Cancer vaccines differ from conventional vaccines for infectious diseases in as much as they are given after the antigenic insult similar to the old hyperimmune serums. Their main targets are tumor antigens (1). Historically the first vaccines consisted of extracts from pyogenic bacteria or mycobacteria which elicit an immune response (1). An example would be the use of BCG vaccine for equine sarcoid, which when given intra-tumorally stimulated an antitumor immune response in a paracrine fashion. Two recent advances have helped the cause of cancer vaccines and these are; immunological understanding of lymphocyte activation and cytokines; and gene transfer technology (1). The targets of cancer vaccines have been identified in some cancers, mainly melanomas and renal carcinomas. Molecular identification of cancer antigens have identified a number of antigenic sites these are; tissue specific antigens, reactivated embryonic gene products, mutated gene products and viral gene products. The main response to cancers is via the cell mediated immune response. The cells that are responsible for CTL (cytotoxic Tcell lysis) are CD8⁺ T-lymphocytes (1). In order for there to be a good antitumor effect three important processes must occur; adequate presentation of the cancer antigen must be presented by the MHC-I on the tumor cell or antigen presenting cell; in addition this must occur with costimulation by such complexes as B7-CD28; local elaboration of cytokines (1). Insufficient tumor antigen presentation on MHC-I or lack of costimulation leads to cell anergy, which a mechanism potentially responsible for induction of tolerance to "self" antigens. Other reasons could be lack of recognition of tumor antigen on MHC-I or inadequate CD8⁺ T-cell activation by "helper" arm (CD4⁺ T-cells) (1).

Research in the direction of cancer vaccines has followed the following courses (1):

 Genetically altered whole-cell tumor vaccines

 Allogenic

 Autologous

 Antigen-Based vaccine strategies

 Vaccination with MHC-1 binding peptides

 Recombinant Viral Vaccines

 Recombinant Bacterial Vaccines

 Nucleic Acid Vaccines

 Dendritic cell vaccines

 Heat shock proteins as carriers of antigen

Historically whole-cell tumor vaccines were used together with adjuvants. These adjuvants often consisted of BCG or Corynebacterium parvum. Although some studies showed moderate clinical responses, these were uncharacterized. The way these vaccines worked appeared to correlate with delayed hypersensitivity reaction to the autologous tumor cells (1). Currently genetically altered whole-cell tumor vaccines seem to be more successful and the response can be characterized. The technique of gene transfer utilizes in most cases a viral vector to implant genes into the cancer cell to enhance the Tcell immune response. Two methods have been studied. The first is genetically altering the cancer cell to produce cytokines (e.g., GM-CSF) which enhance the attraction of antigen presenting cells (APC) such as dendritic cells to the tumor. At the University of Florida we are currently exploring the use of cytokine modified tumor vaccine for cats with vaccine associated soft-tissue sarcomas. The second method is to genetically alter the cancer cell to become a professional APC, expressing the ability to present tumor antigen on major histocompatibility complexes (MHC). These techniques use both ex-vivo and in-vivo methods. A number of technical problems exist, in broad terms these are; limiting viral induced genes or gene products to the tumor (2); and switching off production of the induced cytokine. Because growing cancer cells ex-vivo can be difficult, bystander cells such as human vero cell can be utilized. These cells are histoincompatible and therefore for example, the cats own immune system would ultimately destroy the cells thus switching off the cytokine production. This technique has been employed in the veterinary field to treat vaccine associated sarcoma in cats (3). Results using this technique were promising. Limitations of autologous whole-cell tumor vaccines relates to the expansion of tumor cells from individuals, which can be technically difficult, in addition whole cells constitute an inefficient source of antigen which is required for tumor rejection (1). Currently, allogenic whole-cell tumor vaccines maybe more effective, they are prepared ex-vivo from existing tumor cell lines. Research has shown that the technique is effective (1).

Another approach is antigen-based vaccines. These strategies have three main requirements to be successful designs, these are; identification of common antigens recognized by T-cells and expressed by the majority of cancer patients; identification of a single antigen that can serve as a tumor rejection target in-vivo; development of recombinant vaccine strategies that can generate antigen specific immunity (1). Identification of common antigens has been successful, mainly for cancers such as melanoma. The techniques used to identify common antigens include genetic and biochemical approaches (1).

As mentioned previously tumor antigens fall into four main categories these are:

 Tissue specific antigens, for example these are commonly shared antigens among malignant melanoma cancers.

 Reactivated embryonic gene products. Mage-1 is a common product found in melanomas. These products must be specifically recognizable by T cells.

 Mutated gene products. These can be oncogene or tumor suppressor gene products e.g., p53.

 Viral gene products. Examples of these would be Burkitt's lymphoma and Epstein - Barr virus, possibly FeLV could also be targeted in feline lymphoma.

