Immune Evasion Strategies of Cancer

INTRODUCTION

EVASION FROM ANTIGENIC RECOGNITION

tissue localization

failure of lymphocyte homing

shedding of antigens and antigenic modulation

EVASION FROM IMMUNOGENIC RESPONSE

release of immunosuppressive cytokines

secretion of prostaglandins

downregulation of MHC I expression

lack of costimulation

interference with apoptotic pathways

tumor growth kinetics

CANCER LINKS

BACK TO CANCER VACCINES

REFERENCES

BIO 160 MAIN PAGE

Immune responses in cancer patients are often far from ideal. Because cancer cells are altered-self cells, one would expect cancerous cells to elicit a cell-mediated response. That is, the immune system should target cancerous cells and destroy them. There are three processes that must occur for tumor elimination. The immune system must "see" the cancer, activate lymphocytes, and the cancer cells must be susceptible to killing. In order for this to take place, lymphocytes should be able to infiltrate to the tumor site. CD4+ T-helper1 (Th1) lymphocytes should then recognize tumor-specific antigens in association with MHC II molecules on the surface of professional antigen presenting cells and receive signals from costimulatory molecules such as B7. As a result, Th1 lymphocytes should be activated and release appropriate cytokines including interleukin-2, interferon-gamma and tumor necrosis factor-alpha. These cytokines, in addition to stimulation by tumor-specific antigens presented on cancer cell surface MHC I molecules, should activate cytotoxic "killer" T lymphocytes (CTLs) to lysis cancerous cells. B lymphocytes should also be activated to secrete neutralizing antibodies that aid in cancer cell phagocytosis by antigen presenting cells, although their role in tumor immunity is less important. CTL-mediated lysis of a cancerous cell, the ultimate action of an effective immune response against cancer, is shown below. If any of the processes necessary for the induction of a cell-mediated response fail, tumor elimination may not be effective.

The ideal immune response described above often does not occur in cancer patients because cancer cells evolve mechanisms to evade the body's defenses. The immune surveillance theory hypothesizes that cancerous cells arise regularly but the body eliminates them before they become harmful to surrounding tissue; only those that evade surveillance develop into tumors. This theory is supported by the increased incidence of cancer in immunosuppressed people such as AIDS patients. If the immune surveillance theory is correct, there is a strong selective pressure favoring cancer cells that can avoid notice or somehow prevent themselves from being killed by the immune system.

Evasion strategies can either be preexisting or arise through outgrowth of escape mutants. Because cancer cells arise from normal cells, many of their evasion strategies are shared by healthy tissue. Ordinarily, the immune system does not want to kill the body's own, healthy cells. These self-cells are protected from immune responses by a variety of mechanisms including self-tolerance, sequestration of tissue from surveillance, antigen shedding, lymphocyte killing, secretion of immunosuppressive cytokines, lack of MHC II expression, lack of costimulatory molecules and local secretion of prostaglandins neuropeptides. Many of these evasion strategies are maintained in or exploited by cancer cells. However, because cancer cells contain altered proteins, the immune system may be able to recognize tumor-specific peptides as foreign antigens. As a result, many cancer cells have additional mechanisms to escape host immunity which may include up regulation of the evasion mechanisms shared by healthy cells, downregulation of MHC I expression or peptide presentation, antigenic modulation and failure of lymphocyte homing. These strategies arise through the selection and outgrowth of mutant cells that escape the immune system. Particular evasion strategies are discussed in more detail below.

Cancer is not a single disease; the word cancer refers to approximately 150 diseases. These diseases share two characteristics in common, uncontrolled cell growth and the ability to damage normal tissue. However, depending upon the particular mutations accumulated and the location of the tumor, cancers can be, in many respects, quite diverse. The diversity of cancer applies to their immune evasion strategies as well. The mechanisms that cancers use to escape host immune responses may vary among different types of cancers and even different cancers of a particular type. In general, tumors that are better at escaping host immunity may be more malignant than those that inefficiently evade immune responses.

Knowledge of the particular evasion strategies that tumors employ is necessary to design vaccines that specifically target the ability of tumors to escape immunity and may also be useful to predict responsiveness to available radiation or chemotherapy. For example, a tumor secreting high levels of immunosuppressive cytokines may be less responsive to treatment with Th1 stimulating cytokines such as interleukin-2 and IFN-gamma than a tumor that does not secrete immunosuppressive cytokines. For more on cancer vaccines see the cancer vaccines website.

