Cell-mediated immunity to cancer principally relies on the specificity of the interaction between T cells and tumor antigens presented by tumors and antigen-presenting cells (APC), although other types of cells may be the ultimate effectors. Many tumor antigens have been discovered in recent years. The immune system has been shown to recognize antigens that are shared between cancer and normal tissue, as well as those that are created by genetic alterations in the tumor. In immunocompetent hosts, many tumors may be recognized and destroyed at an early stage, and those that progress may be forced to evolve mechanisms to evade an immune response. Much effort has been directed toward developing strategies of immunizing individuals so that their immune systems can recognize and respond to established malignancies.
It has long been accepted that the immune system evolved to protect the host from microbial invasions. It is harder to explain how the immune system might evolve to fight cancers. In fact, as recently as this past decade, immunologists questioned whether the immune system could recognize cancer cells and whether the immune system was capable of mounting a response that could reject tumors.
In the setting of a microbial invasion, the immune system must only differentiate "self" from "foreign" in order to direct its attack. Cancers, however, arise from the host's own tissues, and thus almost all the genes expressed by tumors are also expressed by normal tissues. Can the immune system attack "altered self" while sparing "self"? Another question is why the immune system would evolve to fight tumors. Although microbial invasion can affect an individual of any age, malignancy is more common after childbearing, and thus the ability to protect against cancer might not exert the same sort of evolutionary pressure as the ability to protect against infection.
Nevertheless, immune recognition of cancers has now been well documented in humans. Also, it has been shown that the immune system is capable of rejecting cancer in experimental systems. Further, evidence shows that the immune system may have a role in protecting the host from developing tumors (the concept of immune surveillance), and that in order for cancer to progress beyond microscopic loci, it must acquire mechanisms to evade an immune response.
Historically, the immune system has been said to have two arms. Humoral immunity was originally described as immunity that could be transferred from one individual to another through the transfer of the serum fraction of blood and is mediated by antibodies. Cellular immunity was thus defined as immunity that could be transferred through the cells of blood. An antitumor immune response can involve either or both arms of the immune system. This article focuses on cellular immunity to cancer.
II. CELLS OF THE IMMUNE SYSTEM
Specific immunity resides in B lymphocytes and T lymphocytes. The ability to react and respond to specific antigens is called adaptive immunity. B lymphocytes are responsible for the secretion of immunoglobulin, or antibodies. Antibodies directly bind to antigen and usually are directed at extracellular or cell surface antigens, such as bacteria, toxins, or viral proteins expressed on the surface of cells. T lymphocytes can only recognize antigens that are presented to the T cell by another host cell (called an antigen-presenting cell) that is genetically identical. The T-cell receptor is expressed on the surface of T cells and has a similar structure to immunoglobulins. However, T-cell receptors do not directly bind to antigens.
Rather, T-cell receptors bind to peptides (between 8 and 16 amino acids in length) presented on the surface of APCs; these peptides are processed from longer proteins inside APC to form the antigens recognized by T cells. Peptides (or in some cases lipids or carbohydrates) are presented to T-cell receptors by major histocompatibility (MHC) molecules on APCs. MHC molecules in humans are called HLA and are extremely polymorphic, distinguishing one individual from another (except for identical twins). A T cell activated by recognizing antigen can either destroy the host cell presenting the antigen or secrete immune substances known as cytokines that stimulate inflammation and recruit other immune cells. T lymphocytes are responsible for recognizing intracellular pathogens (because the antigens recognized by T cells must be processed inside the cell). B cells develop in the bone marrow and T cells develop in the thymus.
Each B or T cell can recognize a single antigen. B or T cells capable of broadly recognizing host tissues are normally deleted from the immune repertoire during development or paralyzed (a process called anergy), but B or T cells capable of recognizing "nonself" are allowed to fully develop. However, B cells or T cells that do not strongly recognize self but can bind to specific types of tissues in the body may be allowed to develop.
T lymphocytes can be further divided into CD4+, or helper, T cells and CD8+, or cytotoxic, T cells. CD4+ T cells recognize antigens presented by MHC class II (class II MHC) molecules. Class II MHC molecules present antigens derived from proteins broken down in lysosomal/endosomal compartments within cells. The cellular compartments sample molecules from the cell surface and outside the cell. Class II MHC is expressed by only specialized APCs that are capable of activating T cells. CD4+ T cells have been called helper T cells because their primary response when activated is to secrete cytokines that enhance the response of CD8+ T cells, B cells, inflammatory cells (especially macrophages), and other immune/ inflammatory cells. In contrast, CD8+ T cells, recognize antigens in the context of class I MHC molecules.
