapy

Abstract

Adoptive cell transfer (ACT) therapy involves transfer of therapeutic lymphocytes to patients mostly for the treatment of cancer and viral infections. One modality to generate therapeutic lymphocytes is to genetically engineer them to express a chimeric antigen receptor (CAR) capable of recognizing the desired target. Current ACT approaches employ the patient's own (syngeneic) lymphocytes, which is both economically and technically challenging. Using foreign (allogeneic) lymphocytes in ACT is problematic because of the severe immunological reaction that occurs between genetically mismatched individuals. However, recently our group has developed a protocol, which allows for safe and effective ACT therapy in a murine model of metastatic disease using allogeneic T cells redirected with a human EGFR2/neuregulin (Her2/neu)-specific CAR. Mild preconditioning of the recipient delayed the rejection of the allogeneic donor T cells such that they had enough time to destroy the tumor, but not enough to cause significant damage to the host. By modulating lymphocyte migration using FTY720, we were actually able to exploit the allogeneic anti-host reaction in order to augment therapeutic benefit while concurrently improving the safety of the treatment. Therefore, we suggest that CAR-based allogeneic ACT therapy could be universally used as a safe and potent ‘off-the-shelf’ treatment for cancer.

Keywords:

• adoptive cell therapy (ACT)
• allogeneic
• cancer
• chimeric antigen receptor (CAR)
• egress
• FTY720
• graft-versus-tumor (GvT)
• graft-verus-host-disease (GvHD)
• hematopoietic cell therapy (HCT)
• Her2/neu
• host-versus-graft (HvG)
• T-body
• tumor

1. Adoptive cell transfer (ACT) therapy

Adoptive cell transfer (ACT) is a procedure in which therapeutic lymphocytes are transferred to patients with the aim of treating disease. Traditionally ACT has used antigen-specific T effector cells (Teff) for the treatment of cancer or viral infections [1], yet recently our group has demonstrated the utility of regulatory T cells (Treg) in ACT for the treatment of autoimmunity [2]. Although the basic principle is the same for ACT therapies, the specifics, such as the transferred cell type, the antigen specificity and the protocols used, can very considerably. Recently, several ACT-based treatments against cancer have shown promising results in early clinical trials against melanoma, neuroblastoma and chronic lymphocytic leukemia (CLL) [3-6].

2. Chimeric antigen receptors (CARs)

One ACT modality, pioneered by our laboratory, uses CAR-redirected T cells, which we call ‘T-bodies’ [7]. CARs are fusion receptor proteins usually composed of single-chain antibody fragment linked to signaling motifs capable of activating T cells. A CAR-redirected T cell can be activated either through a MHC-restricted endogenous T cell receptor (TCR) or through the CAR in an MHC-independent manner. Virtually any antibody could be used in the context of a CAR so T cells could be redirected in this manner against any target antigen. A key advantage of this approach over conventional T cells is the ability to incorporate multiple signaling domains into the modular CAR thus endowing it with supra-physiological potency [7,8]. The potency of this approach has recently been demonstrated in early clinical trials with impressive results [3,5].

An approach that has recently emerged is to redirect antigen-specific T cells via a CAR such that each T cell has two specificities. In addition to being able to target two different tumor-associated antigens concurrently, this approach can also be used to augment ACT in other ways. In a pilot clinical trial Pule et al. redirected both polyclonal and Epstein-Barr virus (EBV) specific T cells with a GD2-specific CAR, and showed that virus-specific T cells persisted longer in vivo and provided more therapeutic benefit than polyclonal T cells [3]. The rationale for this approach was that latent infection with EBV would provide in vivo stimulation to the virus specific T cells thus promoting their survival and function, which would improve their response against the tumor.

3. Allogeneic ACT

One limitation of ACT therapies is that each treatment is specifically fabricated for one individual. Most ACT therapies rely on obtaining the patient's own cells, manipulating them in some manner and then infusing them back into the patient. One exception to this general rule is a modality called donor lymphocyte infusion (DLI), which is practiced following hematopoietic stem cell transplantation (HCT) [9]. DLI therapy involves transferring lymphocytes from the original donor into the patient, and is fairly effective against chronic myeloid leukemia (CML). Because the host is already engrafted with the donor's hematopoietic system, there is no host-versus-graft (HvG) response, and the transferred cells are not rejected. However basing allogeneic ACT on allogeneic HCT is still problematic for two reasons: first, MHC-matched donors are available for only a fraction of the patients, and even then there is a significant risk that the allogeneic cells will attack the patient and cause severe graft-versus-host-disease (GvHD). Second, because the cells must be transferred from the original donor, this therapy could not be used as an ‘off-the-shelf’ treatment for cancer.

Another way to delay rejection of allogeneic cells is to ablate the host's immune system with either irradiation or chemotherapy, and then rescue the host with autologous HCT instead of allogeneic HCT. However, even when using intense preconditioning protocols the surviving host lymphocytes may be able to reject the allogeneic cells albeit at a much slower rate. The problem with this approach is that allogeneic T cells can cause lethal GvHD in immunocompromised hosts. Nevertheless two recent studies used this approach in preclinical murine models of cancer with each study relying on a different strategy to prevent GvHD [10,11]. One of the studies showed that allogeneic cells provided less benefit than syngeneic ones [11], while in the other one no comparison was made [10].