There are several sites in the body, such as the central nervous system, which are inaccessible to the immune system. Tumors in such areas of the body are invisible to immune surveillance and thus cannot be targeted by immune reactions. Residence of a tumor in immune privileged sites allows them to be essentially non-antigenic because the immune system is not even aware of their presence.

The immunogenicity of a tumor antigen also seems to be affected by the location of the tumor antigen. It has been shown that fibroblasts, which lack costimulatory molecules and cytokines, can activate T cells only after a few of them have drained into the vicinity of the lymphoid organs. In other sites in the body, small numbers of fibroblasts are non-immunogenic and can therefore go unnoticed. This indicates that, depending on the carcinogen, a tumor may or may not go unnoticed in a particular area of the body. Since many tumor cells lack costimulatory properties, they can only be detected if they are in the appropriate environment within the body. Cancer cells that manage to avoid the lymphoid organs may be able to sneak through and develop into large tumors. After having reached this stage, it is extremely difficult for the immune system to effectively combat the tumor. This phenomenon is similar to the idea of sequestration in that the tumor may be positioned in a place where the immune system will not mount an immune response against it due to the fact that in that particular location, the numbers of tumor cells are not great enough to be immunogenic.

Tumor evasion of immune surveillance also relies on modulation of expression of various cellular adhesion molecules (CAMs). Lymphocytes normally migrate from the blood when such as L-selectin and alpha-4beta-7 integrin adhesion receptors on these cells bind to ligands expressed on the venular endothelium. In tumors that were not infiltrated by lymphocytes, the L-selectin and alpha-4beta-7 endothelial ligands were not expressed. It is the repressed expression of these ligands that correlates with the failure of lymphocytes to home to the tumor site and infiltrate it. When there is chronic inflammation at a site, the high endothelial venules (HEV) usually become induced to express a variety of cellular adhesion molecules which allow circulating lymphocytes to roll along the vessel wall, firmly adhere to the vessel and then transmigrate from the blood vessel to the inflamed tissue. L-selectin ligands and mucosal addressin cell adhesion molecule-1 (MAdCAM-1) on the HEVs are two of the major adhesion molecules involved in this homing process.

Experiments have shown that the HEVs in tumors that did not get infiltrated did not express significant levels of peripheral lymph node addressins (a subset of L-selectin ligands) and MAdCAM-1. This leads one to believe that tumors posses some mechanism by which they interfere with addressin expression. One possibility is that the tumor secretes paracrine inhibitors that act on the tumor vessel endothelium and prevent the induction of tumor HEVs. Another possibility has been shown in a recent study where VCAM expression was suppressed by soluble factors secreted by a solid tumor within the lungs.

There are also other CAMs, whose expressions, are affected by tumor mechanisms. Expression of the adhesion molecule ICAM-1, is often abnormal in a variety of tumors. Colonic cancer cells fail to express the LFA-3, another of the major cellular adhesion molecules.

The cellular mechanisms in immunologic tolerance involving antigens is still unclear it is thought that antigen shedding plays a role. Under normal conditions thymic negative selection is important and dominant mechanism for both CD4+ and CD8+ cells for those antigens and the peptides which gets expressed in the thymus. For tumor antigens to remain hidden it must be presented in such a way that after being processed by antigen presenting cells it does not get recognized as foreign. One way to avoid the problem all together is for the tumor cell to shed its antigen. Large scale shedding of tumor associated antigens into the serum and lymph creates a substanial degree of tolerance to the antigen in both the T and B cell population. Thus since these T and B cell populations are tolerant of it, an immune response is not triggered.

Cancer cells are masters of deceit and disguise. They can readily alter themselves to evade immunologic recognition and attack. Tumor cells alter their characteristics to evade attack by the immune system. They are capable of generating variants lacking features that mark them for destruction by T cells, killer cells and antibodies. This process is called antigenic modulation or immunoselection.

Some tumor specific antigens disappear from the surface of tumor cells when serum antibody is present. These antigens then reappear after the antibody has dissipated from the region of the tumor cell. When there are high concentrations of a antibodies against cancerous T cells, the antibodies will bind to the antigens of the cancerous T cell and induce endocytosis and/or shedding of the antigen-antibody complex. In order to avoid this fate, cancerous T cells do not display the particular antigen or antigens for which the antibodies are specific.

