Discovery Guides Areas


Cancer Vaccines
(Released January 2006)

  by Preeti Gokal Kochar  


Key Citations

Web Sites



Review Article

It is believed that any person who lives long enough will eventually get cancer. In the past, cancer was thought to be invincible. Today with advances in surgery, chemotherapy and radiation therapy, mortality has been reduced. Still, world wide 7 million deaths per year can be attributed to cancer, 12.5% of total deaths. More recently traditional treatments have been supplemented with newer treatments like chemoprevention and cancer vaccines.

The immune system has developed to protect the body against invasion by microorganisms and prevent disease. As more information about immune cells is discovered, scientists have realized that the immune system plays a crucial role in preventing cancer.1 The implication of this finding is that, by designing cancer vaccines, it is possible to boost the immune system to enable it to combat cancer more effectively.

Traditional vaccines have successfully prevented infectious diseases like small pox. Recently, great progress has been made in the development of vaccines against cervical cancer, caused by human papilloma virus. However, vaccine development of other types of cancers poses more challenges, since most cancers are believed not to be caused by infectious agents, but rather, defects in cellular proteins. Since these proteins are very similar to those found in normal cells, it is difficult to develop vaccines targeting the cancer cells while sparing normal cells. Indeed, most cancer vaccines will be useful for treating cancers in patients already afflicted and not for preventing cancers. The concept of a universal vaccine against cancer is not realistic since there are many types and many causes of cancers.

The idea behind the first cancer vaccine is attributed to Coley who, a century ago, observed that his cancer patients benefited from bacterial infection. This prompted him to treat the patients with bacterial extracts. It was not till the 1980s that development of cancer vaccines progressed further. While certain cancers have been successfully treated, progress has been relatively slow. The pace should increase substantially as scientists gain a deeper understanding of how the immune system fights tumors and as the success of the cancer vaccines now available is evaluated.

To learn how cancer vaccines work it is essential to start with insight into the working of the immune system. This review begins with a primer on tumor immunology: the identity and role of each cell involved in recognizing and fighting cancer. The next section explains how tumors actively try to evade the immune system, while the concluding section highlights the strategies used in designing cancer vaccines.


Immunology Primer

The immune system can be divided into two branches; both are involved in fighting cancer.

Innate: Barriers that human beings are born with, including special cells, to fight specific bacteria and other invaders. Natural killer cells are the most significant innate cells that fight cancer directly.

Adaptive: Response that the immune system generates to fight threats as they occur. Lymphocytes are cells involved in fighting these threats. There are two types of lymphocytes, B cells and T cells. Some lymphocytes are cells that trigger immunity, while other lymphocytes are memory cells that allow this adaptive response to occur repeatedly. Cytotoxic T cells are the adaptive cells that directly fight cancer. However, they cannot always recognize cancers and need antigen-presenting cells, dendritic cells, to help them do so.

Properties of Tumors

  • Since they resemble normal cells, tumors tend not to trigger the immune system.
  • Tumors also actively evade the immune system in different ways.
  • Immune cells are not very efficient in detecting tumor growth.

Cancer Vaccines

  • Vaccines boost the immune system.
  • Preventive vaccines may be used to prevent cancers induced by viruses, such as cervical cancer.
  • Most cancer vaccines will probably be therapeutic, used for patients who already have cancer.
  • Cancer vaccines modify the immune system response to evoke a strong and specific immune response.
  • Types of cancer vaccines include:
    - Tumor antigens used to familiarize the body against the cancer
    - Monoclonal antibodies to mimic tumor antigens
    - Stimulation of the cytotoxic T cells to fight the cancer

I. Tumor Immunology

A. Innate Immunity
The mammalian immune system consists of two broad arms: innate immunity and adaptive immunity. Innate immunity is constitutive, non-specific and swift.2 It consists of natural anatomical barriers, such as skin and mucous membranes, and physiological barriers like elevation of temperature and acid in the stomach to digest harmful bacteria. Innate immunity is effective against infectious agents that have common features recognized by phagocytic cells and their extracellular and intracellular components. Another feature of the innate immune system is complement; a group of inactive proteins in the blood. These are activated in the presence of pathogens and cause cell lysis. Tumor cells have complement regulatory proteins on their cell surface - proteins that inhibit the activation of complement - and thus escape complement-mediated lysis. Pattern-recognition receptors, present on the cell surface, and antimicrobial proteins present inside cells, are part of the innate immunity and both target pathogenic bacteria. Phagocytic cells, namely, natural killer cells, dendritic cells and macrophages are the components of the innate immune system most directly involved in tumor immunology. These cells also participate in the adaptive response and form a bridge between the two arms of the immune system. Innate immunity evolutionarily precedes adaptive immunity and is therefore present in all vertebrate and many invertebrate species.

