Yoram Reiter

The clinical success of cancer immunotherapy has driven ongoing efforts to identify novel targets that can effectively guide potent effector functions to eliminate malignant cells. Traditionally, immunotherapies have focused on surface antigens; however, these represent only a small fraction of the cancer proteome, limiting their therapeutic potential. In contrast, the majority of proteins within the human proteome are intracellular, yet they are represented on the cell surface as short peptides presented by MHC class I molecules. These peptide-MHC complexes offer a vast and largely untapped resource for cancer immunotherapy targets. The intracellular proteome, including neo-antigens, presents an exciting opportunity for the development of novel cell-based and soluble immunotherapies. Targeting these intracellular-derived peptide-MHC molecules on malignant cell surfaces can be achieved using specific T-cell receptors (TCRs) or TCR-mimicking antibodies, known as TCR-like (TCRL) antibodies. Current therapeutic strategies under investigation include adoptive cell transfer of TCR-engineered or TCRL-T cells and CAR-T cells that target peptide-MHC complexes, as well as soluble TCR- and TCRL-based agents like bispecific T cell engagers. Recent clinical developments in targeting the intracellular proteome using TCRL- and TCR-based molecules have shown promising results, with two therapies recently receiving FDA approval for the treatment of unresectable or metastatic uveal melanoma and synovial sarcoma. This review focuses on the processes for selecting and isolating TCR- and TCRL-based targeting moieties, with an emphasis on pre-clinical and clinical studies that explore the potential of peptide-MHC targeting agents in cancer immunotherapy.

The development and application of human TCR-like (TCRL) antibodies recognizing disease-specific MHC–peptide complexes may prove as an important tool for basic research and therapeutic applications. Multiple sclerosis is characterized by aberrant CD4 T-cell response to self-antigens presented by MHC class II molecules. This led us to select a panel of TCRL Abs targeting the immunodominant autoantigenic epitope MOG_35-55derived from myelin oligodendrocyte glycoprotein (MOG) presented on HLA-DR2, which is associated with multiple sclerosis (MS). We demonstrate that these TCRL Abs bind with high specificity to human HLA-DR2/MOG_35-55-derived MHC class II molecules and can detect APCs that naturally present the MS-associated autoantigen in the humanized EAE transgenic mouse model. The TCRL Abs can block ex vivo and in vivo CD4 T-cell proliferation in response to MOG_35-55stimulation in an antigen-specific manner. Most significantly, administration of TCRL Abs to MOG_35-55-induced EAE model in HLA-DR2 transgenic mice both prevents and regresses established EAE. TCRL function was associated with a reduction in autoreactive pathogenic T-cell infiltration into the CNS, along with modulation of activated CD11b+ macrophages/microglial APCs. Collectively, these findings demonstrate the combined action of TCRL Abs in blocking TCR-MHC interactions and modulating APC presentation and activation, leading to a profound antigen-specific inhibitory effect on the neuroinflammatory process, resulting in regression of EAE. Our study constitutes an in vivo proof of concept for the utility of TCR-like antibodies as antigen-specific immunomodulators for CD4-mediated autoimmune diseases such as MS, validating the importance of the TCR-MHC axis as a therapeutic target for various autoimmune and inflammatory diseases.

The development and application of human T-cell receptor (TCR)-like antibodies (TCRL) recognizing disease-specific peptide-MHC complexes may prove an important tool for basic research and therapeutic applications. Multiple Sclerosis is characterized by aberrant CD4 T cell response to self-antigens presented by class II MHC molecules. This led us to select a panel of TCRL Abs targeting the immunodominant autoantigenic epitope MOG35-55 derived from Myelin Oligodendrocyte Glycoprotein (MOG) presented on HLA-DR2 which is associated with Multiple Sclerosis (MS). We demonstrate that these TCRL Abs bind with high specificity to human HLA-DR2/ MOG35-55 derived MHC class II molecules and can detect APCs that naturally present the MS-associated autoantigen in humanized EAE transgenic mouse model. The TCRLs can block ex vivo and in vivo CD4 T-cell proliferation in response to MOG35-55 stimulation in an antigen-specific manner. Most significant, administration of TCRL to MOG35-55 induced EAE model in HLA-DR2 transgenic mice both prevents and regresses established EAE. TCRL function was associated with reduction of autoreactive pathogenic T cells infiltration into the CNS, along with modulation of activated CD11b+ macrophages/microglia APCs. Collectively, these findings demonstrate the combined action of TCRL Abs in blocking TCR-MHC interactions and modulating APC presentation and activation, leading to a profound antigen-specific inhibitory effect on the neuroinflammatory process, resulting in regression of EAE. Our study constitutes an in vivo proof-of-concept for the utility of TCR-like antibodies as antigen-specific immunomodulators for CD4-mediated autoimmune diseases such as multiple sclerosis (MS), validating the importance of the TCR-MHC axis as a therapeutic target for various autoimmune and inflammatory diseases.