Immune evasion strategies that prevent immunogenicity (induction of an effective immune response) are described below.

Cytokines are low-molecular weight proteins that use their ability to act as intercellular communicators to regulate the immune response. A variety of cell types, principally T-helper lymphocytes and macrophages, can secrete cytokines in response to stimuli. Cytokines can act upon either the cells secreting them (autocrine) or on neighboring cells (paracrine) and elicit biological activities in the targeted cells. The functions that cytokines induce can both turn on and turn off particular immune responses. For example, interleukin-2 activates a cell-mediated immune response, while interleukin-10 suppresses cell-mediated responses. The immune modulating activities of cytokines are a result of their influence on gene expression, protein synthesis, membrane expression and antigen shedding from the cell surface.

Many types of cancers have taken advantage of the regulatory role of cytokines to down-regulate appropriate immune responses targeted at destroying cancer cells. They do this by secreting immunosuppressive cytokines that induce generalized and specific inhibition of immune responses. Cancer patients often fail to mount sufficient or appropriate immune responses to their cancers. This is due, in large part, to these immunosuppressive cytokines. Many of the phenotypic evasion strategies of cancerous cells are the direct result of immunosuppressive cytokines. Immunosuppressive cytokines secreted by cancer cells include transforming growth factor-beta (TGF-beta), interleukin-10 (IL-10) and vascular endothelial growth factor (VEGF).

Secretion of TGF-beta has been found in a variety of cancer types including malignant gliomas, breast cancers, prostate cancers and leukemias. TGF-beta is one of the most potent immunosuppressive cytokines yet characterized. It is capable of affecting the proliferation, activation and differentiation of cells participating in both the innate and acquired immune response. The proliferation of thymocytes, T cells, B cells, natural killer (NK) cells, monocytes and macrophages is inhibited by TGF-beta. In addition to suppressing proliferation, TGF-beta has been shown to induce apoptosis (cell death) in B and T cells.

Of particular importance is TGF-beta's affect on cytotoxic T lymphocytes (CTLs), which are important for anti-tumor immunity because of their cytotoxic effects. TGF-beta down-regulates many of the processes necessary for CTL activation. This is done by shifting the Th1-Th2 balance towards Th2, inhibiting Th1 cytokines including IL-12, downregulating IL-2 receptors on T cells, inhibiting antigen presentation on MHC class II molecules and downregulating adhesion and costimulatory molecules. Inhibiting antigen presentation prevents T cells from recognizing the cancer cells as foreign and downregulating adhesion molecules prevents T cells from even getting to the tumor site. If the T cells do manage to get to the tumor site and recognize the tumor, a lack of costimulation can cause the Th cells to become anergic instead of activated. Shifting the Th balance towards Th2 changes cytokines produced to ones such as IL-4 and IL-10 that mediated a humoral response appropriate for intercellular invaders, not cancer cells. Inhibiting Th1 cytokine production further prevents the appropriate, cell-mediated immune response from taking place. Finally, if Th1 cytokines such as IL-2 are secreted, the downregulation of IL-2 receptors makes them ineffectual. The immunosuppressive effects of TGF-beta on various immune cells in a malignant glioma are shown below.

While TGF-beta has potent inhibitory effects on tumor immunity, in contradiction to its immunosuppressive role, TGF-beta also can act in an autocrine fashion to inhibit tumor cell growth in vitro. However, because TGF-beta expression correlates strongly with tumorigenesis in vivo, the immunosuppression that TGF-beta induces must outweigh any anti-proliferative effects.

Many tumor types also secrete IL-10. IL-10 shares many of the same immunosuppressive effects on T lymphocytes as TGF-beta. Through its inhibition of cytokine production by macrophages, IL-10 indirectly reduces cytokine synthesis by Th1 cells. This down-regulates the functions of Thl cells including the activation of CTLs. Secretion of IL-10 in the vicinity of a tumor can render the tumor totally insensitive to CTL-mediated lysis. However, the effect of IL-10 on tumors is not entirely clear. In mouse models, tumor-specific immune responses were elicited in tumors transfected with an IL-10 expression vector. This directly contradicts the expected immunosuppressive role of IL-10. Therefore IL-10 may possess other roles in the development of anti-tumor immune responses.

