While vaccines are primarily thought of in terms of their use for prevention of infectious diseases, they can potentially be used to prevent or treat cancer. This manuscript explores the rationale for vaccines and immunotherapies for cancer from both the scientific and the global needs perspectives. Pathogens that are aetiologic agents of certain cancers provide perhaps the most obvious successful examples of the prophylactic utility of vaccines (such as the hepatitis B vaccine) to prevent not just the infectious disease (hepatitis), but the potential subsequent cancer (hepatocellular carcinoma). The use of monoclonal antibodies illustrates the effectiveness of the immune system for cancer therapy. In addition, the increased understanding of the role and mechanisms of the immune system in the processes of immune surveillance, as well as of its failure during immunosuppression, have yielded better insights into how to design cancer vaccines and immunotherapies. Examples of targets for cancer vaccines will be discussed, as will the challenges and few successes in this arena.
Vaccines to prevent and immunotherapies to treat cancer have been the focus of significant efforts. While often not listed in the top of priorities for global health, there are a variety of reasons for including cancer as a focus of discussion during this meeting on the acceleration of vaccines. This manuscript will explore the rationale for making cancer vaccines from both a scientific perspective and a global needs perspective. The targets for such vaccines will be discussed along with providing an explanation for the challenges and few successes in this arena.
The immunologic rationale for trying to develop cancer vaccines is as shown in box 1. The most successful rationale to date has been to prevent infection with pathogens that can result in the development of cancer. Two such human vaccines have been made; these target hepatitis B virus (HBV) and human papilloma virus (HPV). HBV infection can result in hepatocellular carcinoma, and the global introduction of a preventive vaccine, the recombinant hepatitis B surface antigen (HBsAg) has been important for both the prevention of hepatitis and liver cancer. The HPV vaccine, consisting of a recombinantly made L1 protein that forms a virus-like particle, is strain specific. There are over 100 strains of HPV, with different strains being responsible for causing warts, and others being oncogenic as reviewed in Schiller & Lowy  and Jansen & Frazer . HPV infection is responsible to varying degrees for several cancers, such as cervical, anal, penile, and head and neck cancers. Although there are cofactors such as smoking for head and neck squamous cell cancer, HPV infection is thought to be responsible for about 500 000 cancer deaths globally each year . A challenge for making a prophylactic vaccine is that there are many strains of virus. Nevertheless, by targeting the strains that are considered to be the most prevalent of the cancer-causing strains, the vaccine is intended to prevent around 70 per cent of the cases of cervical cancer by preventing infection with just two oncogenic strains, HPV 16 and 18 . One of the HPV vaccines is also effective for preventing warts caused by two non-oncogenic strains of HPV.
While the majority of pathogens that can cause cancer (table 1) are viruses, Helicobacter pylori, which has a higher incidence in poorer countries, is an important cause of gastric cancer , and schistosomiasis may also play a role in cancer as well [7,8]. The mechanism for the induction of cancer can be transformation of cells by the pathogen, but this does not occur in all cases. It is thought that the chronic inflammation may play a role in the induction of cancer  for pathogens such as hepatitis B and C, and H. pylori. Nevertheless, whether or not the pathogen causes cellular transformation, prevention of infection would prevent cancer and any clinical disease caused by the infection.
Macfarlane Burnett first described the concept of immunosurveillance as a mechanism whereby the body's immune system could recognize and destroy neoplastic cells before they multiplied to the point of causing clinical cancer . This concept was supported by the observations in immunosuppressed patients described below, as well as by anecdotal reports of spontaneous tumour regressions. Other anecdotal reports of satellite tumours disappearing following biopsy or partial excision of the main tumour  raised the idea that the surgical procedure caused inflammation that stimulated the immune responses, leading to destruction of the distant lesions. Thus, the development of cancer has been considered by some to be a failure of the immune system to keep up with the destruction of the neoplastic cells either owing to an inadequate immune response, or owing to the tumour escaping from those responses, much in the same manner as pathogens mutate antigens and escape immunity.
Rationale for cancer vaccines.
Rationale for cancer vaccines: why do we think they will work?