Since most work has been done on melanomas it is hoped that other tumors will be similar and have common antigens. Two approaches have evolved in the development of antigen-based tumor vaccines, these can be divided into DNA-based vaccines that deliver the gene encoding the antigen and peptide- or protein-based vaccines that deliver antigens pulsed on to APCs or mixed with adjuvants (1). An example of DNA-based vaccine include the injections of "naked" or plasmid DNA intramuscularly using needle-free jet delivery device. Currently a phase one clinical research trial in dogs injected with DNA plasmid encoding for a mouse melanoma antigen was found to be effective with durable remission of metastatic disease in one dog (4). Vaccines employing APCs pulsed with tumor associated antigen have also been successful (5). Once again melanomas in dogs were used. Recombinant viruses and bacterial vaccines have also been developed. Utilizing the natural cytotoxic effect that viruses have which attracts APCs is one mechanism. Viruses would have to be restricted to the tumor for this to be effective. It has been known for number years that spontaneous remission have occurred in people that have had viral infections or vaccination. In addition modified viruses with introduced genes can target bone-marrow derived dendritic cells causing them to express tumor associated antigen or encoded costimulatory molecules and enhance the tumor killing effect. Recombinant bacterial vaccines involve the use of genetically engineered bacteria. Bacterial strains of Salmonella, BCG and Listeria monocytogenes have two characteristic which are beneficial. They posses the ability to infect the host via the enteric route, thus providing for oral vaccine use. Secondly, recombinant L. monocytogenes has a two-phase intracellular life cycle. The bacterium infects monocytes and macrophages and occupies phagolysosomes. The bacterium secretes Listeria lysin O which destabilizes the lysosome and allows the bacteria to enter the cytoplasm. Based on the known ways that APCs present antigen either via MHC-1 (cytosolic phase) or MHC-2 (phagolysosome phase) the bacteria can be manipulated to present tumor specific antigen either via MHC-1 to CD-8 T-lymphocytes or via MHC-2 to CD-4 T lymphocytes.

Dendritic cell (DC) vaccines have recently received attention as cancer vaccine because of their ability to stimulate antigen specific T-cells (1;6). Dendritic cells are 100 to 1000 more potent than macrophages in stimulating a response in T-cells, this is due to their higher expression of MHC, cytokine and costimulatory receptors. Based on these characteristic DC, may serve as good presenters of tumour specific antigen to CTL. Currently a number of studies have investigated loading DC with autologous cancer antigen ex-vivo and vaccinating the patient. A newer technique is to use fusion proteins, which are combination of cancer cells and DC.

Heat shock proteins (HSP) can also be used as carriers of antigen (1). HSP are proteins produced by genes that are induced when intracellular conditions promote denaturing of proteins e.g., heat. They act to protect other proteins and aid in the refolding of denatured proteins. HSP can serve as natural biological adjuvants; they also have the capacity to bind a wide array of proteins. The HSP that are an extract from a specific tumour e.g., HSP70 are unique to that tumor and can stimulate an immune response (1). In conclusion cancer vaccine will in the near future become more of a reality as more tumor antigens are identified. Results from research in experimental animals and human and animal clinical trials are promising, however these vaccines are likely to be used as an adjunct to therapies currently used in oncology.

Definitions

 Autologous--same animal

 Allogeneic--same species

 Xenogeneic--different species.

References

1.  Jaffee EM, Pardoll DM. Cancer-Specific Vaccines. In: Mendelsohn J, Howley PM, Israel MA, Liotta LA, editors. The Molecular Basis of Cancer. Philadelphia: W.B. Saunders Company, 2001: 573-588.

2.  Weld KJ, Mayher BE, Allay JA, Cockroft JL, Reed CP, Randolph MM et al. Transrectal gene therapy of the prostate in the canine model. Cancer Gene Ther 2002; 9(2):189-96.

3.  Quintin-Colonna F, Devauchelle P, Fradelizi D, Mourot B, Faure T, Kourilsky P et al. Gene therapy of spontaneous canine melanoma and feline fibrosarcoma by intratumoral administration of histoincompatible cells expressing human interleukin-2. Gene Ther 1996; 3(12):1104-12.

4.  Wolchok JD, Houghton AN, Bergman PJ. Of Mice and Men (and Dogs): Xenogeneic DNA Vaccines for Melanoma. 2002. Ref Type: Personal Communication

5.  Rodriguez-Lecompte JC, Gyorffy S, Majumdar A, Gauldie J, Wan Y. Dendritic cells transduced with adenovirus expressing human TERT or gp100 tumor-associated antigens as cancer vaccines. 2002. Ref Type: Personal Communication

6.  Onaitis M, Kalady MF, Pruitt S, Tyler DS. Dendritic cell gene therapy. Surg Oncol Clin N Am 2002; 11(3):645-60.