Selective stimulation of neoantigen-specific T cells by immunotherapy

From the Riddell Lab, Clinical Research Division

Immune checkpoint blockade (ICB) is a powerful immunotherapy that has revolutionized the treatment of many cancers, including metastatic melanoma. Unlike traditional chemotherapies or targeted inhibitors, ICB yields long-term benefits for many melanoma patients. Understanding the mechanisms of response to ICB, and, inversely, why some patients fail to respond to this therapy, will inform efforts to improve response rates amongst patient populations. In an informative case study, Dr. Joshua Veatch, a physician scientist in the Riddell Lab in the Fred Hutch Clinical Research Division and practicing medical oncologist at the Seattle Cancer Care Alliance, and colleagues performed in-depth characterization of anti-tumor immune responses following ICB in a patient with metastatic melanoma. Their findings, which reveal selective mobilization of immune responses against cancer-specific mutations, were recently published in the Journal for ImmunoTherapy of Cancer.

T cells are important effectors of anti-tumor immunity, mediating both direct and indirect mechanisms of tumor cell killing. A T cell recognizes potential targets through its T cell receptor (TCR), the product of a complex succession of DNA sequence rearrangements that occur during T cell development, which confers upon it a unique specificity to a precise molecular target, or antigen. T cells can recognize tumor targets through two broad classes of antigens: 1) neoantigens: non-self-proteins that are generated through mutations occurring specifically in tumor cells, and are not found elsewhere in the body; 2) tumor associated antigens: self-proteins which are aberrantly expressed in tumor tissues at very high levels, and may be expressed in normal tissues only at low levels or sometimes only during development. This means that T cells with the capacity to recognize self-proteins exist in the body, but mechanisms have evolved to restrain their responses to avoid collateral damage associated with inappropriate or overactive T cell function. For example, immune checkpoints are pathways that serve to dampen T cell activation by directly counteracting activating signals received through the TCR. Many tumors exploit these checkpoint pathways by over-expressing the proteins that trigger them or by recruiting other cell types that express checkpoint proteins. ICB functions by inhibiting checkpoint proteins, thereby releasing tumor-reactive T cells from their suppressive effects.

Killer T cells surround a cancer cell.
Killer T cells surround a cancer cell. National Institutes of Health (NIH), CC PDM 1.0

There has been some debate regarding the relative contributions of neoantigen-specific vs. self-antigen specific T cell responses in anti-tumor immune responses stimulated by immunotherapy. To address this, Dr. Veatch and colleagues collected T cells from blood and tumor biopsy samples taken from an ICB-responsive patient with metastatic melanoma prior to and after initiation of therapy. Using whole exome sequencing of tumor tissue collected prior to therapy (compared to healthy tissue), tumor-specific mutations were identified and, from these, candidate neoantigens were predicted. They then stimulated T cells from the blood with peptides encompassing each predicted candidate neoantigen, and evaluated T cell reactivity by the production of interferon (IFN)-γ. Using this strategy, the authors were able to detect T cell responses against 4 different tumor-specific neoantigens. By expanding these T cell populations against each neoantigen, followed by targeted DNA deep sequencing, they were able to identify TCR sequences (called ‘TCR clonotypes’) associated with neoantigen specificity. Finally, they used parallel approaches to identify TCR clonotypes against 5 different tumor associated self-antigens.

With TCR clonotypes in hand, the authors performed TCR sequencing of blood and tumor samples collected before and after ICB in order to track antigen-specific T cell responses over the course of therapeutic response. Encouragingly, prior to therapy, neoantigen-specific clonotypes were enriched 150-fold in the tumor compared to the blood (0.71% and 0.005% of TCR templates present, respectively). The frequency of neoantigen-specific clonotypes increased robustly in the blood (5-fold) following ICB. Total T cell transcript levels increased dramatically in the tumor post-ICB, and neoantigen-specific clonotypes were modestly enriched in frequency there as well (30% increase). Interestingly, self-antigen-specific clonotypes did not preferentially localize to the tumor and did not expand after ICB in either tissue. These findings indicate that neoantigen-specific T cells already present in the tumor prior to therapy expanded in response to ICB and may have played a greater role in the therapeutically induced anti-tumor immune response than self-antigen-specific T cells in this patient.

“While we are limited by this being a single patient, this shows that the T cells that see mutations can already be there before treatment, waiting to be unleashed, and are harder to find in the blood,” said Dr. Veatch, “and even though we are only beginning to scratch the surface of all the antigens that can target T cells to cancer, in this patient’s great response to treatment the mutations seemed to be more important T cell targets than the self-antigens we looked at.” The authors are careful to note that they cannot rule out a role for T cell responses against self-antigens that were not identified in their assays, and it will be important to compare responses across groups of patients. While tumor-reactive T cells may be quite rare and difficult to isolate from the blood, this work presents a case study in which the identification of neoantigen- and self-antigen-specific clonotypes, and thus the ability to track them over the course of therapy, was possible. These findings begin to shed light on the underpinnings of patient responses to ICB and will inform strategies to advance immunotherapy for improved patient benefit.

This work was funded by the National Institutes of Health and generous support from the Bezos and Lembersky Families.

UW/Fred Hutch Cancer Consortium members Scott Tykodi and Stanley Riddell contributed to this work.

Veatch JR, Singhi N, Jesernig B, Paulson KG, Zalevsky J, Iacucci E, Tykodi SS, Riddell SR. Mobilization of pre-existing polyclonal T cells specific to neoantigens but not self-antigens during treatment of a patient with melanoma with bempegaldesleukin and nivolumab. J Immunother Cancer. 2020 Dec;8(2):e001591. doi: 10.1136/jitc-2020-001591. PMID: 33298619; PMCID: PMC7733177.