Class I MHC molecules, which are present on all cells, present primarily cytoplasmic and nuclear antigens. The primary response of activated CD8+ cells is to lyse cells presenting the specific antigens, and thus they are known as cytotoxic T cells (CTLs). A third type of lymphocyte is neither a B cell nor a T cell and is known as a natural killer (NK) cell. These cells can directly lyse a host cell without recognizing a specific antigen and can also produce cytokines.
Although the exact mechanism by which NK cells recognize cells is not entirely understood, there are signals for both activating and inhibiting NK cells. One of the best understood inhibitory signals is self class I MHC molecules. NK cells have been described as cells that recognize "missing self." Thus virus-infected cells or tumor cells that lose MHC expression in an effort to evade a T-cell response are susceptible to lysis by NK cells. NK T cells form another set of lymphocytes that can mediate tumor immunity. These cells express both NK and T-cell markers and are able to kill target cells like CTLs and also produce cytokines.
Adaptive immunity, driven by B cells and T cells, develops in response to antigens. The frontline defense of the immune system does not require prior recognition of antigen, which is called innate immunity. NK cells are part of the innate immune system. Macrophages, eosinophils, basophils, and mast cells are other cells involved in innate immunity, participating in inflammatory responses, and will not be further discussed here. A critical step of the adaptive immune response by T cells is presentation of antigen. The initial response of T cells to antigens that leads to activation is called priming. Although any APC can present antigenic peptides, only "professional" APCs can activate T-cell responses. Once T cells are primed, they can recognize nonprofessional APCs. Professional APCs are specifically designated to activate T lymphocytes.
Thus antigen presented on an APC can license a T cell to respond to subsequent exposures to antigen presented on other host cells. In contrast, antigen presented to a naive T cell by a host cell that is not a professional APC can potentially tolerize that T cell such that it is rendered anergic to subsequent exposures to that antigen (one of the mechanisms for peripheral tolerance). The dendritic cell is the most potent and best described professional antigenpresenting cell. Dendritic cells are derived from the bone marrow and reside in epithelial surfaces, lymph nodes, and the spleen. They can capture and process antigens from a variety of sources, including bacterial and viral infections and tumors.
III. ANTIGEN PROCESSING AND PRESENTATION
T cells can only recognize antigen presented to them by MHC molecules. MHC molecules have been divided into two classes. Almost all cells express class I MHC, with red blood cells and spermatozoa being exceptions. The class I MHC molecule is translated into the endoplasmic reticulum where it associates with peptide fragments of cytoplasmic and nuclear proteins. The class I MHC/peptide complex is then transported to the cell surface, where it is accessible to T cells. Specialized cells, including B cells, macrophages, and dendritic cells, express class II MHC. The class II MHC molecule is routed to specialized lysosomal/ endosomal compartments after translation. There it associates with fragments of proteins that are also routed to that compartment for degradation or arrive in that compartment from outside the cell through phagocytosis. The class II MHC/peptide complex is also transported to the cell surface.
Humans have three sets of genes for class I MHC molecules, known as HLA-A, HLA-B, and HLA-C. There are also three sets of genes for class II MHC molecules, known as HLA-DR, HLA-DP, and HLADQ. Because each of these genes is highly polymorphic and each individual has two alleles for each of the class I MHC molecules and two alleles for each of the two chains of the class II MHC molecules, the number of possible combinations of MHC molecules is enormous. Because each polymorphic MHC molecule has a different affinity for peptide, a peptide that can bind to the HLA-A1 molecule may not bind to the HLA-A2 molecule. Thus the sets of peptides that can be seen by the immune systems of two genetically different individuals are distinct.
In order for an antigen to be presented by an MHC molecule, it must first be processed. For class II MHC, this is accomplished mainly by the proteolytic degradation of proteins in the endosomal/lysosomal compartment by resident proteolytic enzymes. For class I MHC, this is slightly more complex, as a full-length protein present in the cytoplasm or nucleus must be converted into peptide fragments before transport into the endoplasmic reticulum for binding to MHC molecules. The proteasome is a complex of enzymatic proteins responsible for degrading intracellular proteins.
