Cancer Immunology

The immune system plays a central role in defending the body from infection, and also can play an important role in the host response to neoplasia. The ability of the immune system to detect and kill tumor cells has been called immune surveillance.^{1, 2} While many effective antitumor immune mechanisms have been described, it is not entirely clear whether antitumor immune responses operate in immune surveillance, at least as this concept was defined originally. For example, patients who are immune-deficient, such as people who have AIDS, do have a higher frequency of cancer. However, the tumors that develop in these patients tend to be lymphoproliferative tumors, such as lymphoma, or unusual forms of cancer that are associated with virus infection, such as Kaposi sarcoma.³ Therefore, the natural role of the immune system in preventing cancer may be more restricted than originally envisioned.⁴ However, there is no doubt that the immune system can interact with tumor cells in various ways, and that immune responses, whether natural or induced, can lead to tumor regression. In fact, a recent report provides compelling evidence that the immune system can maintain tumors (methylcholanthrene-induced sarcomas in a murine model) in a state of equilibrium, and that this equilibrium is mediated by tumor-specific adaptive immune system responses.⁵  As more is learned about the interactions between cancer and the immune system, opportunities for new immunotherapeutic and immunodiagnostic approaches become apparent. In this chapter, a brief overview of immunology and of biologic therapies relevant to gynecologic cancers is presented.

Various types of human immune response can be used to target and respond to tumor cells.^{2, 6, 7, 8} Immune responses historically were categorized as humoral or cellular, a distinction based on the observation in experimental systems that some immune responses could be transferred by serum (humoral) and others by cells (cellular). Generally, humoral responses refer to antibody responses. Antibodies are antigen-reactive, soluble, bifunctional molecules that are composed of specific antigen-binding sites that are associated with a constant region that directs the biologic activities of the antibody molecule, such as binding to effector cells or complement activation. Cellular immune responses are mediated directly by activated immune cells rather than by the production of antibodies. Most immune responses involve both humoral and cellular components; specific immune responses involve the coordinated activities of populations of lymphocytes, operating in concert with each other and with antigen-presenting cells, resulting in some effector function. Cellular interactions involved in immune responses include direct cell-to-cell contact as well as cellular interactions mediated by the secretion of and response to cytokines, biologic messenger molecules that play important roles in the genesis, amplification, and effector functions of immune responses.

Immune responses also can be categorized as involving innate or adaptive immunity. Adaptive immunity refers to antigen-specific responses, which include the evolution of antigen-specific immunologic memory, and the expansion and amplification of antigen-specific effector functions. Innate responses refer to a variety of nonantigen-specific mechanisms, which are present at all times and are not amplified after repeated exposure to a given antigen.² An example of an adaptive immune response would be the generation of specific cytolytic T cells (CTL) directed to a tumor-associated antigen. An innate antitumor response would be the killing of tumor cells by natural killer (NK) cells. Both innate and adaptive immune responses can exert potent antitumor activity.

T Lymphocytes and Antitumor Immunity

T lymphocytes play a pivotal role in the generation of immune responses, by acting as helper cells in the generation of humoral and cellular immune responses, and as effector cells in cellular responses.^{6, 7} T-lymphocyte precursors mature into functional T lymphocytes in the thymus, where they learn to recognize antigen in the context of the major histocompatibility complex (MHC) molecules expressed by each individual. Most T lymphocytes with the capability of responding to 'self' are removed during thymic development, eliminating those T cells that might exert autoimmune responsives. T lymphocytes can be distinguished from other types of lymphocytes by their biologic activities and by the expression of distinctive cell surface molecules, including the T-cell antigen receptor and CD3 molecular complex. The expression of lymphocyte cell surface molecules can be quantified in the clinical immunology laboratory by flow cytometry with fluorochrome-labeled monoclonal antibodies that can bind these molecules specifically. There are two major subsets of T lymphocytes: T helper/inducer cells, which express the CD4 cell surface marker, and T cytotoxic cells, which express the CD8 marker. CD4 T lymphocytes can provide help to B lymphocytes, resulting in antibody production, and also can act as helper cells for other T lymphocytes. Much of the helper activity of T lymphocytes is affected by the production of cytokines, such as interleukin-2 (IL-2). The CD8 T-lymphocyte subset includes cells that are cytotoxic (i.e. can kill target cells directly).