The cytokine VEGF is produced by most tumors. It inhibits the differentiation of CD34+ cells into dendritic cells--professional antigen presenters. Since antigen-presenting cells are necessary for the activation of T-helper (Th) lymphocytes and these cells are needed to activate CTLs, a lack of dendritic cells could inhibit the generation of tumor-specific CTLs. The cytokines IL-6 and macrophage colony-stimulating factor (M-CSF) have been shown to have similar effects on dendritic cells as VEGF (Menetrier-Caux et al 1998). The expression of IL-6 has been found in tumors such as renal cell carcinomas.

The effects of TGF-beta, IL-10 and VEGF on tumor infiltrating lymphocytes are summarized in the table below:

Cytokines secreted by cancer cells play other roles in immune evasion in addition to suppressing the development of an immune response. The tumor-derived cytokines Fas-ligand and tumor necrosis factor (TNF) are involved in the killing of infiltrating CTLs. See Interference with apoptotic pathways for more information about these processes.

Immunosuppressive cytokines may prove to be the biggest obstacle to cytokine therapy of cancers. The success of cytokine therapy, the treatment of tumors with Th1 stimulating cytokines such as IL-2 and IFN-gamma, has been disappointing. Approximately just 5% of IL-2 or IFN-gamma treated patients achieve long term remission. This may in part be due to antagonistic effects of immunosuppressive cytokines secreted by cancer cells.

The role of prostaglandins immunoregulation is complex. Prostaglandins affect cell differentiation as well as target cell interaction. In humans the cells responsible for prostaglandin production is the monocyte and macrophages. Monocytes and related cells play a key role in regulationg the interleukin cascade leading to T cell proliferation and finally to the immune response. They can either amplify the response by producing interleukin 1 or shut it down mainly by releasing prostaglandin E2. Prostaglandins may play an important role in immune response/tumor cell interaction based on the following reasons. Firstly, a variety of prostaglandins are produced by cells that are themeselves active in the expression and regulation of immune response activity. Secondly the production of prostaglandins have been linked with the production of effector lymphocytes and tumor targets. Thirdly, prostaglandins produced during these interactions have been shown to influence the expression of lymphocytes and macrophage cytoxicity against tumor targets.

Prostoglandin synthesis is regulated by the COX gene expression. Two separate gene products are produced COX-1 and COX-2. COX-1 and COX-2 are expressed in high levels by tumors. More specifically COX-2 is expressed in high levels in intestinal tumors in both humans and rodents.

Prostaglandins are produced by tumor cells at increased levels when they interact with effector lymphocytes. In vitro, prostaglandins have been shown to inhibit lymphocyte mitogenesis, cytolysis and antibody production. Clinical studies have demonstrated that macrophages from patients with Hodgkin’s disease produces excess amounts of prostaglandins. Also administration of prostaglanin inhibitors to patients with breast cancer cased macrophages in vitro to acquire enhanced cytotoxicity for tumor target cells. Thus prostoglandin pruduction by tumor cells has been suggested as a mechanism by which tumor cells can escape the host’s immune surveillance.

Neuropeptides may also be involved in immune evasion. The neuropeptide adrenomedulin (AM) plays critical roles in a large number of physiological processes and diseases including cancer and heart disease. AM is expressed in many tumor cell lines such as lung, colon, ovary, prostate, brain and skin cancers. Experiments have shown that when neutralizing monoclonal antibodies to AM are administered, it affects the growth of certain human tumor cells. It seems to have a dichotomous role in cell growth, stimulating some cell types while suppressing others. Structurally, AM has arginine-rich residues, which suggests that it has defensive properties.

In a study with breast carcinoma cells, their secretion of TGF-B in vivo suppresses host immune surveillance and curtails the MHC I response. The interaction of the tumor cell TGF-B secretion with endogenous host TGF-B is involved in the progression of human breast carcinoma cells in vivo.

Mechanism of Action: Cancer growth is anchorage-independent. Tumor cells may grow in the absence of an ECM(extracellular matrix); they do not need to be anchored to a basement membrane to grow. Cancer cells may multiply free form while floating in the plasma or intracellular fluids. TGF-B comprises a family of several multifunctional structurally related polypeptides first identified for their ability to induce anchorage-independent growth of fibroblasts that are not affected by cancer. TGF-B suppresses the immune system by inducing angiogenesis, promoting the formation of stroma, and allowing the maintenance and progression of tumor cells in an intact host.