The innate and adaptive mammalian immune system
Figure 1. Immune System

In addition to the phagocytic function of engulfing bacterial and damaged cells, the cells of the innate branch have specific roles:

(1) Natural Killer (NK) Cells
NK cells are part of innate immunity that possesses the ability to kill tumor cells without a previous encounter.3 These cells have killer activating receptors and cause lysis of target cells using specialized enzymes, perforin and granzymes. Killer inhibitory receptors are also present on the NK cell surface, which prevent lysis of cells with MHC molecules. Target cells for NK cells include virally-infected cells and tumor cells. NK cells do not require binding to MHC-antigen complex, so they can kill tumor cells that have low levels of MHC molecules.4 NK cells play a key role in tumor immunology. NK cells lack both CD4 and CD8 antigens and are identified as CD4-CD8- cells.

(2) Dendritic Cells (DCs)
Dendritic cells are present in tissue that is the first line of defense, i.e. skin, mucosal and respiratory membranes.5 In the blood, DCs are found in the immature state. Using their pattern-recognition receptors (e.g. toll-like receptors) they detect bacteria and viruses. When encountered, the pathogens are phagocytosed and their proteins are processed and inserted into the DC surface to be presented to T cells. Dendritic cells are professional antigen-presenting cells (APC). They activate helper T cells and Cytotoxic T cells and also activate B cells. They are very important in tumor immunology. DCs are CD34+ meaning the CD34 differentiation antigen is present on their cell surface.

(3) Macrophages
Macrophages are derived from monocytes and have several different names (e.g. Kupffer's cells, histiocytes, alveolar macrophages) depending on the tissue in which tissue they are found.6 Macrophages have granules filled with digestive enzymes. They are important in fighting bacteria and also ingest damaged cells by phagocytosis. In addition to being strongly phagocytic, they present antigens to T cells and thus have a role in adaptive immunity. Macrophages can destroy tumor cells and play a crucial role in the inflammatory response.

B. Adaptive Immunity
The adaptive (or acquired) responses of the immune system are very specific and slower than the innate response. The adaptive response follows the innate response and is dependent on specific recognition of antigen by antigen receptors present on the cell surface. There are two types of adaptive immunity: cell-mediated immunity and humoral immunity. T lymphocytes are responsible for cell-mediated immunity and B lymphocytes for humoral immunity. Immunological memory is a feature of adaptive immunity—after the initial immune response, B and T memory cells present in the blood are triggered to mount a stronger and more effective immune response when they encounter the same intruder again.

(1) B cells are produced and mature in the bone marrow. On the cell surface are antigen receptors or B cell receptors (BCR), proteins that recognize and bind to soluble antigens in the blood. The antigens are then taken up by the cells and processed. The fragments of the digested antigens are displayed on the surface of the cell bound to MHC class II molecules. This induces a T helper cell to bind and secrete lymphokines (a type of cytokine, see section ID). The lymphokines cause the B cell to mature and divide into a plasma cell. The mature B cell switches into an antibody-producing plasma cell. Antibodies bind with very high specificity to the antigen. For flash animation of a humoral response against a bacterium, see In tumor immunity, B cells play a role in destroying tumor cells by two different means. The first is complement-mediated lysis. In addition they facilitate antibody-dependent cell-mediated cytotoxicity in which antibodies recognize and bind to a tumor cell and then trigger cell lysis of target tumor cell by several different immune cells.