A major biologic function of CTLs is the lysis of virus-infected cells. However, CTLs can directly mediate the lysis of tumor cells, presumably by recognizing unique antigens presented by tumor cells. CTLs can kill tumor cells by signaling the induction of apoptosis in the target cells, and by the secretion of perforin, a pore-forming protein.⁶ T cells also can contribute to antitumor immune responses by producing cytokines such as tumor necrosis factor-α (TNF-α), which induce tumor cell lysis and can enhance other antitumor cell effector responses.

T lymphocytes recognize specific antigens by interactions that involve the T-cell antigen receptor.^{6, 7} In terms of its general structure and molecular organization, the T-cell antigen receptor is reminiscent of the antibody molecule. However, there are major differences between the antigen receptor molecules on B lymphocytes and those on T lymphocytes; for one, the T-cell receptor is not secreted. Also, the B-cell antigen receptor interacts with antigen in a different way than the T-cell receptor; the T-cell antigen receptor can see antigen only when it is presented to it in association with self MHC molecules, on the surface of an antigen-presenting cell. The B-cell antigen receptor can bind antigen directly, and therefore is not restricted in this way.

B Lymphocytes and Antibodies

B lymphocytes are the cells that produce and secrete antibodies.² After exposure to antigen and appropriate activation signals, they differentiate to become plasma cells, terminally differentiated cells that produce large quantities of antibodies. Pre-B cells originate from progenitor stem cells after the rearrangement of immunoglobulin genes from their germ cell configuration to the configuration that result in a functional immunoglobulin molecule. Mature B lymphocytes use cell surface immunoglobulin molecules as antigen receptors. In addition to producing antibodies, B lymphocytes play another important role: they can serve as efficient antigen-presenting cells for T lymphocytes.

The production of antitumor cell antibodies does not appear to play a central role in host antitumor immune responses, but monoclonal antibodies that are reactive with tumor-associated antigens may prove useful in antitumor therapy, or for tumor detection. Immunotoxin-conjugated monoclonal antibodies directed to antigens expressed by human ovarian adenocarcinoma effect tumor cell killing in experimental animal systems.⁹ Several obstacles had to be overcome before monoclonal antibodies were of practical clinical utility, especially the modification of monoclonal antibodies to minimize host immune responses directed to foreign antigens on monoclonal antibodies of murine origin. Because most monoclonal antibodies are of murine and not human origin, the host's immune system can recognize and respond to murine monoclonal antibodies. The preparation of humanized murine monoclonal antibodies (i.e. genetically engineered monoclonal antibodies composed of human constant regions with specific antigen-reactive murine variable regions) has alleviated many of the problems associated with the administration of murine monoclonal antibodies. Monoclonal antibody-based agents are being used increasingly in antitumor therapies.

Macrophages, Monocytes, and Dendritic Cells

Monocyte/macrophages play important roles in immune responses. Macrophages, which form an important part of innate immune responses, also play a key role in the generation of adaptive, antigen-specific, lymphocyte-mediated immune responses, because they can act as effective antigen-presenting cells. Helper/inducer (CD4) T lymphocytes, bearing a T-cell receptor of appropriate antigen and self specificity, can be activated by antigen-presenting macrophages that display processed antigen combined with self MHC molecules (MHC class II molecules). Antigen presenting cells also provide important costimulatory signals that are important for the induction of T-lymphocyte activation. In addition to serving as antigen-presenting cells, macrophages can ingest and kill microorganisms, and can act as cytotoxic, antitumor killer cells. Also, macrophages and monocytes produce various cytokines, including IL-1, IL-6, chemokines (RANTES, MIP1α, MIP1βIL-8), IL-10, and TNF, which can be involved in both innate and adaptive immune responses, and which can have direct biologic effects on tumor cells, both as growth-inducing and growth-inhibiting factors.