Experimental Models: Experimental inoculation of TGF-B cells in mice with no thymus decreases MHC I activity. Intraperitoneal injections of antibody to TGF-B (which neutralized this growth factor in vivo) suppressed intraabomdinal tumors and lung metastases. These anti-TGF-B antibodies transiently inhibited growth of established breast carcinoma subcutaneous tumors and resulted in immediate escalation in the normal host response to tumor (ie. MHC I activity)

Mechanism of Action: Upon viral infection or malignant cell transformation, the immune system generates novel sets of peptides (derived from the cervical cancer cells) which bind to MHC-I products. The stable MHC-I expression requires loading of the heavy chain/light chain dimer with antigenic peptide, which is delivered to MHC-I molecules in the endoplasmic reticulum by the peptide transporter, encoded by the TAP 1 and 2 genes. Once the MHC-I is loaded with antigenic peptide, it is now the target for elimination by CD8+ CTLs and can play a pivotal role in the eradication of virally or transformed cells.

Experimental Models: Immunohistochemical results from a polyclonal antiserum used to analyze TAP-1 function revealed a loss of TAP-1 in 37/76 carcinomas. This data demonstrates that malignant cells evade immune surveillance through the CTL pathway by inhibiting peptide transport(through the downregulation of TAP-1).

Applicability to other models: Absence of TAP-1/2 expression accompanied by loss of MHC-I expression has also been reported in small cell lung carcinoma cell lines. Preliminary results on lung, colon, mammary carcinomas indicate that the loss of immunoreactivity for TAP-1 is not only restricted to cervical carcinomas but is a generalized phenomenon applicable to many cancer pathologies.

A. Melanoma tumor cells are immunogenic; theoretically, they should cause an immune response but they do not stimulate an effective anti-tumor immune response in vivo.. Melanoma tumors may be capable of delivering antigen-specific signals to T cells, but do not deliver the costimulatory signals necessary for full activation of T cells because of the lack of B7 expression on their surface. T cell activation requires two distinct signaling events. The first signal originates from the binding of the TCR to its antigen-MHC ligand, and provides the specificity of the interaction. The second signal is either provided by soluble factors such as IL-2 or the interaction of cell-surface molecules on the T cell with their ligands on APCs. This second signal is thought to provide the necessary costimulation to the TCR-mediated signaling event. Binding of the TCR with peptide-MHC complexes in the absence of costimulation can result in T cell inactivation or anergy, which is associated with a block in the IL-2 gene transcription.

Expression of B7 on the surface of a cell is the costimulatory signal necessary to allow for the cytolytic CD8+ T cell attack on the tumor. The costimulation results from an interaction of the CD28 molecule on the T cell surface with its ligand,B7, on the surface of an antigen-presenting cell(APC). B7 display renders tumor cells capable of effective antigen presentation, leading to their eventual eradication.

Mouse melanoma cells were transfected with an expression vector containing mouse cDNA encoding B7. The B7-expressing transfectant was effective in providing costimulation to T cells and induced an immediate immune attack on the melanoma cells in vivo. Although melanoma rejection was not complete, provision of the costimluatory B7 significantly enhanced the anti-tumor response. This response was mediated by CD8+ T cells; CD4+ T cells were not required. Transfection with costimulatory molecules provides a means to elicit effective CTL responses to established tumors in the immunotherapy of primary or metastatic disease.

B. Influenza nucleoprotein tumors are immunogenic and theoretically should provoke an immune response. Like melanoma cells, they do not because of the lack of B7 expression on their surface. The B7 family of molecules is the most potent among costimulatoratory molecules. The B7-1 and B7-2 can interact with their counterreceptors CD28 and CTLA-4 on the T cells. In the complete absence of exogenous help, the B7-CD28 is sufficient and completely necessary for the costimulation of CD8+ CTL generation. While activities of B7-1/CD28 are highly dependent on the tumor system studied, the greatest effect of B7-1 transfection occurs with more inherently immunogenic tumors. The current hypothesis is that B7-1+ tumor cells can provide both signals one and two directly to activate native CTL in the absence of other bystander help.