(2) T cells (or T lymphocytes): During hematopoiesis, bone marrow stem cells that are destined to become T cells migrate to the thymus to complete their development and maturation. The thymocytes or developing T cells are protected from the contact with antigens in the blood.7 CD antigen markers are added to the T cell surface as they differentiate (i.e. develop into the specific type of T cells). Flash animation: cellular response against virus

Types of T cells
In addition to Memory cells there are four types of T cells:

(a) Cytotoxic T cells (CTL) CTL are effector T cells.8 These cells play a central role in tumor immunology since they destroy tumor cells, which they recognize by virtue of the tumor cell surface antigens. Immature CTLs require activation by antigen-presenting cells. Mature CTLs recognize major histocompatibility complex or MHC- antigen complex on the surface of a target cell and destroy the cell. Tumor cells that do not display MHC-antigen complex are destroyed by NK cells. Together CTL and NK cells are the two cell types that destroy tumor cells. CTLs have the CD8 marker on their cell surface; they are CD8+.

(b) Helper T cells (Th cells) Helper cells are so called because they help other immune cells to perform their function7 primarily by secretion of lymphokines. Th cells, CD4+ cells in the thymus, are destined to be helper T cells. These, along with CTL, are effector T cells. Presentation of antigen and the appropriate cytokine lead to differentiation into type 1 (Th1) or type 2 (Th2) cells. Th1 cells participate in cell-mediated immunity in controlling infections and Th2 cells participate in humoral immunity since they cooperate with B cells.9

(c) Suppressor T cells or Regulatory T cells (Tsupp or Treg cells) Suppressor T cells prevent an uncontrolled immune response of effector cells, which could lead to autoimmunity.10 There are subpopulations of suppressor cells, CD4+ CD25+ and CD8+, with different modes of action,11 including secretion of TNF-beta, and interference of effector cell T cell receptor (TCR) binding to MHC-antigen complex.

(d) Natural Killer T (NKT) cells
Not to be confused with NK cells, NKT cells play a role in searching for tumor cells (called immune surveillance) and in preventing metastasis.12 These cells are cytotoxic cells that have characteristics of both NK cells and T cells. NKT cells express TCR (a specific subtype of TCR not found in other T cells) on their cell surface, and in addition express NK receptors, making them unique since they display receptors of both NK cells and T cells. NKT cells secrete IL-4 or gamma-interferon and are able to activate T helper cells, making NKT cells a link between innate and adaptive immunity.

C. Antigen Presentation and T cell Activation
When cells in the body present MHC class I coupled antigens, T cells recognize and destroy the cells. However, T cells do not recognize many tumor antigens directly on the cells. Instead, special antigen-presenting cells (APCs) present the antigen for recognition, which triggers activation of T cells, enabling the T cells to mount a response to destroy the tumor cells. Dendritic cells are proficient at processing and presenting antigens and are thus called professional antigen-presenting cells. DCs have high levels of Major Histocompatibility Complex (MHC), costimulatory molecules and cell adhesion molecules,13 all of which are essential for antigen presentation.

What is the function of MHC molecules? Named for immune rejection of incompatible transplanted tissue, they play a vital role in cell-mediated immunity. When there are problems within cells, such as an infection or break down of cellular material prior to cell death, the degraded proteins of the cell are displayed on the cell surface bound to the MHC molecule. These protein fragments or antigens are displayed and recognized by circulating immune cells in the blood, allowing the immune system to survey the status inside of cells since healthy cells would not have the degraded protein and the fragments displayed on the cell surface.

MHC class I molecules are displayed on almost all the cells of the body and are recognized by CTL cells (Fig. 2). MHC class II molecules are displayed on immune cells, namely dendritic cells and macrophages, and are recognized by helper T cells. The MHC molecules are responsible for self-tolerance and mounting an immune response against foreign and potentially harmful agents.

MHC Class I Antigen Presentation Pathway
Figure 2a. Antigen Presentation. Viral proteins are processed in an infected cell. The protein fragments bind to MHC I molecules (see inset Fig. 2b) and are displayed on the cell surface. These are recognized by cytotoxic T cells, which then destroy the infected cell.

Figure 2b. Major histocompatibility complex (MHC) class I is magnified. The MHC molecule (orange) has a groove to bind the processed antigen (red) for display on the cell surface. Beta 2-microglobulin (pink) stabilizes the molecule. The whole complex is embedded into the cell membrane.