— prophylaxis against pathogens that result in cancer
— observation that certain cancers arise in immunosuppressed states
— success of monoclonal antibodies
— limited demonstrated clinical efficacy of immunotherapeutics
A rationale for the development of cancer vaccines is the observation that when the immune system is compromised, a variety of cancers can develop. The most striking example of this may be in patients infected with HIV who go on to develop AIDS, because the development of certain cancers was one of the aspects of the clinical presentation of the patients that led to the recognition of the emergence of a new syndrome and then the pathogen . While Kaposi's sarcoma and non-Hodgkin's lymphoma were among the cancers first noted for their frequency, invasive cervical cancer has now also been added to the list of conditions which constitute the very definition of the clinical syndrome of AIDS .
In patients immunosuppressed following solid organ transplantation, the occurrence of malignancies is also increased . Particular cancers are noted with higher frequency, including lymphomas, skin cancer, Kaposi's sarcoma and lung cancer. The incidence of lymphomas is such that a particular syndrome has even been named: post-transplant lymphoproliferative disorders . It has also been observed that the incidence and severity of the cancers are related to the extent of immunosuppression. The incidence of cancer in patients who suffer from certain hereditary (i.e. primary) immunodeficiencies is also increased, notably with Hodgkin's disease and non-Hodgkin lymphoma, seen in association with infection with oncogenic viruses, such as Epstein–Barr virus (EBV) .
The development of products that stimulate immune cells of either the adaptive or innate immune systems provides evidence for the capabilities of immunotherapy after a tumour has grown to the point of causing clinical disease. The cytokine IL2, a stimulator of cytolytic T cells, has been licensed for the treatment of melanoma and renal cell cancer . Bacillus Calmette–Guérin (BCG), a vaccine for the prevention of tuberculosis, is licensed to prevent recurrence and progression of bladder cancer. BCG is administered intravesically, and while the exact mechanisms of action are not fully known, it appears to be multi-factorial involving cytokines and innate responses .
5. Monoclonal antibodies
The successful use of monoclonal antibodies for a variety of cancers has demonstrated that even when one arm of the immune system is used in isolation, it can be effective for therapy of a variety of cancers [18,19]. It is worth noting, however, the long developmental times needed to bring the simple concept of targeting tumours with highly specific antibodies to fruition as useful therapies, perhaps not least owing to the need to find critical tumour antigen targets. These targets for which monoclonal antibodies have been effective immunotherapies have not only been proteins that are unique tumour neoantigens, but also include the growth factor VEGF (vascular endothelial growth factor), EGFR (epidermal growth factor receptor), HER2/Neu (human epidermal growth factor receptor 2) and cell markers such as CD20 and CD33 [18,19]. Thus, the mechanisms for the efficacy of the monoclonal antibodies may include mechanisms such as blockade of angiogenesis in the case of targeting VEGF, or targeting a growth factor receptor, in the case of antibodies against EGFR, rather than simply an antibody-mediated killing of a cell.
6. Licensed cancer vaccines
A limited number of products have been approved for use in various countries (reviewed in ), although not usually in multiple countries. These have been based on a variety of approaches, but have mainly involved administration of cellular products rather than traditional off-the-shelf biologicals. An example is sipuleucel-T which was recently licensed for the treatment of advanced prostate cancer . It involves taking antigen presenting cells from a patient, stimulating them with tumour antigen and granulocyte-macrophage colony stimulating factor (GM-CSF) and growing them in vitro, then re-infusing them into the patient. This therapy is obviously more complex than an off-the-shelf vaccine, and likewise is quite expensive with the proposed cost per patient being US$93 000 per treatment. In contrast, a recently licensed veterinary vaccine for the treatment of dog melanoma  is a plasmid DNA vaccine, consisting of DNA encoding a tumour antigen. In this case, the antigen is actually the human version of tyrosinase, and it is thought that the effective immune response may be due in part to the xenogenic nature of the antigen enabling the dog's immune system to break tolerance to the dog tumour antigen, to thus mount an effective response .