The impressive clinical success of cancer immunotherapy has motivated the continued search for new targets that may serve to guide potent effector functions in an attempt to efficiently kill malignant cells. The intracellular proteome is an interesting source for such new targets, such as neo-antigens and others, with growing interest in their application for cell-based immunotherapies. These intracellular-derived targets are peptides presented by MHC class I molecules on the cell surface of malignant cells. These disease-specific class I HLA–peptide complexes can be targeted by specific TCRs or by antibodies that mimic TCR-specificity, termed TCR-like (TCRL) antibodies. Adoptive cell transfer of TCR engineered T cells and T-cell-receptor-like based CAR-T cells, targeted against a peptide-MHC of interest, are currently tested as cancer therapeutic agents in pre-clinical and clinical trials, along with soluble TCR- and TCRL-based agents, such as immunotoxins and bi-specific T cell engagers. Targeting the intracellular proteome using TCRL- and TCR-based molecules shows promising results in cancer immunotherapy, as exemplified by the success of the anti-gp100/HLA-A2 TCR-based T cell engager, recently approved by the FDA for the treatment of unresectable or metastatic uveal melanoma. This review is focused on the selection and isolation processes of TCR- and TCRL-based targeting moieties, with a spotlight on pre-clinical and clinical studies, examining peptide-MHC targeting agents in cancer immunotherapy.

Adoptive cell immunotherapy with chimeric antigen receptor (CAR) showed limited potency in solid tumors, despite durable remissions for hematopoietic malignancies. Therefore, an investigation of ways to enhance the efficacy of CARs' antitumor response has been engaged upon. We previously examined the interplay between the biophysical parameters of CAR binding (i.e., affinity, avidity, and antigen density), as regulators of CAR T-cell activity and detected nonmonotonic behaviors of affinity and antigen density and an interrelation between avidity and antigen density. Here, we built an evolving phenotypic model of CAR T-cell regulation, which suggested that receptor downmodulation is a key determinant of CAR T-cell function. We verified this assumption by measuring and manipulating receptor downmodulation and intracellular signaling processes. CAR downmodulation inhibition, via actin polymerization inhibition, but not inhibition of regulatory inhibitory phosphatases, was able to increase CAR T-cell responses. In addition, we documented trogocytosis in CAR T cells that depends on actin polymerization. In summary, our study modeled the parameters that govern CAR T-cell engagement and revealed an underappreciated mechanism of T-cell regulation. These results have a potential to predict and therefore advance the rational design of CAR T cells for adoptive cell treatments. See related article on p. 872

Chimeric antigen receptors (CARs) are immunoreceptors that redirect T cells to selectively kill tumor cells. Given their clinical successes in hematologic malignancies, there is a strong aspiration to advance this immunotherapy for solid cancers; hence, molecular CAR design and careful target choice are crucial for their function. To evaluate the functional significance of the biophysical properties of CAR binding (i.e., affinity, avidity, and antigen density), we generated an experimental system in which these properties are controllable. We constructed and characterized a series of CARs, which target the melanoma tumor–associated antigen Tyr/HLA-A2, and in which the affinity of the single-chain Fv binding domains ranged in KD from 4 to 400 nmol/L. These CARs were transduced into T cells, and each CAR T-cell population was sorted by the level of receptor expression. Finally, the various CAR T cells were encountered with target cells that present different levels of the target antigen. We detected nonmonotonic behaviors of affinity and antigen density, and an interrelation between avidity and antigen density. Antitumor activity measurements in vitro and in vivo corroborated these observations. Our study contributes to the understanding of CAR T-cell function and regulation, having the potential to improve therapies by the rational design of CAR T cells. See related article on p. 946

SS1P is an anti-mesothelin immunotoxin composed of a targeting antibody fragment genetically fused to a truncated fragment of Pseudomonas exotoxin A. Delayed responses reported in mesothelioma patients receiving SS1P suggest that anti-tumor immunity is induced. The goal of this study is to evaluate if SS1P therapy renders mesothelioma tumors more sensitive to cytotoxic T-lymphocyte-associated antigen 4 (CTLA-4) immune checkpoint blockade. We evaluated the ability of SS1P to induce adenosine triphosphate (ATP) secretion and calreticulin expression on the surface of AE17M mouse mesothelioma cells. Both properties are associated with immunogenic cell death. Furthermore, we treated these tumors with intra-tumoral SS1P and systemic CTLA-4. We found that SS1P increased the release of ATP from AE17M cells in a dose and time-dependent manner. In addition, SS1P induced calreticulin expression on the surface of AE17M cells. These results suggest that SS1P promotes immunogenic cell death and could sensitize tumors to anti-CTLA-4 based therapy. In mouse studies, we found that the combination of anti-CTLA-4 with intra-tumoral SS1P induced complete regressions in most mice and provided a statistically significant survival benefit compared to monotherapy. The surviving mice were protected from tumor re-challenge, indicating the development of anti-tumor immunity. These findings support the use of intra-tumoral SS1P in combination with anti-CTLA-4.