Antigen presentation leads to activation of T cells that can recognize and destroy tumors. As shown in Figure 3, the MHC class II-antigen complex on the APC cell surface is recognized by the T cell antigen receptor (TCR). The binding of the two is very specific and is called an immunological synapse. Next, the costimulatory receptor, CD28, binds DC cell adhesion molecules. The formation of the TCR-MHC-antigen complex sets off a cascade of events within the T cell resulting in gene activation, and the ultimate result is cell proliferation or differentiation or anergy or apoptosis.14 Cytokines, specifically IL-2 and also other interleukins, play an important role in cell activation. Activation of CTL and Th cells is similar.15 Differentiation of the T cell into an effector cell (a T helper cell or CTL) is required to produce a cell-mediated immune response. It is noteworthy that T cell activation is a complex event; failure to obtain a sustainable activation signal results in failure to recognize the tumor antigen, leading to failure to attack the tumor.

The 2 signals of T Cell activation: antigen and costimulation
Figure 3. T cell activation by an antigen-presenting dendritic cell. Antigen presenting cells use class II MHC molecules in contrast to other cells.

D. Role of Cytokines
Cytokines are essential in any type of immune response. These are proteins secreted by cells of the immune system to signal other cells of the immune system. Typically they are produced locally and affect cells in the near vicinity. The following table lists some of the cytokines involved in generating an immune response.

Interleukin-2 activated Th1 cells, NK cells Stimulate proliferation of B cells and activated T cells, NK functions. Activate anti-tumor CTL response
Interleukin-4 Th2 and mast cells B cell proliferation, eosinophil and mast cell growth and function, IgE and class II MHC expression on B cells, inhibition of monokine production
Interleukin-8 Macrophages and other cells Stimulation of cell migration of neutrophils and T cells to inflammation site
Interferon-alpha and -beta macrophages, neutrophils and some somatic cells Antiviral effects, induction of class I MHC on all somatic cells, activation of NK cells and macrophages. Stimulation of B cell class-switching
Interferon-gamma activated Th1 and NK cells Stimulation of MHC molecule expression on APCs and somatic cells. activation of macrophages, neutrophils, NK cells, anti-tumor CTL and antiviral effects
Tumor necrosis factor-alpha Activated macrophages, damaged tissue Induces expression of other growth factors and induces cell proliferation
Tumor necrosis factor-beta
CTL cells

Induces cell death in many cell types
Transforming growth factor-beta Th1 cells and NK cells Anti-inflammatory, promotes wound healing, inhibits macrophages and lymphocyte proliferation
Colony-stimulating factors: Granulocyte Macrophage-Colony Stimulating Factor Induce cell proliferation in bone marrow stem cellsActivation of DCs
Table 1. From Basic Science of Tumor Antigen and Immune activation, Handbook of Cancer Vaccines and

II. Immune properties of Tumors
Tumors do not induce a strong immune response for two major reasons. First, early in embryonic development, all the lymphocytes that recognize normal cells' antigens—"self antigens"—are destroyed, to prevent an immune response against one's own tissue. This is called self tolerance. Since tumor cells are not dramatically different from normal cells, tumor cells are also immunologically tolerated (i.e. ignored by the immune system). Second, tumors actively evade the immune system (described below in IIB).

To examine the immune properties more closely this section is subdivided into three: tumor antigens, the mechanism of immune evasion by tumors and the detection of cancerous growth (called immune surveillance).

A. Tumor antigens
In general, tumors are poorly immunogenic and do not trigger an immune response. There are two types of tumor antigens:16 Some tumors have unique antigens (not found on normal cells), called tumor-specific antigens (TSA) . TSA are present in tumors induced by infectious agents (e.g. EBNA-1 antigen from Epstein Barr virus-induced Burkitt's lymphoma) and mutated genes found only in tumor cells (e.g. mutated caspase-8 enzyme found in head and neck cancer, which is different from the normal caspase-8). However, many tumors have antigens, called tumor-associated antigens (TAA) , similar to regular cells but either modified or produced in greater quantities.17 Some TAA are limited to a specific tumor type (e.g. melanocyte differentiation antigens, MAGE antigens, are limited to melanomas and some normal tissue), while other TAA are found in several different tumors (e.g. cancer testis antigen are found in normal testis and a variety of cancers such as prostate carcinoma, breast carcinoma, and lung carcinoma).