Proteins destined for the proteasome are cleaved into peptides that then can be transported into the endoplasmic reticulum by the transporter associated with antigen processing (TAP). It should be noted that cleavage of a protein by the proteasome into certain peptide fragments and transportation of certain peptide fragments by TAP into the endoplasmic reticulum are also responsible for the repertoire of peptides that can be seen by CD8+ T cells.
In order to activate a T-cell response, antigen must be presented in the context of costimulatory signals, only present on professional APCs. How do tumor antigens then activate an immune response if cancer cells are not professional APCs? Tumor antigens from necrotic or apoptotic cancer cells can be taken up and presented by APCs. This process is called crosspresentation.
By the process of phagocytosis or other fluid uptake, tumor antigen can access the class II MHC compartment of an APC, but it is not obvious how exogenous tumor antigens might be presented by the class I MHC processing pathway (which typically involves cytoplasmic and nuclear proteins). One possibility is that exogenous peptides can bind to heat shock proteins, which are stress proteins produced in abundance by many cancer cells. Heat shock proteins bind to receptors on professional APCs and can efficiently deliver peptides into the class I MHC pathway.
This process may be one mechanism for cross presentation. In addition, heat shock proteins can activate professional APCs, leading to more efficient stimulation of T cells. Thus tumor antigens released from damaged or dying cells can be presented to the immune system by professional APCs.
IV. T-CELL RESPONSE
T cells recognize antigen on the surface of APCs in the context of self-MHC molecules. The T-cell receptor confers the specificity of this response. Each T cell has a unique T-cell receptor that interacts with the anti- genic peptide bound to an MHC molecule. CD8 or CD4 molecules on the surface of the T cell served as coreceptors. CD8 interacts with class I MHC molecules and CD4 interacts with class II MHC independently of peptide binding; thus CD8+ T cells recognize peptides presented by class I MHC and CD4+ T cells recognize peptides presented by class II MHC. When a T cell encounters its cognate peptide/MHC complex, the T-cell receptor binds, leading to signals through the T-cell receptor that activate the T cell.
A naive T cell, one that has never encountered its cognate peptide/MHC complex, needs costimulation in order to be fully activated. If it does not receive costimulatory signals, it can be rendered tolerant to subsequent exposures to antigen (a form of peripheral tolerance). Molecules such as B7-1/B7-2 and ICAM- 1 on the surface of professional APCs typically provide costimulation. Costimulation may also be provided by cytokines secreted by the APC or by other cells nearby (see Fig. 1). Once a T cell has experienced antigen in the setting of sufficient costimulation, it may proliferate. Thus a clone of T cells with identical T-cell receptors is generated. They may become effector or helper T cells, and if the signal is sufficient a small subset may become memory T cells. Effector and helper T cells respond to antigen without the need for further costimulation, but are themselves destroyed in the ensuing immune response in a process called activation-induced cell death. Memory T cells can also respond to antigen without the need for costimulation, but these remain quiescent during the initial immune response and are only activated upon repeat exposure to antigen.
FIGURE 1 Activation of a naive T cell by an antigen-presenting cell (APC). A T cell recognizes antigen presented on the surface of an APC. The T-cell receptor binds to a peptide fragment, known as an epitope, complexed to an antigen-presenting MHC molecule. The coreceptors for MHC, CD4, and CD8 on the surface of the T-cell facilitate this interaction. CD4+ T cells recognize epitopes complexed with class II MHC molecules, and CD8+ T cells recognize epitopes complexed with class I MHC molecules. Each T cell has T-cell receptors of one specificity that bind a particular peptide epitope and MHC complex. In order to activate a naive T cell (one that has never before experienced antigen), the antigen/MHC complex must be present within the context of sufficient costimulation. Costimulation can be provided by B7-1 or B7- 2 on the APC binding to CD28 on the T cell, and by LFA-1 on the APC binding to ICAM-1 on the T cell. Costimulation can also be provided by cytokines in the local environment. Mature dendritic cells are professional APCs and thus are able to provide sufficient costimulation.
CD4+ and CD8+ T cells classically have different effector responses. CD4+, or helper, T cells secrete soluble cytokines that act locally and at a distance to enhance the function of other immune cells, such as B cells, CD8+ T cells, or macrophages. Activated CD4+ T cells can be categorized by the profile of cytokines they secrete and the types of cells they stimulate when activated. Th1 helper T cells enhance the function of macrophages and CD8+ T cells through the cytokines they produce, most classically IFN-γ but also TNF-β. Thus Th1 cells are central players in cell-mediated immunity. In contrast, Th2 cells produce a humoral immune response through a set of cytokines, IL-4, IL-5, and IL-10, that enhance B cells and may inhibit T-cell-mediated immune responses. CD8+ T cells are typically cytotoxic T lymphocytes because their classic effector function is to lyse cells.
