Destruction Development of Cancer Vaccines Challenges to Cancer Vaccines References
Interest in activating the immune system to control tumors dates to more than 100 yr ago and is often attributed to William B. Coley who used the inflammatory response to bacterial products as a form of immunotherapy (1). This initiated the era of attempts to nonspecifically activate the immune system with the hope that tumors would be targeted as part of the overall, predominantly inflammatory, response. Following the abandonment of the crude bacterial products in Coley's toxin, other bacteria or their cell wall products such as bacille Calmette-Guein (BCG), Corynebacterium parvum, nocardia rubra, OK-432, and others have been tested. In some circumstances, these inflammatory activators have had success as antitumor agents. For example, BCG has activity in superficial bladder cancer in situ (2). Later, cytokines such as interleukin (IL)-2 that could activate T and natural killer (NK) cells were extensively tested and demonstrated activity in renal cell carcinoma and melanoma. In the past two decades, a new era in cancer immunotherapy has opened with the discovery of how the immune system can recognize and destroy tumors. This has included the description of tumor-expressed antigens that can be recognized by immune effectors and a characterization of how immune effectors are primed to recognize these antigens as a signal to destroy the associated tumor. By harnessing these observations, it has been possible to design vaccines with the necessary ingredients for activating tumor antigen-specific immune responses in vivo.
It has long been realized that many tumors are poorly immunogenic. That is, if they are merely disaggregated and reinjected, they frequently grow unabated and do not activate a protective immune response. Subsequently, it was demonstrated that tumors could be modified to increase their immunogenicity. After immunization with these modified tumors, the immune system could recognize and destroy the original unmodified tumor (3). These data demonstrated that tumors could be recognized by T cells once the T cells had been activated.
It is now known that activation of nave, tumor antigen-specific cytotoxic T cells (CTLs) so that the T cell proliferates and becomes capable of tumor destruction requires at least two signals. The first signal is the presentation of tumor antigen in the form of an 8-10 amino acid epitope within the major histocompatibility complex (MHC) class I (in humans, called human leukocyte antigens [HLA] A, B, C) molecule to the T-cell receptor (TCR). The second signal is engagement of the T-cell CD28 molecule with costimu-latory molecules (CD80 and CD86) on the stimulating cell. Without this costimulation, antigen-specific T cells enter a state of anergy. Although tumors frequently express MHC molecules containing tumor peptides, they do not possess the necessary costimulatory molecules (4). In vivo, these two signals are provided by antigen-presenting cells such as dendritic cells (DCs). Therefore, strategies for activating immune responses against tumors involve either delivering antigen to antigen-presenting cells (APCs) or modifying tumors so that they can provide both signals for T-cell activation.
3. components of the immune system involved in tumor destruction
A model of antitumor immunity that only includes cytolytic T cells is of course too simplistic. First, the activity of CTL is modified by positive influences from CD4+ helper T cells and negative influences from regulatory and suppressor T cells. Second, non-MHC-restricted tumor killing by NK cells may also be important for tumors that escape recognition by downregulation of MHC. Third, the other arm of the immune system, humoral or antibody-mediated immunity, may play an important role. Indeed, in some vaccine strategies, outcome correlates most with antibody titer. Fourth, cytokines and chemokines, elaborated by both immune effectors and tumors, may dramatically modify the activity of cellular components of the immune response.
Cytotoxic CD8+ T cells recognize MHC-restricted peptides on the tumor surface via their T-cell receptor. In contrast, mature NK cells express an array of germline-encoded activating and inhibitory receptors on their surface and it is the balance between these receptors (5) that determines the cytolytic activity of the NK cell. In particular, the inhibitory receptors such as the human killer cell Ig-like receptors (KIRs) recognize HLA-A, -B, and -C with bound peptide and suppress NK-cell lysis of target cells that express these class I MHC molecules, but allow the lysis of class I negative cells. Thus, tumors that downregulate MHC become less susceptible to attack by CTLs, but more susceptible to attack by NK cells.
CTLs and NK cells share two different mechanisms for tumor-cell killing: the Fas-Fas-ligand (FasL) pathway (6), and the perforin-granzyme pathway (7). FasL, a member of the TNF (tumor necrosis factor) family, is expressed predominantly on activated T cells, macrophages, and neutrophils. Binding of a FasL trimer on the effector cell surface to a Fas trimer on the target cell membrane causes the formation of the death-inducing signaling complex (DISC) around the cytoplasmic chain of Fas. This results in interaction of Fas with the adaptor protein Fas-associated death-domain protein (FADD) and the initiation of a chain of events eventuating in apoptosis of the tumor cell target. In the perforin-granzyme pathway, perforin, a pore-forming molecule stored in cytotoxic granules together with granzymes, is released with the granules upon recognition of a target cell. Perforin monomers insert themselves into the target cell membrane and create pores that cause osmotic lysis of the target cell and allow granzymes to enter the target cell and induce apoptosis through various downstream effector pathways.