Mature dendritic cells also are efficient antigen-presenting cells for T cells.¹⁰ In fact, the dendritic cells found in lymphoid tissue are the most potent stimulators of naive T cells. Mature dendritic cells express very high levels of MHC class I and class II molecules, as well as high levels of adhesion molecules and cell surface lymphocyte stimulatory molecules, such as the B7 molecule. Also, these dendritic cells secrete chemotactic factors (chemokines) that attract T cells. Together, these features explain, in part, their ability to act as highly efficient antigen-presenting cells for T lymphocytes.

Natural Killer Cells and Antibody-Dependent Cellular Cytotoxicity

NK cells have characteristically large granular lymphocyte morphologic features. NK cells do not express the CD3–T-cell receptor complex, but they can express some markers that are shared with T lymphocytes or other types of lymphocytes, in addition to other NK-associated markers. NK cells can lyse target cells, including tumor cells, unrestricted by the expression of antigen on the target cell.^{6, 7, 11} Therefore, NK cells are effector cells in an innate (nonantigen-restricted) type of immune response, and they play a vital role in the nonspecific killing of tumor cells or virus-infected cells. While NK cells represent an innate form of immunity that does not require an adaptive memory response for biologic function, NK responses can be augmented by cytokines, such as IL-2. NK cells can be stimulated to kill target cells that do not express MHC class I molecules. MHC class I molecules, which normally are expressed on all nucleated cells, play a central role in the display of endogenously synthesized, intracellular antigens to CD8⁺ CTLs. Many viruses subvert CTLs by down-regulating the expression of MHC class I molecules, effectively hiding such virus-infected cells from T-cell killing. However, NK cells can then detect and kill such MHC class I-negative cells. Therefore, NK cells provide an important fail-safe role in the detection of virus infected host cells that have evaded T-cell immunity.

The cells that can effect antibody-dependent cellular cytotoxicity (ADCC) are NK-like cells. ADCC can result in the lysis of tumor cell targets in vitro. Although the mechanisms of tumor cell killing in ADCC are not clearly understood, close cellular contact appears to be required between the ADCC effector cell and the target cell.

Cytokines, Lymphokines, and Immune Mediators

Cytokines are soluble mediator molecules that induce, enhance, or affect immune responses. Cytokines are heterogenous, representing several molecular superfamilies, typically have multiple and redundant biologic actions, are produced by various types of cells, and play critical roles, not only in immune responses but also in biologic responses outside of the immune response, such as hematopoiesis or the acute-phase response. Most cytokines are small or medium secreted proteins, are produced transiently and locally, and interact with specific cellular receptors, resulting in signal transduction followed by changes in cellular proliferation or differentiation in the target cell.

Although cytokines are heterogeneous and generally share little structural or amino acid homology, some cytokines appear to have evolved from common ancestral precursors, and are classified as forming members of a cytokine molecular superfamily. Cytokine superfamilies include the hematopoietins (IL-2, IL4, IL-6, and related cytokines, interferons), the TNF-like cytokines (TNF-α, lymphotoxin and related molecules), chemokines (IL-8, RANTES, and related cytokines), the IL-1-like cytokines (IL-1α, IL-1α, IL-1RA, IL-18), and the IL-17 family of cytokines (Table 1).

Table 1. Cytokines: molecular superfamilies

IL-2 induces T-lymphocyte proliferation, and is produced primarily by activated T lymphocytes.⁶ T lymphocytes secrete IL-2, express the IL-2 receptor, and respond to IL-2 by proliferating. Much of the helper activity of T lymphocytes is mediated by IL-2. IL-2 interacts with specific IL-2 receptors on the surface of the target cell; the high-affinity IL-2 receptor is composed of three polypeptides, the CD25α, CD122β, and CD132γ chains, which unassociated have much lower binding affinity for IL-2. T-cell activation results in greatly increased numbers of the high-affinity receptor for IL-2 on the surface of these cells. IL-2 has other biologic activities, including the induction of B-lymphocyte activation and maturation, and monocyte and NK-cell activation.