However, it should be noted that secretion of cytokines, mainly IFN-γ, by CD8+ T cells is also a function of activation. CD8+ T cells lyse their targets through two main mechanisms: one involves the release of cytotoxic proteins known as perforin and granzymes from specialized vesicles, and the second involves induction of programmed cell death through the stimulation of fas on the target cell. Effector CD8+ T cells are armed with lytic granules, and upon activation they can release the contents of these granules. One of these is the protein perforin, which can destroy the integrity of the cell membrane through the formation of pores. These pores allow entry of proteolytic enzymes, including granzymes, which are also released from the lytic granules. Another way cytotoxic T cells arm themselves is through the expression of fas ligand on their cell surface. The interaction of fas ligand on the CD8+ T cell with fas on the target cell induces apoptosis of the target. Some cancer cells expressed fas, although the degree of actual expression of fas in vivo by cancers remains controversial.
V. TUMOR ANTIGENS
T cells can recognize antigens presented by cancer cells, although cancer cells are not typically professional APCs and do not easily elicit T-cell immunity. The greatest advance in cancer immunology has come from the molecular identification of antigens recognized by the immune system. Several competing classification schemes have been proposed to categorize tumor antigens. No one scheme has been widely accepted. The classification used here encompasses most of the tumor antigens identified and is based on the presence or expression of antigens recognized on tumors in normal host tissue (see Table I).
TABLE I Tumor AntigensClass Type Example Alterations or atypical gene products New epitope by mutation p16/INK4A
β-Catenin Altered trafficking by stability CDC27 Psuedogene HPX42B Antisense DNA RU2 Translocation GDP-L-fucose:β-D-galactoside-2--L-fucosyltransferase Cancer-testis antigens MAGE family
NY-ESO-1 Differentiation antigens Melanosomal differentiation antigens Tyrosinase
MelanA/MART-1 Signal transduction molecules HER2/neu
EphA3 Viral antigens Epstein–Barr virus
Human T-cell leukemia virus
A. Alterations or Atypical Gene Products
Cancer results from the accumulation of genetic mutations, so it may seem obvious that at least some of these mutations could be recognized by the immune system. However, the manner in which the immune system detects genetic change would not be easy to predict.
In some cases, a mutation can create a new epitope or increase the immunogenicity of an existing epitope. In the case of genes encoding p16/INK4A, p53, and β-catenin, the immune system can recognize the products of point mutations that are implicated in the pathogenesis of malignancy. In these cases, T cells potentially strike at the heart of genetic lesions that drive the cancer. In addition to recognizing point mutations, T cells have also been shown to recognize new epitopes created by genetic translocations joining two distinct genes in a fusion protein.
In addition to creating a new epitope, a mutation can alter trafficking and stability by its gene product, thus allowing for MHC presentation of an unaltered epitope that is not seen in normal cells. Cancer cells can also present antigens from alternative transcripts, including those from cryptic start sites and alternative reading frames, and even from parts of the genome that are normally invisible, such as pseudogenes and antisense products from the minus strand of DNA.
B. Cancer Testis Antigens
Another class of tumor antigens are not unique to cancer but are only shared with germline cells, which do not express classical MHC molecules and thus do not present these antigens to the immune system. These antigens are operationally tumor specific. Expression of these normally silent antigens is sometimes introduced in cancer cells, perhaps due to changes in the transcriptional regulation such as methylation of DNA.
The prototype of this family is the MAGE-1 protein, which was the first human gene product recognized by CD8+ T cells from cancer patients. The list of cancer-testes antigens includes the MAGE family and related GAGE and BAGE families, along with the NY-ESO-1 antigen.
C. Differentiation Antigens
Antigens shared by cancer cells and their normal cell counterparts are known as differentiation antigens. Prototypical differentiation antigens are the melanosomal differentiation antigens shared by melanoma and normal melanocytes. In fact, antigens characteristic of melanocytic differentiation were determined, in part, from the recognition of melanoma cells by autologous serum. Most melanosomal differentiation antigens are proteins involved in the synthesis of melanin, such as tyrosinase, TRP-1/gp75, TRP2, and gp100/pmel17.