Mouse influenza nucleoproteins were transfected with B7-1 and studied against mice that do not express B7-1-. For certain tumors that originate from "nonprofessional APC" such as epithelial cancers, the priming of naïve, tumor-specific CTL in vivo is mediated exclusively by the host’s bone marrow-derived APCs.The bone marrow chimeras now carrying B7-1 demonstrated that a tumor lacking the appropriate costimulatory molecules could not directly activate CTL, but does so indirectly by shedding tumor antigens for uptake, processing, and presentation by professional APCs.

-B7-1 expression on the tumor cells make the cell more susceptible to lysis in vivo.

This study demonstrates that the inability of the autologous host to reject resident tumor cells is frequently the result of inadequate generation of tumor-specific T cells. Mouse sarcoma cells here genetically engineered to provide both T-cell activation signals stimulate potent tumor-specific CD4+ T cells that case rejection of both engineered and wild-type neoplastic cells.

-demonstrates that coexpression of B7 by MHC Class II tumor cells induces immunity in the autologous host that is specific for naturally occurring tumor antigens of poorly immunogenic tumors.

-Essentially illustrates the role of B7 activation molecule in stimulating potent-tumor-specific CD4+ T cells that mediate rejection of wild-type tumors and provides a theoretical basis for immunotherapy of established tumors.

Organisms have evolved mechanisms to selectively eliminate unwanted self-cells such as cancer cells that grow uncontrollably and damage surrounding tissue. One of these mechanisms, called apoptosis, is programmed into the aberrant cell itself. Cells are programmed to undergo suicide in response to either a positive stimulus or by negative induction through the loss of a suppressor signal. Members of the tumor necrosis factor family of receptors (TNFR) are responsible for the positive initiation of the apoptotic pathway in response to a stimulus. Among the most common receptors in this family are TNFRI, TNFRII and Fas (also called CD95 or APO-1). While TNFRI and Fas can initiate the intracellular cascade of events that leads to apoptosis, TNFRII cannot. TNFRII is involved only in the activation of the transcription factor NFkB. TNFRI has dual signaling capabilities and can activate NFkB as well. However, its dual functions make it a less effect inducer of apoptosis than Fas.

The mechanism of apoptosis induction is similar for TNFRI and Fas. The binding of ligands (TNF for TNFRI and Fas-ligand for Fas) on the cell surface causes the receptors to trimerize. This in turn causes intracellular domains of the receptors, known as "death domains", to associate resulting in the recruitment of adaptor molecules. Adaptor molecules contain "death effector domains" that activates the caspase FLICE. FLICE then activates downstream effector caspases that commit the cell to apoptosis. Of the two apoptotic pathways, Fas is the more important because TNFRI rarely triggers apoptosis unless protein synthesis has also stopped.

The TNFRI and Fas-mediated apoptotic pathways are two of the methods used by the immune system to kill cancerous cells. They are independent of the perforin/granzyme B system used by CTLs and NKs to kill target cells. Two cell types of the immune system, T lymphocytes--mainly Th1-- and natural killer cells, can induce apoptosis in cancerous cells by releasing soluble ligands (TNF or Fas-ligand) or by expressing the ligands on their cell membrane. Ideally, these ligands bind to Fas and TNFRI on the surface of cancerous cells. However, many types of cancers have strategies to evade and exploit the Fas and also the TNFRI apoptotic pathways. These strategies include Fas-ligand expression by cancer cells, expression of Fas decoy receptors and expression of apoptosis suppressive oncogenes that can lead to receptor loss.

In the past, it was thought that the expression of Fas-ligand was restricted to cells involved in the immune response and certain "immune privileged" sites. Tissues where immune responses and associated inflammatory reactions can cause irreversible damage, such as the eye, have protective mechanisms to suppress the immune response. These sites are known as "immune privileged". Like "immune privileged" sites, many types of cancers have recently been found to express Fas-ligand. The expression of Fas-ligand can cause infiltrating lymphocytes bearing the Fas receptor to undergo apoptosis. In "immune privileged" site this prevents damaging immune responses from occurring. However, in tumors it prevents lymphocytes from destroying the tumor. Instead of the lymphocytes killing the cancer cells, the tumor kills the infiltrating lymphocytes (see diagram above). Fas-ligand expression and the killing of Fas-expressing T lymphocytes has been demonstrated in many tumor types including melanomas, sarcomas, some brain tumors, colon cancers, pancreatic cancers and stomach cancers.