B-lymphocyte activation and differentiation to immunoglobulin-secreting plasma cells are enhanced by cytokines that are produced by helper T lymphocytes or monocytes. Several cytokines, originally described as B-cell-stimulating factors, IL-4, IL-5, and IL-6, are now known to have additional biologic activities. For instance, IL-6, a factor that can induce B-lymphocyte differentiation to immunoglobulin-secreting cells, is a pleiotropic cytokine¹² with biologic activities that include the induction of cytotoxic T-lymphocyte differentiation, the induction of acute-phase reactant production by hepatocytes, and activity as a colony-stimulating factor for hematopoietic stem cells.  There are several IL-6-like cytokines, including IL-11, Oncostatin-M, leukemia-inhibitory factor (LIF), and ciliary neurotropic factor (CNTF).¹² Also, a virus-encoded IL-6 (vIL-6) homologue of human IL-6, the human herpesvirus type 8 (HHV-8) open reading frame (ORF)-K2 gene product, is similar to human IL-6 in structure and function.¹³ Interestingly, several types of tumor cells produce IL-6, and IL-6 has been proposed to act as an autocrine and paracrine growth factor for different types of neoplasms.¹⁴ However, IL-6 may be an effective antitumor agent because of its ability to enhance antitumor T-cell-mediated immune responsiveness.¹⁵

IL-3, a factor that can enhance the early differentiation of hematopoietic cells,¹⁶ may have a role in immunotherapy because of its ability to induce hematopoietic differentiation in people undergoing aggressive chemotherapeutic treatment or bone marrow transplantation. IL-10 is an immune-inhibitory cytokine.¹⁷

IL-12 is a T-cell-stimulating and NK-stimulating cytokine, which is thought to play a key role in the induction of type 1, or TH1, immune responses.¹⁸  IL-23 and IL-27 are IL-12-like cytokines that have unique biological effects, including the induction of inflammatory responses against infection, and the regulation of B and T cell activation.¹⁹ 

IL-17 cytokines are recently described molecules that are involved in inducing and mediating proinflammatory responses.  IL-17 can induce the expression of several other cytokines and chemokines by several types of cells.^{20, 21}

TNF-α is a cytokine that can be cytotoxic for tumor cells, can enhance immune cell-mediated cellular cytotoxicity, and can activate macrophages and induce monokine secretion. Other biologic activities of TNF-α include the induction of cachexia, inflammation, and fever. This cytokine is an important mediator of endotoxin shock.

There are three types of interferons: interferon-α (IFN-α), interferon-β (IFN-β), and interferon-γ (IFN-γ).^{6, 18} Interferons, which form a family of molecules within the hematopoietin superfamily, are cytokines that can interfere with viral production in infected cells and, in addition to this, have various effects on the immune system. For instance, IFN-γ, a T-lymphocyte-produced cytokine, can affect immune function by enhancing the induction of MHC molecules expression, enhancing the activity of antigen-presenting cells, and thereby enhancing T-lymphocyte activation.

As research has provided new information on the biologic activities of cytokines, these factors have appeared to be extraordinarily pleiotropic, with a bewildering array of biologic activities, some of which occur outside of the immune system. However, the recognition of two types of helper T cells, the type 1 (TH1) and type 2 (TH2) subsets of CD4⁺ helper T cells, has brought some order to this otherwise bewildering situation. These helper T cell subsets are defined by the pattern of cytokines that these cells produce.  TH1 and TH2 subsets control the nature of an immune response by secreting characteristic and mutually antagonistic sets of cytokines.⁶ TH1 clones produce IL-2 and IFN-γ, while TH2 clones produce IL-4, IL-5, IL-6, and IL-10 (Table 2). TH1 responses are characterized by enhanced cell-mediated immune and inflammatory responses, including increased macrophage activation and T-cell proliferation and activation. TH2 responses are dominated by enhanced humoral responses, including increased B-cell activation and antibody production. Most immune responses involve both TH1 and TH2 responses, although some responses are dominated by either a TH1 or TH2 response pattern. The pattern of cytokines produced by TH1 or TH2 cells determines the nature of subsequent immune responses. For example, IL-10, a TH2 cytokine, inhibits the production of IFN-γ and other TH1 cytokines. Also, IL-10 down-regulates class II MHC expression on monocytes, resulting in a strong reduction in the antigen-presenting capacity of these cells, and acts as a direct B-cell stimulatory factor. In contrast, IL-12, a potent stimulator of IFN-γ production, promotes TH1-type immune responses.¹⁸ Because some cytokines have direct or indirect antitumor or immune-enhancing effects, several of these factors have been used in the experimental treatment of cancer.