7. Designing cancer vaccines: antigens
A key element of any vaccine in addition to the delivery system, or biological type of vaccine (for example, plasmid DNA, recombinant protein and cells), is of course the antigen. The types of antigens include those that are (i) unique to the pathogen, if the cancer has an infectious aetiology, (ii) unique to the tumour, (iii) over-expressed on the tumour compared with expression on normal cells, or (iv) related to the differentiation state of the tumour cell. HBsAg is an effective target for prophylaxis against hepatitis B infection, and thus HBV-mediated hepatocellular carcinoma. For cervical cancer, in addition to preventing the cancer by preventing HPV infection, another possible approach would be to make a therapeutic vaccine that could be used in people already infected with HPV, but who have pre-malignant or malignant disease. That is, following infection with an oncogenic strain of virus, some women go on to develop pre-cancerous lesions known as intra-epithelial neoplasia. These in turn can progress to invasive carcinoma. But any vaccine that used the L1 protein would not be effective for therapy because, once the virus infects a cell and as the cell transforms, the infected cell does not express the L1 protein. By contrast, the HPV proteins, E6 and E7, play a role in the transformation and proliferation of cells (reviewed in ), and thus could be targets for a therapeutic vaccine. Efforts have been made to induce cellular immune responses against these proteins, and thus gene-based vaccines, designed to elicit MHC Class I-restricted T cells are being explored. It is important since these involve the delivery of genes encoding the antigens, for the vaccines to encode mutated versions of the E6 and E7 so that they do not result in the production of the protein with oncogenic activity. Such a therapeutic vaccine against cervical cancer would be particularly useful in developing countries lacking the ability to screen women for the early phases of cancer, in order to treat the disease before it progresses to carcinoma.
Another example of a tumour antigen is MUC-1, which is considered to be a protein that is over-expressed on a variety of tumours . However, MUC-1 is not simply over-expressed, but its expression is no longer confined to the apical surfaces of cells, and additionally has aberrant glycosylation that may make MUC-1 a better target for immunotherapy .
8. Immune responses to cancer
The key advances in developing cancer vaccines have been in the area of understanding the immune response to cancer. The simplest concept of a cancer vaccine is that one simply needs to find the right antigen and vaccine delivery system that is capable of eliciting antibodies and T cells that can destroy the tumour cells. However, it is now clear that the tumour is not a passive target for the immunity. Instead, it is capable of downregulating antigen expression, of mutating to lose the target antigen, and of making itself less susceptible to killing by the immune cells . In addition, both the tumour and the surrounding stroma can regulate the immune response by the production of factors, and even as a result of the hypoxia that can develop if the tumour outgrows its blood supply. This process is known as immunoediting [26,27], and is divided into the three stages of tumour elimination, equilibrium (when the tumour and immune response both exist with the latter not being able to eliminate the tumour cells) and escape. The escape of the tumour from the immune response occurs, not simply as a result of factors produced by the tumour and stroma that decrease immune responses, but also because of immune cell dysfunction, and suppression mediated by regulatory T cells . It has thus been proposed that effective cancer immunotherapeutic vaccines will need to be not simply potent delivery systems of antigens, but combined with other interventions such as increasing other immunogenic types of cell death and blocking the immunosuppressive activities of the tumours and stroma. While this may make the actual vaccine more complicated, it should be remembered that the treatment of cancer is already multi-modal, and chemotherapy frequently uses a combination of agents.
9. The global need for cancer vaccines
While cancer therapies are often thought of as treatments for resource-rich countries, given the often exceedingly high costs of therapies, there are a variety of rationales for developing vaccines for cancer that can be deployed throughout the world. First of all, while the incidence of cancers is decreasing in certain individual countries, the global incidence of cancer is increasing, and the total numbers of cases and deaths are enormous with 12 667 500 new cases occurring in 2008 (table 2). This is because of a variety of factors including the rise in rates of smoking in many countries, and the increase in lifespans so that more people are surviving to an age where cancer is more likely to occur. Different cancers predominate in different countries and regions, due in part to differences in the presence of aetiologic agents and cofactors. These include not only different rates of smoking, but also such factors as differences in rates of infection with H. pylori, and differences in diets (with salt and pickled foods playing a role in the high rates of gastric cancer in Japan and Korea).
There is a need to develop cancer vaccines for both the prevention and therapy of cancers. Key challenges are finding the antigens and delivery systems to make effective and potent immune responses, particularly in the immunotherapeutic setting. This is complicated by the fact that the immune response is a dynamic process whereby immune cells, the tumour and stroma all interact to both increase and decrease the tumour-specific immunity. However, these biological challenges need to be addressed as the world's population grows and as certain cofactors act to increase cancer incidence in many regions despite advances in prevention (e.g. not smoking), early detection and current modalities of treatment. Vaccines may be the best means to prevent and treat cancer in resource-poor settings.
One contribution of 16 to a Discussion Meeting Issue ‘New vaccines for global health’.
- This journal is © 2011 The Royal Society