B. Immune Evasion of tumors
Tumors actively evade the immune system in several ways, some of which are:

(1) Dysfunction of the immune system is induced by the cancer cells. The immune system of cancer patients and animal models is found to have various immune defects18 such as:

a) Defects in antigen presentation: Tumor cells are genetically unstable and loss of genes and even parts of chromosomes is often seen. Loss of MHC class I genes means that the tumor antigens will not be displayed on tumor cell surface (see Fig 2) since MHC molecules are absent. This leads to escape of the tumor from CTL response. This is one mechanism of immune evasion. Similarly, defects in antigen processing apparatus (e.g. TAP protein defect) can lead to decreased levels of antigen presented to CTL, resulting in immune evasion.

b) T cell defects leading to T cell anergy: Tumor cells produce an enzyme, arginase, that depletes an important amino acid, arginine. Arginine is an essential component of the T cell receptor, therefore depletion of arginine leads to a loss of T cell recognition of the antigen leading to a loss of T cell function. Interestingly bacteria that cause chronic infections, like Helicobacter pylori (stomach ulcers) and Mycobacteria (leprosy and tuberculosis), also employ this strategy of producing arginase.

c) Dendritic cell defects: Since many tumors are not immunogenic, DCs are essential for activating T cells (Fig. 3). Targeting DCs prevents an anti-tumor response. One way tumors inhibit DCs is by secreting growth factors, which inhibit formation of dendritic cells in the bone marrow. Besides inhibiting DC formation, some tumors increase IL-10 levels. IL-10 reduces expression of CD antigen molecules on DCs, CD80 and CD86, both of which are required for a T cell to be activated. In the absence of CD80 and CD86, T cells become tolerant to the tumor and do not mount an immune response against it. A third way of affecting DC function is tumor secretion of nitric oxide and hydrogen peroxide that causes DCs to undergo cell death.

(2) Immune suppression
Cyclooxygenase-2 (COX-2), an enzyme produced by lung cancer cells, causes several effects on the immune system and allows metastasis of cancer. One of the effects is immunosuppression due to decrease of the cytokine IL-10, and increase of IL-12. Mouse models lacking Cox-2 showed reduced tumor growth when transplanted with tumors. Cox-2 inhibitors are used to prevent metastasis18

(3) Cell deletion
Developing T cells in the thymus recognize a specific antigen. When the antigen is recognized in the body the cells undergo active proliferation to mount a response. Some tumors actively kill those developing T cells specific for the tumor antigen, which could potentially mount an anti-tumor response. For example, T cells specific for a plasmocytoid tumor antigen were deleted when introduced into a mouse19 leading to immune evasion.

(4) Immune ignorance
T cells have been seen to mount an immune response against a reactive tumor antigen in vitro (outside the body) but not in vivo (inside the body). Immune ignorance means that the T cells are blind to the tumor antigen. This is different from anergy (or unresponsiveness) in T cells. Ignorance is also different from immune suppression, which is inhibition of T cells, non-specific to an antigen.18

C. Immune surveillance
T cells (NKT cells and CTLs) are believed to carry out an immune surveillance function. They seek out newly transformed cells by recognizing tumor antigens. The tumor antigens do not need to be on the cell surface, since antigen processing and presentation coupled to MHC I molecules ensures that even intracellular (within the cells) antigens are recognized (see Fig. 2a, b) and the cells are destroyed. Tumor-specific CTL have been observed in many types of cancers.18 NK cells are also found in tumors and appear to recognize common characteristics of tumor cells. Humoral immune response is also involved in detecting and destroying cancers, e.g. carcinogen-induced and virus-induced leukemia is recognized by antibodies. These antibodies trigger complement-mediated cytotoxicity. Despite the fact that immune cells recognize tumors and are found within them, tumors do occur. Growth of a tumor can therefore be viewed as failure of immune surveillance.

III. Cancer Vaccines
While there has been significant progress in developing cancer vaccines in the last decade, the concept of stimulating the immune system to combat cancer is a century old. William B. Coley first used streptococcal cultures to treat patients with advanced sarcoma (a type of malignant cancer of the connective tissue) between 1900 and 1936.20 The result was a clinical immune response against the tumor; this was the first cancer vaccine to be used. Today we understand that the components of the bacterial extract stimulated the immune response in a general way, causing Coley's toxin to succeed. Since Coley, bacille Calmette-Guerin (BCG), which works in a similar manner, has been successfully used against bladder cancer. In the 1980s, Rosenberg used interleukin-2 (IL-2) to treat advanced cancers. Tumor regression was reported in 15 to 20% of the patients. IL-2 does not directly affect solid tumor growth; the anti-tumor effect seen is believed to be due to its effect on T cells. All of these vaccines provide a general boost to the immune system, enabling it to respond better against the tumor.