4. The effect of modulating lymphocyte migration on immune responses

A key step in every immune response is the trafficking of lymphocytes to their destination. The first step of trafficking is to exit from the lymph node (LN), and into the blood, a step called lymphocyte egress. One mechanism which accounts for lymphocyte egress from the LN, is chemotaxis of lymphocytes in response to a gradient of the chemical sphigosine-1-phospahte (S1P), which is present at high levels in the blood and low levels in the LN [12]. FTY720 is an agonist of the S1P receptors, and can prevent lymphocyte egress, thus effectively sequestering lymphocytes in the LN [13]. Because of its sequestering properties, FTY720 can potentially inhibit any immune reaction and indeed it has shown efficacy in clinical trials for kidney transplantation [14]; it recently received approval for the treatment of multiple sclerosis (MS) [15], and it has shown efficacy in preclinical models of GvHD [10,16].

5. Universal allogeneic ACT

Recently, our group has demonstrated that allogeneic CAR-redirected T cells could be used as a universal treatment for cancer [17]. Because the CAR is not MHC-restricted, redirected cells could be transferred to any patient and act as 'universal effector cells'. We proposed that mild preconditioning would delay the rejection of allogeneic T cells and allow them enough time to destroy the tumor on one hand while on the other hand their eventual rejection would prevent the development of GvHD. Using this approach in a mouse model of experimental metastases, fully MHC-mismatched allogeneic T cells redirected with a human EGFR2/neuregulin (Her2/neu)-specific CAR were able to provide as much therapeutic benefit as syngeneic cells. We demonstrated that the GvH reactivity of the allogeneic T cells caused extensive proliferation, augmenting the anti-tumor response, but that the allogeneic T cells were indeed eventually rejected and no GvHD developed.

Our second working hypothesis was that allogeneic ACT presents a unique opportunity to employ lymphocyte sequestration to augment the efficacy of immunotherapy rather than suppressing it. The rationale for this hypothesis is that both the HvG and GvH reactions originate in the lymph nodes where resident T cells encounter MHC-mismatched antigen-presenting cells (APC), undergo massive proliferation, and then egress out of the lymph nodes into peripheral organs. ACT on the other hand, introduces cells directly into the blood, circumventing egress, and allowing the cells to reach the tumor directly. We therefore administered FTY720 for the first ten days following irradiation and adoptive cell transfer. Consistent with its ability to inhibit the HvG response, adding FTY720 to the treatment protocol significantly augmented the survival and the efficacy of allogeneic ACT while autologous ACT was not affected because syngeneic cells are not subject to HvG.

6. Expert opinion

CAR-redirected allogeneic T-cells could be used as 'universal effector cells', and hold great promise as an 'off-the-shelf' cellular treatment for cancer. This approach harnesses the power of the GvH response to drive the proliferation of transferred cells, and augment the efficacy of the treatment. Furthermore, using the lymphocyte sequestering agent FTY720 enhances therapeutic efficacy even further such that allogeneic cells were superior to syngeneic ones.

Using CAR offers several advantages over conventional TCR-based approaches aside from the obvious advantage of being MHC-unrestricted. First, tumors frequently downregulate MHC proteins or ligands of co-stimulatory receptors and the CAR approach circumvents this problem altogether. Second, improvements to the CAR can enhance its potency [18] leading to stronger anti-tumor responses without boosting the GvH response, which is mediated by the endogenous TCR. Finally, it should be noted that CAR-based approaches are not mutually exclusive with TCR-based ones because antigen specific T cells can be transduced to express a CAR, as has been recently demonstrated in an early clinical trial [3].

While this work provides only a proof of principle, we strongly believe that this approach is widely applicable for two main reasons. First, by using CAR against different antigens, many different cancers could be targeted of both solid and hematological origin, as has been recently shown in pilot clinical trials against neuroblastoma and CLL [3,5]. Second, the HvG and GvH effects occur independently of the tumor and, since we were able to overcome this problem in our study, the same approach could be used regardless of tumor type.

While in our model we were able to attain significant therapeutic benefit without serious adverse effects, administration of allogeneic T cells to patients following lymphodepletion does carry the risk of serious GvHD as was seen in some patients which experienced GvHD following transfusion of unirradiated leukocytes (TA-GvHD) [19]. Clinical translation would of course require further verification as well as additional safety measures such as suicide genes, which could be used to eliminate allogeneic cells should GvHD develop [20,21]. One possible consequence of this therapy would be that the host would be immunized to allogeneic cells, which could potentially limit future platelet transfusion. However, when one considers that one of the approved treatments for hematological malignancies is allogeneic HCT, which is both risky and not available for all patients due to lack of compatible donors, there clearly is a niche for this type of treatment.

In conclusion, allogeneic adoptive cell therapy may become the treatment of choice both because of its obvious technical and economical advantages over autologous ACT, and due to its greater efficacy.

Declaration of interest

The authors state no conflict of interest and have received no payment in preparation of this manuscript.