CD4+ T-helper (Th) cells promote the activity of other immune cells, by both cytokine release and direct cell-cell interactions. Their TCRs recognize antigen in the form of peptides presented by MHC class II molecules. Since tumors generally do not express MHC II, Th cells are primed by APCs such as DCs or B cells. Different subsets of Th cells, designated Th1 and Th2, have pronounced differences in their effect on the immune response. The Th1 response, characterized by secretion of IL- and interferon-y (IFN-y), is stimulatory for CTLs, whereas the Th2 response, characterized by secretion of IL-4, IL-5, IL-6, and IL-10, is important for isotype switching and antibody production by B cells. Activated Th cells also provide direct signals for maturation of DCs, via expression of CD40-ligand, which binds to CD40 on DCs. CD40L is also an important signal for B-cell isotype switching. Whether a Th cell follows the Th1 or Th2 pathway is dependent on signals from APCs such as IL-12, which skews the response toward Th1. Because numerous studies have demonstrated that the induction of Th1 responses can slow or prevent tumor growth, whereas Th2 responses may permit tumor growth, it is generally thought that anticancer vaccines should contain adjuvants that induce Th1-type cytokines. There is also increasing interest in designing vaccines that contain MHC class II epitopes in addition to MHC class I epitopes to ensure the participation of Th cells in activating the immune response (8).
B cells recognize tumor antigen as epitopes on proteins or carbohydrates via the B-cell receptor, a surface-expressed, monospecific immunoglobulin. Although B cells can be directly activated by large antigens that bind simultaneously to multiple antibody receptors, this T-cell-independent activation results in mainly IgM production and poor memory induction. T-cell-dependent responses, in contrast, usually require two signals for activation, antigen binding to the B-cell receptor and cytokine secretion from Th cells. Antigen either binds to the B-cell receptor or is presented to B cells by APCs such as macrophages or follicular DCs. Antigen is also processed by APCs and presented to Th cells. Furthermore, antigen taken up by B cells can be processed and presented to Th cells via MHC class II molecules. The Th cells supply IL-2, IL-4, IL-5, and CD40-ligand that lead to B-cell proliferation, class switching, and the B cell eventually becomes an antibody-secreting plasma cell or a memory B cell. Although the natural antibody response against tumor cells is weak, monoclonal antibodies are now being used to treat lymphomas and breast cancer successfully. Furthermore, some of the most promising tumor vaccines in later stage clinical trials activate antibody responses. There is now increasing emphasis on attempting to include humoral immunity in anticancer vaccine approaches (9). Similarly, immunologic monitoring of vaccine clinical trials is assuming increasing importance.
Antibody-mediated killing of tumors may occur by antibody-dependent cellular cytotoxicity (ADCC) or complement-dependent cytotoxicity (CDC). ADCC occurs when the Fc region of an antibody interacts with Fc receptors (FcRs), FcRI and FcRIII, on immune effector cells such as macrophages, NK cells, and neutrophils. The effector cell then destroys the tumor by phagocytosis or lysis. In CDC, recruitment of the complement component C1q by IgG bound to the tumor-cell surface triggers a proteolytic cascade that results in cell death in one of two ways. First, the proteolytic cascade may ultimately result in formation of a membrane attack complex that kills the target cell by rupturing its cell membrane. Second, tumor-cell-bound C1q can bind to complement receptors, such as C1qR, CR1 (CD35), and CR3 (CD11b/CD18), on macrophages, NK cells, and neutrophils, triggering phagocytosis or lysis.