MelanA/MART-1 is another melanoma differentiation antigen of unknown function. Immunity against melanosomal differentiation antigens is, in fact, autoimmunity and destruction of melanocytes as well as tumor cells can ensue. It is thought that immunity against differentiation antigens can develop because these antigens are not expressed in the thymus, so reactive T cells are not deleted and peripheral tolerance is incomplete. In melanoma, which is the best studied cancer type in humans in the field of cancer immunology, differentiation antigens are the most frequent antigens recognized by T cells.
D. Signal Transduction Molecules
Another group of antigens includes the signal transduction molecules, which do not fit in any of the categories just described. These molecules are present in normal tissue but can be overexpressed by cancers. In cancer patients, T cells reactive to HER2/neu, a receptor tyrosine kinase that is overexpressed in several types of cancer, including breast, ovarian, and lung, have been identified. Another tyrosine kinase, EphA3, is recognized by autologous T cells in a human patient.
E. Viral Antigens
Viruses are implicated in the pathogenesis of some types of cancers, including hepatoma (hepatitis virus), lymphomas and nasopharyngeal cancer (Epstein–Barr virus), and leukemias (HTLV). Viral-infected cells can express strong antigens that are recognized by T cells.
VI. IMMUNE SURVEILLANCE AND IMMUNE EVASION
Lewis Thomas originally proposed the concept that the immune system protects the organism against the development of malignancies, known as immune surveillance. This idea fell out of favor in the 1970s, but has reemerged. Experiments with animals having specific deficiencies in their immune systems have shown a higher rate of tumor formation. If it is true that the immune system identifies many tumors at an early stage before they become clinically apparent, then it follows that tumors escaping surveillance can develop mechanisms to evade the immune system. Mechanisms by which the tumors can escape immune attack include loss of MHC molecules, loss of immunogenic antigens, and loss of antigen-processing machinery. It appears that the most common of these mechanisms is the loss of MHC molecules. In fact, many tumors have been shown to have lost expression of some or all of their MHC I genes.
VII. INDUCTION OF CELL-MEDIATED IMMUNITY TO CANCER
The roots of cancer immunology are in the therapy of cancer. Treatment of "tumors" by injection of infected purulent materials is documented in both Eastern and Western ancient medical writings and dates back at least several millennia. It can be argued that the birth of modern tumor immunology began with the treatment of cancer with bacterial extracts by William B. Coley around the end of the 19th century. With a better understanding of the molecular basis of immunity, the field of immunotherapy is rapidly growing. Approaches to the induction of cell-mediated immunity against cancer can be grouped into three broad categories: (1) immune modulation, (2) passive/adoptive immunotherapy, and (3) active immunotherapy/vaccination (see Table II). Most efforts using these strategies are in clinical trials, so our discussion is brief.
TABLE II Strategies for Immunotherapy of Cancer Using Cell-Mediated ImmunityStrategy Type Example Immune modulation Cytokines IL-2
IFN-β Passive/adoptive immunotherapy Transfer of T cells
Transfer of NK/LAK cells Active immunotherapy/vaccination Whole tumor cell/tumor cell preparation vaccines Tumor cells transduced with cytokine
Dendritic cells pulsed with tumor preparations
Dendritic cells pulsed with RNA from tumors
Heat shock protein preparations Antigen-specific vaccines Recombinant protein
Naked plasmid DNA
A. Immune Modulation
The strategy of immune modulation involves attempting to activate the immune system against the cancer through the administration of immunostimulatory substances. In particular, approaches involve systemic treatment with the cytokines IL-2, IL-12, IFN-α, and IFN-β. IL-2 is a primary growth factor for T cells and thus can potentially expand existing T cells that recognize the tumor. IL-12 is a cytokine known to bias the immune system toward a Th1, or cell-mediated, immune response. Recombinant human IL-2 is routinely used in the treatment of melanoma and renal carcinoma. Experimental evidence suggests that cell-mediated immune responses are more potent than humoral immune responses against cancer and that IL-12 can induce tumor regression. However, IL-12 is still being tested in early clinical trials. The interferons IFN-α and IFN-β bind to the same receptor and have similar effects that can enhance cell-mediated immunity. IFN-α and IFN-β both increase expression of the antigen-processing and -presenting machinery, as well as activating NK cells. The interferons are widely used for cancer therapy, including treatment of melanoma, renal cancer, and leukemias.