Table 2. Cytokines: biological activities

T cell subsets other than TH1 and TH2 cells have been identified, in recent work.  Two significant subsets are regulatory T cells (Treg) and TH17 cells.  Treg cells are a functionally-defined subset of T cells that can inhibit other immune cells, which have been the focus of much attention.  Treg cells, which are CD4+, CD25+ and Foxp3+ T cells, produce IL10 and TGFβ, cytokines that down-regulate many immune responses.²²  It is believed that Treg play an important role in regulating immune responses, especially in preventing autoimmune anti-self responses by maintaining tolerance.²³  Treg dysfunction also may contribute to the growth of cancer.²⁴

IL-23, in conjunction with IL-6 and TGF-β1, stimulates naive CD4 T cells to differentiate into a novel subset of T cells, TH17 cells, which are characterized by the production of IL-17, are distinct from the TH1 and TH2 subsets, and which are associated with autoimmune responses.  The development of TH17 cells is inhibited by IFN-γ and IL-4, TH1 and TH2 cytokines, and these cells are resistant to supression by TH1 or TH2 cytokines.²⁵

The precise roles of cytokines in antitumor immune responses have not been described completely. Because cytokines are pleiotropic, they may exert antitumor effects through many different direct or indirect activities. In fact, it is possible that a single cytokine could both enhance tumor growth directly by acting as a growth factor, while enhancing anti-tumor immune responses. However, the potential of cytokines to enhance antitumor immune responses has been exploited in various strategies for the experimental treatment of cancer. These include the enhancement of host cytokine production induced nonspecifically by exposure to biologic response modifiers; direct treatment with recombinant cytokines; the use of adoptive immunotherapy, in which patient peripheral blood cells or tumor-infiltrating lymphocytes (TIL) are exposed to cytokines and activated ex vivo, generating activated cells with antitumor effects that can then be readministered to the patient; and gene therapy-based approaches in which tumor cells are transduced with a cytokine gene, the expression of which will presumably enhance antitumor immune responses, or even by the modulation of local cytokine production induced by drugs. Also, it is important to note that the antitumor effects of cytokines might be modulated by soluble receptors or blocking factors; blocking factors for TNF and for lymphotoxin were found in ascites of patients with ovarian cancer.²⁶

Cytokines can have growth-enhancing effects for tumor cells: cytokines can act as autocrine or paracrine growth factors for human cancer cells, including tumor cells of nonlymphoid origin. For instance, IL-6, which is produced by various other types of human tumor cells, can act as an autocrine growth factor for human myeloma, Kaposi sarcoma, and renal carcinoma.^{27, 28, 29, 30} Interestingly, vIL-6, the viral homologue of human IL-6 that is encoded by HHV-8, may act as a paracrine growth factor for multiple myeloma cells³¹ and in Castleman disease,³² as well as for Kaposi sarcoma.³³

Clearly, cytokines are of great potential value in cancer treatment and also may play roles in the pathogenesis of various cancers by acting as tumor growth or viability-enhancing factors. Because of their multiple, even conflicting, biologic effects, a thorough understanding of cytokine biology will be essential for the successful use of these molecules in cancer treatment.

Tumor development has been associated with the inappropriate expression or mutation of various genes: oncogene overexpression, amplification, or mutation; tumor suppressor gene (antioncogene) expression or mutation; or the expression of  genes for cytokines and growth factors and the cellular receptors for these molecules. Subversion of host antitumor immune responses also may play a role in the development of cancer.

