Introduction

For cancer cells to be recognized by the immune system, they must express non-self or altered self-antigens. The concept that tumor cells express antigens, which are recognizable as non-self by the immune system, has been supported by experimental evidence. However, despite the efforts that have been directed at developing good immunotherapy strategies for cancer, this approach is still far from efficient.

Tumour Antigens and the Immune System

As with other antigens, Tumour Specific Antigens (TSAs) can elicit both protective and suppressive immune responses. B cells serve the host by the production of TSA reactive antibodies. These antibodies may help to define TSAs and aid in diagnosis but tend not to be protective for the host. Where they have been shown to have a protective function is through their ability to activate complement pathways to induce complement mediated lysis or cause antibody-dependent cell mediated cytotoxicity (ADCC). Macrophages, neutrophils, NK cells and some lymphocytes can function as ADCC cells. One exciting area of development of this approach is the use of anti-idiotype vaccines. For example, in lymphoma, the clonal expression of antibody can be used as an idiotype to generate anti-idiotype antibodies for therapy.

Research has identified that the cell-mediated arm of the immune system probably offers the greatest protection against tumour-associated antigens. Macrophages and NK cells can effect tumour cell killing via non-antibody dependent means and in a non-MHC restricted fashion. However, cytotoxic T lymphocytes (CTLs) are the most effective mediators of tumour rejection. CTLs are TSA specific and become activated following antigen presentation by antigen presenting cells (in association with MHC Class II molecules) and subsequent further activation via helper T cells. CTLs or tumour infiltrating lymphocytes (TILs) are then able to recognize tumour cells expressing TSAs in association with MHC class I molecules. Some tumour cells also express MHC class II molecules and can themselves act as antigen presenting cells. The role of the CD4+ T helper cells is to provide the correct cytokine environment to drive cell mediate responses and activation of the CD8+ TILs.

The Role of T-Regulatory Cells

Naturally occurring regulatory T cells (T-regs) have been shown to suppress immune responses to self-antigens, thereby limiting autoimmunity. Although T-regs are heterogeneous they characteristically express CD25 and secrete the immunosuppressive cytokine IL-10. In the case of tumours, where immune responses to self-antigens are beneficial and lead to elimination of the tumour, such suppressive activity is actually detrimental to the host. Manipulation of T-regs holds great promise for the immunotherapy of cancer. Several studies, performed using rodent models, indicate that T-regs cells inhibit effective anti-tumour immune responses and that their removal promotes tumour rejection. The increasing number of studies of T-regs in patients with cancer also point to a role for these cells in promoting disease progression. In light of evidence showing that T-reg cells suppress anti-tumour immune responses in mice, preliminary studies in humans using the markers (CD25, foxp3 and IL-10) in conjunction with functional T-reg assays investigated the frequency of these cells in cohorts of healthy controls and patients. Overall, the conclusion that T-reg frequency is increased in malignant disease has been verified both in tumour samples and in peripheral blood. Many mouse studies have shown that depleting CD25⁺ cells promotes immune responses induced by tumour cells or antigen-pulsed dendritic cells as well as immune-mediated rejection of tumour cell-lines. More impressively, intra-tumoural injection of CD4-specific antibodies promoted regression of established tumours. A potential problem associated with the use of CD25-specific antibodies is the simultaneous depletion of conventional CD25⁺ effector T cells, whose loss may compromise the beneficial effect of depleting the T-regs. Other conventional approaches have included the use of standard anti-cancer chemotherapeutic drugs and also inhibitors of COX-2. Both of these are used already in the treatment of canine tumours. An alternative approach would be to transiently inhibit IL-10 as this is a key molecule. The critical point is that inhibition of T-regs can only be short-term so as not to elicit problems of autoimmunity.