At present animal models are available for many cancers and mouse models have been used to study cancer vaccines. Clinical trials have been conducted on some cancer vaccines and more are in the process. For updates on clinical trials see

A. Preventive vs. Therapeutic Vaccines
Vaccines to prevent infectious diseases are prophylactic (or preventive). In contrast most cancer vaccines are expected to be therapeutic (destroying a tumor that has already developed). The exceptions are vaccines to prevent cancers induced by infectious agents, such as Human papilloma virus (HPV) that causes cervical cancers and Helicobacter pylori bacteria believed to cause gastric cancers. HPV16 and HPV18 together account for 70% of cervical cancers, and a vaccine recently developed against these two strains shows great promise.21 A number of other cancers are virally-induced, and the feasibility of using preventive vaccines is being studied. Theoretically, preventive cancers should also be effective in cancers with a mutated gene that produces different antigens, e.g. a mutated hormone receptor in breast carcinoma.

Therapeutic vaccines
Many cancers are not caused by infectious agents. For these, designing cancer vaccines offers more challenges than vaccines against infectious agents.22 One challenge is that most tumor antigens are self antigens. So it is hard to induce long-term immunological memory against tumor antigens without producing autoimmunity. Overcoming immunosuppression is another factor. Causes of immunosuppression are the direct effect of the cancer (see section IIC) or previous cancer treatments or aging. For these reasons most cancer vaccines are for treating cancer or therapeutic vaccines.

B. Designing Cancer Vaccines
An ideal cancer vaccine induces a strong anti-tumor response in the host while sparing normal tissue. Therefore two essential factors of the cancer vaccines are specificity and strength of the immune response. Specificity refers to the recognition of a tumor antigen by the T-cell receptor or antibody. Each of the strategies described in the following section, IIIC, uses a specificity to target the tumor cell for destruction. Since tumors are poorly immunogenic, the second essential element of a cancer vaccine is to increase the strength of the immune response. This is often achieved by non-specific (or general) immunostimulators, as described next.

Use of General Immune Stimulation in Cancer Vaccines
Before examining specific cancer vaccine strategies it is appropriate to address the advantages of two important types of molecules that boost immunity non-specifically, i.e. the effect is not specific for a particular tumor.

(1) Adjuvants are potent immunostimulatory agents. The success of Coley's toxin is now attributed to two adjuvants in the bacterial extract, DNA and lipid A. Examples of adjuvants used in tumor vaccines are keyhole limpet hemocyanin (a protein from limpet mollusks), bacterial lipopolysaccharides (lipid-carbohydrate molecules), and BCG. A chemopreventive (tumor-preventing) agent, curcumin, from the turmeric plant shows promise and may be used in the future. Some of the specific cancer vaccines use adjuvants to increase the strength of the immune response.23

(2) Cytokines are soluble factors used for signaling by immune cells (see section ID) and individual cytokines are often used in conjunction with other treatments.23 A complication of cytokines is toxicity. The three most commonly used cytokines at present are: (i) IL-2, the cytokine most often used to treat cancer. IL-2 acts by stimulating cytotoxic T cells to respond to tumors. In clinical trials, IL-2 caused long term regression for melanoma, but 20% of the patients developed vitiligo, an autoimmune skin disease21. While effective, IL-2 treatment reveals another complication in using immunotherapy, namely that onset of autoimmunity can be a side effect. (ii) Interferon-alpha has been used to treat hairy cell leukemia and Kaposi's sarcoma.24 Cytokines are either used directly or the gene encoding the cytokine is introduced into the patient's cells to enable the patient's own cells to manufacture the cytokine. (iii) GM-CSF gene, for instance, is used to direct the synthesis and secretion of Granulocyte Macrophage Colony Stimulating Factor to stimulate dendritic cells. GM-CSF has been used in conjunction with several cancer vaccines.