B. Passive/Adoptive Immunotherapy
Adoptive immunotherapy involves the ex vivo activation and expansion of T cells or NK cells and passive transfer of these cells into the cancer patient. This is an experimental approach with some risks and needs to be explored more intensively. Cells for adoptive therapy can be obtained either from the patient or from a separate individual. In the case of adoptive transfer between different individuals, the goal here is for T cells from the donor to attack the recipient's cancer in a process called graft versus tumor. The MHC haplotype of the transferred T cells, of course, must match the patient to be treated.
Adoptive transfer of T cells against viral antigens expressed by the Epstein–Barr virus has been successfully used to treat lymphomas induced by the Epstein–Barr virus following bone marrow transplantation. Other experimental approaches use T cells expanded from tumor lesions (called tumor-infiltrating lymphocytes by some) and T cells expanded by exposure to specific tumor peptides plus APCs. In addition to T cells, NK cells may be used for adoptive therapy. The entity originally called lymphocyte-activated killer (LAK) cells used in some of the earliest adoptive therapy trials has been shown to consist mostly of activated NK cells.
C. Active Immunotherapy/Vaccination
Investigation into active immunotherapy or vaccination of cancer patients is the most active area of cancer immunotherapy. It should be noted that in tumor immunology, the word "vaccination" is used to mean the deliberate induction of specific immunity to treat a cancer and does not typically indicate that the vaccine is given prophylactically to prevent disease, as is the case in infectious disease. Similar to vaccines against infectious diseases, vaccines against cancer have been developed both where the vaccine does not target any specific antigens and where the vaccine is based on a specific antigen (whole tumor cell/tumor cell preparation vaccines versus antigenspecific vaccines).
1. Whole Tumor Cell/Tumor Cell Preparation Vaccines
The first cancer vaccines were based on cancer cells given with adjuvant. Newer approaches using whole cell vaccines involve tumor cells transfected to express cytokines, costimulatory molecules, or allogeneic molecules. In particular, GM-CSF, which is a central growth factor for professional antigen-presenting cells, is receiving attention. Another strategy involves extracting heat shock proteins from autologous tumor cells and administering that preparation as a vaccine. Heat shock proteins have bound peptides from the tumor cell and, when extracted, can chaperone these antigens into the class I MHC pathway of an APC. Dendritic cells have also been used as an adjuvant with tumor cell preparations. Dendritic cells taken from the host can be pulsed with autologous tumor cell lysate or apoptotic bodies from autologous tumor cells and administered back to the patient.
2. Antigen-Specific Vaccines
The last two decades have been characterized by a search for cancer antigens that can be recognized by the immune system. Vaccines based on these defined tumor antigens are being studied in clinical trials. In some approaches, the whole antigen has been delivered to the patient as a recombinant protein, as DNA (either naked DNA are expressed by recombinant virus), or loaded onto dendritic cells and then administered as a vaccine. Naked DNA and recombinant viruses are particularly attractive strategies because of the ease of making and administering the vaccine. Xenogeneic vaccination, which is immunization with the homologous antigen from another species, has in some cases proven more potent in inducing an immune response than vaccination with syngeneic antigen.
Increased understanding of MHC molecules is allowing better determination of potentially immunogenic epitopes of many tumor antigens. Epitopes for class I MHC molecules are being identified at an increasing rate, and class II MHC epitopes have begun to be discovered. Vaccination with these minimal peptide epitopes with or without adjuvant has followed. These peptides can also be loaded on dendritic cells and administered as a vaccine. Epitopes for some tumor antigens, particularly differentiation antigens, are poorly immunogenic. It is thought that T cells against self-antigens have low or intermediate affinity for these antigens because more potent T cells were deleted during thymic T-cell development. Modifying these epitopes at the amino acid level to make them stronger binders to MHC or to the T-cell receptor has increased the efficacy of peptide-based vaccines in experimental models. These enhanced peptides are called heteroclitic vaccines. Although motifs for epitopes have thus far only been identified for some of the more common human MHC alleles, efforts to identify epitopes for other alleles are ongoing.