Because of their central role in activating adaptive immune responses, dendritic cells (DCs) have garnered the most recent attention in the development of anticancer vaccines. Derived from bone marrow precursors, DCs circulate to peripheral sites (such as the epidermis of the skin where they are called Langerhans cells), capture and process antigen, and traffic to draining lymph nodes where they prime CTL, Th, and B cells. Exogenous antigens delivered in vaccines can enter both the class II and class I MHC pathways within DCs resulting in their ability to activate both CD4+ Ths and CD8+ CTLs. The potency of DCs as T-cell stimulators derives from several factors, the high level of expression of MHC, costimulatory, and adhesion molecules, that permit the formation of strong "immu-nologic synapses" (10) at the site of T-cell-DC interactions. Because only immature DCs efficiently take up and process antigen, whereas mature DCs are the most effective at antigen presentation, the most successful vaccine strategies may be those that interact with DCs to achieve the desired maturation state. Furthermore, DCs can be divided into DC1s and DC2s. DC1s are immunostimulatory, but DC2s are frequently described to be tolerogenic. Therefore, vaccine strategies that can promote DC1 effects over DC2, may be important. Finally, the ability to generate DCs ex vivo has also spawned a large number of studies using antigen-loaded DCs as the cancer vaccine.
Natural killer T cells (NKTs), a subset of lymphocytes that express both T-cell and NK-cell markers, recognize glycolipid antigen (GalCer) presented by the MHC-like molecule CD1d. Although few in number, they can produce large amounts of IL-4 or IFN-y, and promote humoral or cellular immunity. In common with NK cells and CTLs, they kill tumor cells by a perforin-dependent mechanism. NKT are thought to play a role in biasing the immune response toward TH1 or TH2, and have been implicated in immune surveillance and protection against carcinogenesis (11). A regulatory or immunosuppressive subset of NKT cells has also recently been reported. It is still uncertain how best to integrate an understanding of these cells into vaccine approaches, but the availability of the NKT ligand alpha-galactosylceramide (KRN7000) has permitted further study.
As is the case with all biologic systems, the immune response is characterized by the need for regulation to prevent excessive or aberrant immune responses from causing auto-immunity. Several cell types mediate regulation of immune responses. First, CD8+ suppressor T cells, stimulated by DC2s (12), release IL-10 and suppress the reactivity of other T cells. Another recently described regulatory T-cell subset, the CD4+/CD25+ T cells, can mediate suppression by cell-cell contact and this suppressive effect is dependent upon surface-bound transforming growth factor-P (TGF-P) on the regulatory T cells (13). Finally, some signals delivered from APCs to T cells may actually downregulate the immune response. For example, CD80 binding to CTLA-4 on T cells causes downregulation of T-cell activity. Although vaccines may in fact activate potent immune responses, it will likely be necessary to interfere with these regulatory and suppressor influences. Vaccine studies are now beginning to integrate addition manipulations to abrogate regulatory responses including the use of antibodies to deplete CD4+/CD25+ T cells or interfere with CTLA-4 activation.
Cytokines are critical for signaling between immune effectors. In vivo, cytokines released at the site of tissue damage from tumors, such as tumor necrosis factor-a (TNF-a), CD40 ligand (CD40L), and IFN-y, cause DC maturation. Granulocyte macrophage-colony-stimulating factor (GM-CSF) released from macrophages, and T and B cells, also is involved in DC differentiation. DCs produce cytokines such as IL-12, IL-15, and IL-18, which drive Th1 responses. In turn, the Th1 cytokines IFN-y, IL-2, and TNF-a appear to be important for the generation of antitumor, CTL activity as described above.
Because of the importance of cytokines to in vivo immune responses, they have been added to vaccine strategies as adjuvants or to augment the immune response following immunization. GM-CSF has been used as an adjuvant for cancer vaccines because of its ability to activate DCs. IL-2 has been administered following immunization because it induces T-cell proliferation and cytokine production, and augments TCR-mediated and TCR-independent cytolysis. IL-12 (14) has gained increasing recognition as a possible adjunct to active immunotherapy because of its pleiotropic effects including stimulation of Th1 cells, NK-cell proliferation, and augmentation of B-cell IgG production. IFN-a has been shown to enhance humoral immunity and promote isotype switching, in part through its effects on DCs. IFN-y can augment the immunogenicity of tumor cells through the upregulation of MHC class I expression, which may be important for tumors with downregulated MHC expression.
4. development of cancer vaccines
4.1. Differences Between Tumor and Infectious Disease Vaccines
There are a number of conceptual differences between existing, clinically effective vaccines approved for use in humans and those that are being developed and tested in the setting of cancer immunotherapy. In general, vaccines approved for clinical use in the United States are applied as prophylaxis against infectious pathogens, such as hepatitis B virus and pneumococcus. Generally, these vaccines must be administered prior to the onset of an infection (as primary prophylaxis) in order to be effective, although there are rare exceptions such as the treatment of rabies (in which both specific immunization and immune globulin are administered to achieve postexposure prophylaxis). In contrast, cancer immunotherapy vaccines presently are aimed at the eradication of either a microscopic or, more commonly, macroscopic burden of tumor cells.