C. Types of Cancer vaccines
(1) Tumor antigen-based cancer vaccines
The rationale behind tumor antigen-based vaccines is that the patient's immune system will recognize the same antigen on the tumor cells, mount an immune response against the tumor, and eventually destroy it. Recent advances in protein chemistry and genomics have led to the identification of tumor antigens with potential cancer immunotherapy use. Two approaches introduce the tumor antigen into the patient or animal model. A third approach is a variation in which tumor-associated carbohydrate antigens are used.

a) Tumor antigen DNA vaccines
The first approach is to introduce DNA that encodes tumor antigen into an animal or person. The DNA is taken up by antigen-presenting cells (APCs) and the tumor antigen is produced in the cell. It is then presented on the cell surface as MHC-antigen complex and activates cytotoxic T cells (see section IC). The activated T cells can now recognize and destroy the tumor with the same antigen on the cell surface of the tumor cells. Research is ongoing to find the ideal vector for the tumor antigen encoding gene. Various viral vectors have been tried, with different properties with the aim of prolonging the gene expression of the vaccine. Surprising challenges have been encountered,25 e.g. viral vector-based vaccines were successful in producing tumor regression in animal models but not in clinical trials. The failure in humans is due to viral vectors that were used in immunization programs against infectious agents decades ago, such as smallpox, so that patients' immune systems now react against the vectors.

b) Tumor Antigen synthetic Peptide Vaccines
A portion of a tumor antigen, a synthetic peptide (an artificially-produced short protein sequence), is directly introduced into a patient. The ideal peptide vaccine should not be abundant in any normal tissue. Preliminary results for a 9 amino acid peptide vaccine show promise in clinical trials of myeloid leukemia patients.26 Other peptide vaccines have shown mixed success. In melanoma, despite obtaining good preliminary results with gp100-derived peptide, the tumor response was not strong enough to combat the disease and many patients eventually succumbed.25 Adjuvants are added to peptide vaccines to stimulate the immune system.

c) Antibody-inducing vaccines against carbohydrate antigens
Individual tumor cells and early metastases are eliminated by inducing an antibody response against cell surface carbohydrate antigens.12 The Thompsen-Freidenreich antigen, a small molecule found in carcinomas but not in normal cells, is expressed in early stages of cancer transformation. The use of this tumor antigen is an attractive option since it is shared by many types of cancers.

(2) Monoclonal antibodies. While the first strategy utilizes antigens, the second strategy uses antibodies for cancer therapy, specifically monoclonal antibodies. Monoclonal antibodies are derived from a single antibody-producing cell and recognize a single antigen.27 They are used in immunotherapy for different purposes. Two examples are:

a) Anti-idiotype vaccines
Antibodies are proteins and can be antigens for other antibodies. An anti-idiotypic antibody is an antibody against an individual structural determinant of variable region of other antibodies. Vaccines using these types of monoclonal antibodies are called anti-idiotype vaccines. When introduced into the patient these antibodies bind to the tumor by recognizing the tumor antigen. The anti-idiotypic antibodies in turn are recognized by other antibodies, thus forming a network and magnifying the immune response.

Idiotype network
Figure 4. Idiotype Network. (A) Monoclonal antibody (Ab1) binds the tumor antigen (B) A different monoclonal Ab2 binds to Ab1. Ab3, a different monoclonal antibody binds Ab2, setting up a network cascade of antibodies.

This triggers antibody-dependent cell cytotoxicity (See IB) and T cell response, stimulating both humoral and cell-mediated immunity. No adjuvants are needed since the cascade of anti-antibodies ensures a strong and long-lasting immune response.28 Clinical trials with colorectal cancer patients using carcinoembryonic antigen (CEA) show promising results.

b) A different strategy uses monoclonal antibodies against a costimulator for T cell activation. As shown in Fig. 3, binding of CD28 is essential to activate a T cell. In a mouse model, binding of antibody to CD137 provides strong costimulation (see Figure 3) and breaks immune ignorance in mouse models by allowing the T cell activation.18

(3) Cell-based immunotherapy:
The above examples introduce molecules into the patient to generate a strong anti-tumor response. In contrast, the next types of cancer vaccines involve administrating entire cells into the patient.

a) Tumor cell-based vaccine:
Whole cells (containing tumor antigens), taken either from the patient (autologous) or from a different patient (allogeneic), are introduced to stimulate the immune system to recognize the existing tumor and mount a response. Since the tumor cells are destroyed they are not harmful, rather they stimulate the host immune system to recognize the tumor cells. The advantage of this method is that the tumor antigens do not need to be identified. Whole tumor cell vaccines mixed with BCG (as adjuvant) have been used against colorectal cancer, melanoma and renal cell carcinoma.29

b) Dendritic cell-based vaccines:

Dendritic cells that attack cancer
Figure 5. The above figure shows a tumor antigen linked to a cytokine bind to a dendritic cell. The dendritic cell processes the antigen and presents it to cytotoxic T cells. The activated cytotoxic T cells now recognize the cancer cell and destroy it.