Non-Specific Immunotherapy

Immunotherapy is treatment by immunological means and can be divided into active and passive forms. In active immunotherapy, the patient's own immune system is stimulated to respond to the tumour, whereas in passive immunotherapy, there is adoptive transfer of immune cells. Many early protocols involving immunotherapy were based on non-specific immunostimulation using whole cell vaccines and/or injectable BCG vaccine and also muramyl tripeptides as adjuvants to more classical approaches.

Cytokines and Anti-Tumor Immunity

The discovery of cytokines led to early clinical trials injecting protein into cancer patients in an attempt to ameliorate advanced disease. Cytokine proteins essentially act in an autocrine or paracrine fashion. Consequently, in these early trials it was difficult to achieve doses that have biological effect without causing toxicity to the host. In alternative strategies, there have been attempts to isolate tumour-infiltrating lymphocytes (TILs) from biopsy material. TILs are CD8+ cells that are tumour specific and can be expanded ex-vivo by incubation with a cocktail of cytokines including IL-2. Once expanded, these TILs are delivered back to the patient in an attempt to target micrometastatic disease. This has proved useful in some circumstances, but is cumbersome and time consuming. Researchers have tried to establish similar effects by expanding patient dendritic cells by addition of cytokines and/or priming with (or pulsing) with tumour antigens (Dendritic Cell Vaccines).

Vaccination Strategies

It seems likely that effective cancer immunotherapy will require a combination of approaches that may include delivery of cytokine molecules and/or co-stimulatory molecules and/or tumour specific antigens. The advent of technologies such as phage display for identifying tumour specific antigens may hold exciting keys for the development of novel therapeutics. Several canine antigens have been identified which may be tumour specific (e.g., gp100 in melanoma) or more universal (VEGFR2, TERT).

Currently, the only commercially available cancer vaccine for animals has been produced by Merial for canine melanoma. Its license is very restricted and can only be used by specialist oncologists. Long term prospective studies are required to identify the true impact of this vaccine, but early results are encouraging. Despite this, it would appear, as with many immunotherapy strategies, that the greatest chance of success is where there is low tumour burden in the patient.

Stem Cells and Immunotherapy

Over the past 5 years there has been increasing acceptance of the "stem cell" basis of cancer development. In this, cancer is considered to be initiated in somatic stem cells or long-lived cells (e.g., memory B cells). The major significance of this is that these stem cells are highly resistant to the effects of conventional chemotherapy and radiation treatments. However, there is some preliminary evidence from mouse brain tumour models that these cells may offer greater immunogenicity than their more differentiated progeny. If this is the case, then one could envisage a new class of cell-based vaccines developed from more primitive cell types.

Why Does Immunotherapy Fail

From the described processes of tumour immunology described above, it is clear that some cancer cells are able to evade immune recognition.

 Class I MHC molecules may be down-regulated on tumour cells so that they cannot form complexes of processed tumour antigen peptides and MHC molecules required for CTL recognition. Very few tumour cells express MHC class II molecules and so activation of CTLs relies on successful penetration of the tumour by professional APCs. If this does not occur then there will be sub-optimal activation of CTL responses.

 Even in tumors that express MHC molecules, activation of CTL's also requires a second signal provided by co-stimulatory molecules CD28 and B7. Some tumours fail to express B7 and therefore fail to elicit CTL activation and lead to a situation regarded as T cell anergy.

 Some tumours produce cytokine molecules that down regulate the immune system such as TGF-beta.

 Anti-tumour immunity may result in selection of tumour cells within a mass that no longer express immunogenic peptide-MHC molecules. This may arise through mutations or deletions in genes encoding the tumour antigens. Alternatively, immunoselection may favor the growth of tumour cells with deletions or mutations in MHC genes whose products are required to present antigenic peptides. These mutations are more likely in cancer cells because of the high mitotic rate of the cells and also the inherent genomic instability.

 The discovery and characterization of T-regulatory cells suggests that these cells may play a major role in protecting tumours from immune rejection. Possible strategies to reduce T-Regulator cell activity may promote tumour rejection using immunotherapy.