Cytokines and growth factors may play important roles in the development and growth of ovarian cancer; in fact, epithelial ovarian cancer may be a cytokine-propelled disease.^{14, 34, 35, 36} Cytokines, including IL-1 and IL-6, enhance the proliferation of ovarian cancer cells,^{37, 38} and various cytokines are produced by ovarian cancer cells, including macrophage colony-stimulating factor (M-CSF), granulocyte-macrophage colony-stimulating factor (GM-CSF), IL-1, TNF-α, and IL-6.^{39, 40, 41} Many ovarian cancer cells produce both M-CSF and fms, the M-CSF receptor.^{38, 39, 40, 42, 43} Levels of fms messenger RNA correlated strongly with ovarian tumors of high histologic grade and advanced clinical stage, and were associated with a poor clinical outcome.⁴⁰ In addition, elevated plasma levels of M-CSF were seen in 70–80% of patients with ovarian cancer.^{38, 40, 42} Because M-CSF-stimulated macrophages might produce other cytokines, such as IL-1 or IL-6, that can further stimulate tumor cell growth, M-CSF potentially could act as both an autocrine and paracrine tumor stimulatory factor and as a factor that can modify the host environment, resulting in enhanced tumor cell growth.⁴⁰

Role of Interleukin-6 in the Pathogenesis of Ovarian Cancer

IL-6 is a pleiotropic growth- and differentiation-inducing factor^{12, 44} that acts as an autocrine and paracrine growth factor for several types of human tumors, including multiple myeloma cells,^{28, 45} renal cancer cells,³⁰ and AIDS-associated Kaposi sarcoma.²⁹ IL-6 also acts as a programmed cell death-preventing factor for B-cell tumors.⁴⁶

Ovarian cancer cells of epithelial origin show detectable levels of IL-6 gene expression and produce and secrete variable quantities of IL-6.²⁷  Several cytokines (IFN-γ, IL-1, and TNF) can up-regulate IL-6 production by ovarian tumor cells.²⁷ Primary cultures of normal human ovarian epithelium also produced IL-6.⁴⁷ Elevated levels of IL-6 also have been seen in vivo in ovarian cancer; IL-6 was detected in the ascitic fluid²⁷ and serum⁴⁸ of women with ovarian cancer. Ascitic fluids isolated from patients with ovarian cancer had elevated levels of IL-6.²⁷ Elevated serum IL-6 levels were seen in ovarian cancer, and they correlated with the presence of more extensive disease.⁴⁸ In another study, serum levels of IL-6 in ovarian cancer patients were seen to correlate with disease-free survival time, as well as with FIGO stage.⁴⁹  

We studied the in vivo biologic effects of IL-6 by correlating levels of IL-6, acute-phase proteins (C-reactive protein [CRP], haptoglobin), and immunoglobulins (Ig) M, A, and G in the ascitic fluid or serum of patients with ovarian cancer. Significantly elevated levels of IL-6, CRP, haptoglobin, and IgA, and decreased levels of IgM were seen in ovarian cancer ascites (n = 36), compared with a control group (donors with nonmalignant gynecologic conditions, n = 20) (unpublished observations). Again, elevated serum IL-6 levels (6.5 ± 5.2 vs. .2 ± .02 U/mL) were seen in ovarian cancer (n = 7), and serum CRP levels were increased compared with levels in control subjects (n = 7). The levels of CRP in ascites correlated with IL-6 content (not shown). These results suggest that the IL-6 that is present in vivo in patients with ovarian cancer is biologically active because IL-6 can act both as a B-cell-stimulating factor, inducing the production of immunoglobulins, and as a potent inducer for the production of acute-phase reactants.^{12, 50, 51, 52} In a related study, serum levels of various cytokines, including IL-6, were examined in people with ovarian cancer; serum levels of IL-1α, IL-1β, IL-6, TNF-α, sIL-2R, and CRP were significantly increased in patients compared with controls, consistent with the previously reported hypothesis that high IL-6 and/or CRP serum levels may represent an important and independent prognostic factor of the likely outcome in cancer patients.⁵³