Lack of efficient tumor antigen presentation in DCs in cancer patients has led to the use of DC-based vaccines. As Fig. 5 shows the appropriate tumor antigen is bound to the DC cell surface. The tumor antigens are taken up by dendritic cells; they are processed and presented to the T cells along with the appropriate costimulatory signal. Once activated by the DCs the cytotoxic T cells recognize and destroy the tumor cells expressing the tumor antigen.30 What is the source of the DCs in the DC-based vaccine? As shown in the figure below, DCs are collected from the blood of the patient (a process called leukapheresis) and "loaded" with tumor antigens from the patient's own tumor cells. These DCs are then reintroduced into the patient and stimulate the immune system. DC vaccines have been used in patients with metastatic melanoma, renal carcinoma and prostate cancer.13

Generation of dendritic cells in a patient
Figure 6

Recently, dendritic cell-tumor cell hybrid (DCs and tumor cells fused together) generation has been reported,31 since these cells' ability to stimulate T cell responses is better than a mixture of tumor cells and dendritic cells, as shown in Fig. 6.

c) T cell-based vaccines:
Recently it has become feasible to manipulate the cells most critical in cancer immunity, T cells. Development of techniques of culturing T cells, and development of vectors to introduce appropriate genes, into these cells has led to clinical trials using T cells. A recent review details the progress and future directions for T cell-based cancer immunotherapy.1 (i) In the first strategy, T cells from a different person are infused into the patient. The rationale is that MHC antigen differences, together with the tumor antigens, set up a graft-versus-tumor reaction resulting in T cells from the donor destroying the host's tumor. This is an example of allogeneic lymphocyte therapy10. (ii) A different strategy involves manipulation of CTL cells (see Fig. 7). The rationale here is to bypass the involvement of MHC class I on tumor cells and antigen-presenting cells by directly modifying the cytotoxic T cells to recognize a tumor antigen and become activated. For this purpose a chimeric molecule designed, consisting of an antibody and a T-cell receptor. The antibody recognizes the appropriate tumor and the TCR activates the T cell to combat the tumor. These have been successful in treating ovarian tumors in mice.13

A gene encoding a chimeric receptor and a T cell expressing a cytokine
Figure 7. This schematic representation shows two types of genes introduced into cytotoxic T cells using a virus: The top green T cell expresses a gene encoding a chimeric receptor. This gene codes for a fusion protein, part of which is an anti-tumor antibody (orange), while the other part is a T cell receptor (black).13 Upon encountering the tumor, the antibody specifically binds to the tumor antigen and the TCR signals proliferation and activation.
The bottom T cell expresses a cytokine (yellow) to improve survival and anti-tumor efficacy.

d) NK cell-based therapy
Similar in principle to the first T cell-based strategy (Section IIIC c(i)), clinical trials in renal carcinoma patients show that NK cells from a donor produce a graft-versus-tumor effect against the cancer. The donors and hosts were mismatched for killer cell inhibitory receptors18 to produce a strong immune response. In the future it should be possible to develop more NK-cell based cancer therapies.

While we have gained insights into the workings of the immune system, some mysteries remain, notably how tumors escape both T cells and NK cells? Are the signals for NK cell cytotoxicity also destroyed in tumor cells that have lost MHC genes since tumor cells often lose genes randomly? Some challenges in for the future are to develop vaccines that sustain T cell activation and proliferation and prevent T cell anergy to produce a long-term anti-tumor effect. Another challenge is that, if tolerance against TAA self antigens breaks down, autoimmunity can result. This is a complication with serious repercussions and no easy solutions (see III B).

While the development of cancer vaccines has been challenging and more complex than anticipated, its future appears to be promising. It should soon be possible to develop vaccines to prevent the occurrence of cancers caused by infectious agents (HPV-induced cervical cancer and Helicobacter-induced gastric cancer).

In the past cancer was regarded as invincible. With numerous treatment options available, this is no longer the case for many cancers. As the work in this field comes to fruition, treatments customized for the specific type of cancer will some day become routine.

© Copyright 2006, All Rights Reserved, CSA