In addition, cancer cells can express non-functional Fas receptors that bind Fas-ligand but do not initiate the caspase cascade that leads to apoptosis. These non-functional receptors have been termed decoy receptors because they deceive T cells and natural killer cells expressing Fas-ligand into believing that they are initiating apoptosis in the cancerous cells. Cancerous cells can also secrete soluble decoy receptors that similarly neutralize Fas-ligand. The recently identified DcR3 gene encodes one such receptor that lung and colon cancers were found to express (Pitti et al 1998).

Oncogenes expressed in cancer cells can also suppress responsiveness to Fas and TNFRI-mediated apoptosis. Mutation in the p53 tumor suppressor gene is associated with reduced expression of Fas on the surface of cancerous cells, making them less responsive to apoptosis. The overexpression of the bcl-2 gene correlates with malignant transformation and correlates negatively with Fas-mediated apoptosis sensitivity in tumors in vivo. However, the lack of sensitivity to apoptosis in bcl-2 expressing tumors was overcome in the presence of the cytokines IFN-gamma and TNF in vitro (Weller et al 1995). This suggests the potential immunotherapeutic benefit of these cytokines in tumors possessing the bcl-2 oncogene. Another oncogene, v-Rel, prevented TNF induced apoptosis in vitro and may induce the expression of other anti-apoptosis oncogenes such as bcl-2 (Zong et al 1997).

The mechanisms used by cancer cells to escape apoptosis are summarized in the table below:

Tumor growth occurs because cell production is greater than cell loss, whereas in normal adult tissues, the rates are equal. In virtually all types of human cancer, the machinery that regulates cell growth and division has malfunctioned.

The cell cycle has four stages: the G1 phase when the cell increases in size and gets ready to replicate its DNA, the S phase when the cell synthesizes or copies its chromosomes, the G2 phase in which the cell prepares to divide and the M phase when mitosis occurs. When the various growth inhibitory proteins and checkpoint controls which regulate this cycle become disabled due to mutations characteristic of cancerous cells, the cell cycle is no longer under tight regulation. Tumor cells are capable of proliferating so quickly that the immune response is not fast enough to keep their growth in check. The growth of the tumor cells outpaces the immune response.

Lack of cell cycle controls leads to excessive proliferation of tumor cells. Because mounting an immune response is based on the activation and consequent proliferation of normal B and T cells that recognize certain tumor antigens, the response is significantly slower than that of a tumor cell. The cells of the immune system still function under the auspices of the cell cycle clock and therefore are subject to inhibitory proteins which make sure that the cell has the proper environment before they are allowed to complete the cell cycle.

Tumor cells may also use numbers in another way to outsmart the immune system. In experiments where chemically induced cancers were transferred to syngeneic hosts, it has been shown that moderate cell doses of a tumor antigen will usually be rejected by the host immune system, but larger cell doses can break through and grow. In this case the tumor cells are too numerous for the immune system to effectively eliminate them.

Tests have also demonstrated that very small doses of antigenic tumor cells also grow progressively. These small doses are said to "sneak through" the immune surveillance system orchestrated by the host’s circulating lymphocytes. These tumor cells are capable of sneaking through because not enough of them reach the right cells for the body to know that it should elicit an immune response before the levels of tumor growth have become unmanageable.

Effect	TGF-beta	IL-10	VEGF
Inhibition of T-cell growth	+	-	+
Inhibition of CTL differentiation	+	+	+
Inhibition of cytokine production	+	+	-
Induction of T-cell anergy	+	-	-
Downregulation of cytotoxic potential	+	+	-
Inhibition of antigen presentation	+	+	-
Shift in the Th1-Th2 balance towards Th2	+	+	-
Downregulation of adhesion/costimulatory molecules	+	+	-
Resistance to CTL-mediated lysis	-	+	-

MECHANISMS FOR ESCAPE FROM APOPTOSIS
Mechanisms	Effect
Expression of Fas-ligand	T cell killing
Expression of decoy receptors	Blocking of apoptosis
Secretion of soluble receptors	Blocking of apoptosis
Receptor loss (p53)	No response to T cell ligands
Overexpression of Bcl-2 or v-Rel	Inhibition of apoptosis