A second major difference is that antigens targeted by existing vaccines are those of viruses or other pathogens, and hence are highly immunogenic, foreign antigens. No vaccines currently in clinical use target human antigenic epitopes, whether they comprise mutated or nonmutated antigens. That clinically significant immune responses can even be induced by vaccination against tumor-associated antigens (TAAs) is based on evidence derived primarily from animal models rather than from human trials. Therefore, the specific methodology for effectively immunizing humans against a self- or mutated self-antigen is unknown.
Whereas most existing vaccines approved for clinical use are believed to act through the generation of protective humoral immunity, most experts believe that this will not be the case for cancer vaccines. Rather, the induction of tumor-specific cellular immunity will likely be necessary in order to induce a clinically effective immune response. Though the leading hypothesis is that a TAA-specific cytotoxic T-cell response is a critical component of this process, no vaccines approved for routine use in patients are known to exert their clinical benefit primarily through this mechanism. Ironically, although there are no clinically proven antitumor vaccines currently in standard use, two effective monoclonal antibody preparations are currently available for the treatment of malignant disease (trastuzumab and rituximab), challenging dogma that induction of successful antitumor immunity in humans depends primarily on a cellular rather than humoral host response.
4.2. Necessary Ingredients for a Successful Cancer Vaccine
Although cancer vaccines differ from vaccines against infectious organisms in their therapeutic as opposed to prophylactic use, they share some of the same requirements for success. First, there must be an antigen expressed by the tumor and immune effectors that can recognize it as their signal to initiate the cascade of events that lead to tumor destruction. Second, vaccines must present the antigen in a manner that leads to immune activation instead of anergy. Third, they must provide or recruit other sources of cytokines to promote a more vigorous immune response. The major difference between the two types of vaccine approaches is that for tumor vaccines, the target antigens are usually self antigens, not foreign ones, and the immune system is generally predisposed to anergy against self-antigens. Therefore, it has been more difficult to merely provide tumor antigens to the immune system and induce a potent antitumor response. The following sections describe how these requirements may be met for tumor vaccines.
As described above, although most tumors appear to be nonimmunogenic, they in fact express antigens, either native, surface-expressed proteins or carbohydrates, or processed peptides bound within the groove of MHC molecules, that can be recognized by the immune system. Although some antigens are unique to their tumors such as idiotype produced by malignant B cells or viral antigens from virally induced malignancies (Chapter 5), most are "self" antigens either overexpressed or mutated in the tumor. Theoretically, tumor antigens may be derived from any cellular protein. There is no consensus on whether only identified antigens or undefined pools of antigens would be preferred for a cancer vaccine. Vaccines with pools of undefined antigens (such as tumor-cell vaccines) are attractive because they permit immunization against a wider array of antigens increasing the chances that one or more of the antigens will represent a true rejection antigen in the recipient. Nonetheless, they are difficult to standardize and may face more difficult regulatory hurdles during later phases of development. Vaccines that use one or a small number of pure, defined antigens (such as peptide epitopes) are easier to produce and qualify. Immunologic monitoring is also simplified. Therefore, a major effort is ongoing to discover more antigens. Although they were previously identified by testing cDNA libraries of tumors for those that encoded antigens that could be recognized by T cells cloned from patients with malignancies, newer methods of screening for antigens have now been developed (see Chapters 2 and 3).
4.2.2. Delivery of Tumor Antigens in Cancer Vaccines
Merely delivering an antigen by itself is unlikely to activate a potent T-cell-mediated immune response. The antigen must be delivered in such a way that it is presented along with the necessary costimulatory molecules. This has been accomplished by mixing the antigen with an inflammatory adjuvant (Chapter 8), a carrier protein, or a cytokine such as GM-CSF to increase the exposure of the antigen to antigen-presenting cells (APCs). Also, plasmids or viral vectors containing genes for the antigen and costimulatory molecules may be administered so that when they are taken up by APCs, they can modify the APCs to express the antigen and higher levels of the costimulatory molecules. Injecting genes for costimulatory molecules directly into tumors or modifiying tumor cells to express these genes and reinjecting the tumor as a vaccine can also increase the exposure of the antigen to the immune system. Finally, directly loading DCs with tumor antigens increases the likelihood that the antigen of interest will be processed and presented. The diverse array of strategies that accomplish these goals are described in Chapters 10 through 24. Combinations of strategies, most notably by prime-boost immunizations (Chapter 9), may induce the greatest immune responses. Although adoptive immunotherapy is not a vaccine strategy, the cells delivered are frequently stimulated in vivo or in vitro with vaccines and there is increasing interest in vaccinations following adoptive transfer of T cells. Therefore, Chapter 24, "T-Cell Adoptive Immunotherapy," has been included in this text.
