Cancer Immunotherapy by Checkpoint Blockade

1

Introduction

In the last decades there has come into being, without either flourish of trumpets or serious controversy, a general current of belief in what I have come to call ‘immunological surveillance’. – One can therefore picture a form of surveillance by which the body is being continuously patrolled, as it were, for the appearance of aberrant protein patterns. — M. Burnet

The idea that immune responses may be involved in the survival or destruction of emerging malignant cells has risen as a sort of intuition rather than a solid scientific theory. This optimistic belief arose partly from clinical observations in humans; there are rare but well-documented instances where an established, well-proven malignancy has retrogressed on its own, as in the spontaneous remission of sarcoma in patients who develop erysipelas. Somatic mutations occur constantly in the mammalian body, particularly in malignant cells, and an immune response against these new proteins was considered likely, at least theoretically. In 1960s the observation that a considerable range of systemic manifestations, apparently autoimmune in character, occur in various malignancies led to the hypothesis that malignancy-related immune responses may also involve lymphocyte clones that react to autoantigens.

Since then, a large body of evidence has accumulated for adaptive immune responses against cancers in both experimental animal models and humans. The gene encoding the first human tumor antigen specifically recognized by autologous cytotoxic T lymphocytes (CTLs) was cloned in 1991, followed by the identification of a number of tumor antigens in several types of cancer. Some of these are “neoantigens” derived from viruses or somatic gene mutations, which might or might not be directly related to the oncogenesis. However, many of the tumor antigens, including differentiation antigens, overexpressed/amplified antigens, and germ (testis)-related antigens, show no detectable mutations, and the responses are thus considered self-reactive in principle. Such adaptive immune responses, mostly detected in optimized culture conditions, do not automatically indicate their actual contribution to tumor eradication in the body, nor do they constitute axiomatic evidence for cancer immunosurveillance. However, a more recent study demonstrated a marked increase in tumor development, whether in response to a chemical carcinogen or occurring spontaneously with age, in mice with a completely defective adaptive immune system. Most humans infected with the Epstein–Barr virus (EBV) remain asymptomatic throughout life despite EBV’s strong oncogenic potential for B cells; however, in patients with genetic mutations in T-cell signaling, EBV causes aggressive X-linked lymphoproliferative diseases, including malignant lymphoma. There are a number of case reports of the robust development of otherwise occult tumors in organ-transplantation recipients under continuous immunosuppressive regimens, supporting the importance of the immune surveillance “flagship” in cancer prevention.

2

Cancer immunotherapy

Despite the reassuring evidence of adoptive immune responses against cancer cells, the incidence of cancer in humans increases with age. Cancer continues to pose a major threat to human life. It is apparent that immunosurveillance mechanism is not perfect, and some cancer cells may evolve mechanisms of escape from it. Indeed, cancers that develop in immune-sufficient hosts are far more resistant to the immune system than those in immune-deficient hosts; this phenomenon is called cancer immunoediting. There have been numerous attempts to potentiate cancer immunity in humans in the last few decades. Historically, cancer immunotherapy has been based on two approaches—active immunotherapy to reinforce any endogenous adaptive cancer immunity in the host and passive (adoptive) cell therapy, in which immune effector cells are developed ex vivo and supplied back to the original host.

One form of active immunotherapy is therapeutic cancer vaccination (as opposed to prophylactic vaccination for infection), in which likely tumor antigens are injected into the patient to boost immune responses. There have been clinical trials of cancer “vaccines” consisting of crude tumor cells, recombinant tumor antigen proteins/peptides, genetic vaccines (DNA), or antigen-presenting dendritic cells (DCs) loaded with cancer antigens ex vivo. Cytokines in the IL-2 family (IL-2, IL-15, and IL-21) have also been tested extensively in attempts to boost immune responses to cancer. IL-2 was the first cytokine to be approved by the FDA for treating metastatic renal cancer and melanoma; however, the therapeutic effects are modest, and its highly pleiotropic biological activities can cause severe toxicity.

In adoptive immunotherapy, autologous immune effector cells are activated and expanded ex vivo via procedures that involve anti-CD3 antibodies, IL-2 and other stimulants, or autologous cancer cells, after which the cells are returned to the patient. Although one of the most efficient sources of immune cells has proven to be tumor-infiltrating lymphocytes (TILs) isolated from surgically resected tumor tissues, the activated cells often contain heterogeneous effector populations, including nonspecific cytotoxic cells. In more recent studies, T cells have been genetically transduced with cancer antigen–specific T-cell receptors (TCRs) or chimeric receptors of antibody-variable fragments and TCR-signaling domains (CARs), which enables the T cells to recognize cancer antigens in a non-MHC-restricted manner. Some of these approaches have been effective in patients with certain leukemia, but they are not established as standard therapies yet because of the technical laboriousness involved in treatment and the severe adverse effects such as cytokine-release syndrome.

Recently, a third novel concept of cancer immunotherapy emerged, based on a molecular understanding of immune regulation. Acquired immune responses consist of two distinct phases: an initiation phase, in which naive antigen-specific T-cell clones are robustly expanded to form a sufficient population of effector cell progenies, and an effector phase, in which the differentiated T cells exert such effector functions as antibody production, inflammation, and target-cell destruction. In each phase, the immune responses are inherently and tightly controlled, with checkpoints to prevent excessive immune responses and to prevent attacks on normal tissue cells. The immune receptors CTLA-4 and PD-1, which play crucial roles in cancer immunity, have emerged as promising targets for cancer immunotherapy.

3

Regulation of immune responses