The immune system is capable of recognizing and rejecting tumors. Both humoral and cell-mediated immune responses are potentially important for tumor immunity. The cell-mediated immune response is particularly effective in experimental models. Nevertheless, the fact that some tumors grow progressively despite evidence of immune recognition demonstrates the difficulty of controlling large or metastatic tumors with the immune system. Immunotherapy strategies have encountered these obstacles, but ongoing research is aimed at better understanding the principles and more effective manipulation of the cell-mediated immune response against cancer.
Jason S. Gold
Alan N. Houghton
Memorial Sloan-Kettering Cancer Center and Weill Medical School of Cornell University
ANTIBODIES IN THE GENE THERAPY OF CANCER ; DIFFERENTIATION AND THE ROLE OF DIFFERENTIATION-INDUCERS IN CANCER TREATMENT ; IMMUNE DEFICIENCY: OPPORTUNISTIC TUMORS ; MOLECULAR BASIS OF TUMOR IMMUNITY ; T CELLS AGAINST TUMORS ; T CELLS, FUNCTION OF ; TUMOR ANTIGENS
adoptive immunotherapy The strategy of passively immunizing against a tumor by infusing T cells specific to the tumor or natural killer cells activated with cytokines.
antigen-presenting cell A cell that is specialized in activating a T-cell-mediated immune response by presenting antigen in the context of sufficient costimulation to naive T cells.
cancer-testis antigens Tumor antigens that are only expressed on cancer cells and immune privileged germ cell tissue.
cell-mediated immunity Immunity that can be transferred by the cellular fraction of blood but not serum. The specificity of a cell-mediated immune response is determined by the T-cell receptor of T cells.
costimulation Signals that allow activation of a naive T cell upon encountering antigen. Costimulation can be transmitted by the interaction of molecules on the cell surface of the antigen-presenting cell with molecules on the cell surface of the T cell or by cytokines in the local milieu.
differentiation antigen Tumor antigens shared by tumors and similarly differentiated normal tissue. The best described differentiation antigens are melanosomal differentiation antigens, which are shared between melanoma and normal melanocytes.
immune surveillance The concept that the immune system recognizes and destroys incipient neoplasms. Thus hosts with impaired immunity may be more susceptible to cancer, and cancers in immunocompetent hosts must evolve to evade the natural immune response.
MHC molecules Molecules on the surface of cells that present antigen in the form of peptide epitopes so that it can be recognized by T cells. MHC class I molecules typically present cytoplasmic proteins. MHC class II molecules present antigens present in endosomal and lysosomal compartments.
tumor antigens Antigens on tumors that the immune system is capable of recognizing and responding to. Tumor antigens may be present exclusively on cancer cells or may be shared with normal tissue.
T-cell receptors Molecules on the surface of T cells capable of recognizing antigen complexed to MHC. Each T cell has one specificity of T-cell receptors.
vaccination The strategy of actively inducing immunity by giving the host an antigenic substance. Vaccination against cancer can involve administration of whole tumor cells or tumor cell preparations, as well as recombinant or synthetic tumor antigens. In contrast to vaccination against infectious diseases, in the clinical setting, vaccination against cancer usually implies vaccinating against a disease after it has occurred rather than prophylactically.
Banchereau, J., and Steinman, R. M. (1998). Dendritic cells and the control of immunity. Nature 392(6673), 245.
Boon, T., Coulie, P. G., and Van den Eynde, B. (1997). Tumor antigens recognized by T cells. Immunol. Today 18(6), 267.
Bremers, A. J., and Parmiani, G. (2000). Immunology and immunotherapy of human cancer: Present concepts and clinical developments. Crit. Rev. Oncol. Hematol. 34(1), 1.
Delon, J., and Germain, R. N. (2000). Information transfer at the immunological synapse. Curr. Biol. 10(24), R923.
Houghton, A. N. (1994). Cancer antigens: Immune recognition of self and altered self. J. Exp. Med. 180, 1.
Houghton, A. N., Gold, J. S., and Blachere, N. E. (2001). Immunity against cancer: Lessons learned from melanoma. Curr. Opin. Immunol. 13, 134.
Marincola, F. M., Jaffee, E. M., Hicklin, D. J., and Ferrone, S. (2000). Escape of human solid tumors from T-cell recognition: Molecular mechanisms and functional significance. Adv. Immunol. 74, 181.
Pardoll, D. M. (1998). Cancer vaccines. Nature Med. 4(5 Suppl.), 525.
Rosenberg, S. A. (1999). A new era for cancer immunotherapy based on the genes that encode cancer antigens. Immunity 10(3), 281.