Chapters 25 to 33 describe the clinical results for therapeutic vaccines in the most common malignancies. Although it is too early to determine the ultimate role for cancer vaccines, the results do provide an increasingly clear picture of the challenges that require attention. First, it will be necessary to identify from among the many strategies a few vaccines with enough promise to warrant large-scale clinical trials. This will require novel clinical trial designs and intermediate markers of activity such as immunologic assays to determine which induce the most potent antigen-specific immune responses. Recent attempts to reach a consensus on the immune assays to use and how to interpret them (15) should simplify comparison across different studies. Clinical trial design is discussed in Chapter 34. Immunologic assays are described in Chapters 35 and 36. Second, the level of immune response detected by these assays is still fairly low. If one were to assume that the magnitude of the T-cell response necessary to clear viral infections is similar to the magnitude required to destroy tumors, then most cancer vaccines activate T-cell responses two or more orders of magnitude less than is necessary. Third, tumors possess a variety of mechanisms for evading even a high-level T-cell or antibody response. Tumor escape is addressed in Chapters 6 and 7. Finally, before a vaccine can be administered to patients, it will require considerable regulatory scrutiny to ensure that it is safe and effective. Although the regulatory requirements for infectious-disease vaccines have been honed over many years, the use of cellular vaccines poses new issues for the Food and Drug Administration and other regulators. Regulatory requirements for cell-based vaccines are discussed in Chapter 37.
1. Coley WB. The treatment of malignant tumors by repeated inoculation of erysipelas: report of ten original cases. Am J Med Sci 1893; 105:487-511.
2. Meyer JP, Persad R, Gillatt DA. Use of bacille Calmette-Guerin in superficial bladder cancer. Postgrad Med J 2002; 78:449-454.
3. Van Pel A, Boon T. Protection against a nonimmunogenic mouse leukemia by an immunogenic variant obtained by mutagenesis. Proc Natl Acad Sci USA 1982; 79:4718-4722.
4. Abken H, Hombach A, Heuser C, Kronfeld K, Seliger B. Tuning tumor-specific T-cell activation: a matter of costimulation? Trends Immunol 2002; 23:240-245.
5. Radaev S, Sun PD. Structure and function of natural killer cell surface receptors. Annu Rev Biophys Biomol Struct 2003; 32:93-114.
6. Daniel PT, Wieder T, Sturm I, Schulze-Osthoff K. The kiss of death: promises and failures of death receptors and ligands in cancer therapy. Leukemia 2001; 15:1022-1032.
7. Trapani JA, Smyth MJ. Functional significance of the perforin/granzyme cell death pathway. Nat Rev Immunol 2002; 2:735-747.
8. Zeng G. MHC class II-restricted tumor antigens recognized by CD4+ T cells: new strategies for cancer vaccine design. J Immunother 2001; 24:195-204.
9. Reilly RT, Emens LA, Jaffee EM. Humoral and cellular immune responses: independent forces or collaborators in the fight against cancer? Curr Opin Investig Drugs 2001; 2:133-135.
10. Bromley SK, Burack WR, Johnson KG, Somersalo K, Sims TN, Sumen C, et al. The immunological synapse. Annu Rev Immunol 2001; 19:375-396.
11. Brutkiewicz RR, Sriram V. Natural killer T (NKT) cells and their role in antitumor immunity. Crit Rev Oncol Hematol 2002; 41:287-298.
12. Gilliet M, Liu YJ. Generation of human CD8 T regulatory cells by CD40 ligand-activated plasmacytoid dendritic cells. J Exp Med 2002; 195:695-704.
13. Morse MA, Clay TM, Mosca P, Lyerly HK. Immunoregulatory T cells in cancer immunotherapy. Expert Opin Biol Ther 2002; 2:827-834.
14. Lee SM, Suen Y, Qian J, Knoppel E, Cairo MS. The regulation and biological activity of interleukin 12. Leuk Lymphoma 1998; 29:427-438.
15. Keilholz U, Weber J, Finke JH, Gabrilovich DI, Kast WM, Disis ML, et al. Immunologic monitoring of cancer vaccine therapy: results of a workshop sponsored by the Society for Biological Therapy. J Immunother 2002; 25:97-138.
Was this article helpful?