Other observations suggest that IL-6 can act as a growth factor for ovarian cancer cells. Culture supernatants from activated human monocytes, which support ovarian tumor growth, contained significant amounts of IL-6, and anti-IL-6 serum inhibited part of this monocyte-produced growth-supporting activity.³⁷ Therefore, IL-6 has the potential to act as a paracrine growth factor for ovarian cancer cells. We examined the possibility that IL-6 acts as an autocrine growth factor for ovarian cancer cells by inhibiting IL-6 gene expression by exposure to IL-6 antisense oligonucleotides. The result was greatly decreased cellular proliferation.⁵⁴ These results suggest that endogenous IL-6 production is needed for optimal cell growth.⁵³ In other work, it was seen that IL-6 enhanced the resistance of ovarian cancer cells to apoptosis induced by cytotoxic drugs.⁵⁵ Also, IL-6 expression was seen to be elevated in drug-resistant ovarian cancer cells.⁵⁶ In our own work, we saw that pretreatment of ovarian cancer cells with IL-6 resulted in decreased sensitivity to cisplatin-induced cytotoxicity.⁵⁷ In more recent work, it was seen that IL-6 secreted by ovarian cancer cells can serve as a potent angiogenesis-promoting factor.⁵⁸ Together, these results suggests that this cytokine can promote enhanced viability, and resistance to cytotoxic drugs, in ovarian cancer cells, as well as promote angiogenesis, activities which could promote the growth of these cancers.

IL-6 also may enhance ovarian cancer cell growth indirectly, by inhibiting T cell activation. In fact, it has been seen that elevated serum levels of IL-6 and other cytokines were correlated with decreased immune cell activation.⁵⁹ Clearly, IL-6 can act as an autocrine and paracrine growth factor for ovarian cancer cells, but the complete role of IL-6 in the pathogenesis of ovarian cancer is not clear, because IL-6 has the potential to modulate antitumor immune responses.

Role of Interleukin-10 in the Pathogenesis of Ovarian Cancer

Epithelial malignancies of the ovary usually are confined to the peritoneal cavity, even in advanced stages of the disease, so it has been suggested that the growth of ovarian cancer intraperitoneally could be related to the local deficiency of antitumor immune effector mechanisms.⁶⁰ Ascitic fluid from patients with ovarian cancer contained significantly elevated levels of IL-10,⁶¹ a cytokine that originally was identified as cytokine synthesis inhibitory factor because of its activity as an inhibitor of cytokine production.⁶² IL-10, a 35- to 40-kd cytokine, is produced by the type 2 CD4⁺ helper T-cell clones (TH2), and inhibits the cytokine production of the type 1 subset (TH1). We measured the levels of various cytokines in ascites that were obtained from women with ovarian cancer and found elevated levels of various cytokines compared with the levels in peritoneal fluid from women who did not have cancer. Additionally, 50–60% of patients had significant ascitic levels of IL-2, TNF-α, IFN-γ, G-CSF, and GM-CSF. Levels of various cytokines, including IL-6, IL-10, TNF-α, G-CSF, and GM-CSF were significantly elevated (unpublished observations). A similar pattern was seen in sera from women with ovarian cancer, with IL-6 and IL-10 frequently detected.

Our preliminary results showed that ovarian cancer cells do not produce IL-10 (unpublished observations). In a subsequent study, it was seen that monocytes make the IL-10 seen in ovarian cancer ascites, and that such IL-10-producing monocytes inhibited T cell activation.⁶³ In a recent report, the presence of elevated levels of IL-10 in patients with ovarian cancer was confirmed, and it was seen that ovarian cancer cells, including ovarian cancer cell lines, produced IL-10, in contrast to our earlier results.⁶⁴  High levels of IL-10 may result in a peritoneal environment characterized by immune unresponsiveness, thereby promoting tumor cell growth.  

Epithelial ovarian cancer is responsive to many chemotherapeutic agents, but despite improvements in surgery and chemotherapy, the 5-year survival rate remains low, at approximately 20–30% for advanced stage disease.⁶⁵ Also, because the disease typically eludes early detection techniques, most patients present in advanced stage. Therefore, there is a great need to develop useful biologic therapies, and innovative approaches to the control of metastatic residual epithelial ovarian cancer are being sought. Ovarian cancer represents an attractive target for these therapies: the bulk of ovarian cancer occurs in the peritoneal cavity, making the regional administration of biological therapy theoretically attractive. Patients with small-volume residual peritoneal disease are good candidates for biological therapies and immunotherapy, particularly regional peritoneal immunotherapy.^{66, 67, 68} Further, patients with advanced disease are significantly immunocompromised,⁶⁹ suggesting a role for immune-enhancing therapeutic approaches. Advances in molecular biology, biotechnology, immunology, and cytokine biology have resulted in the availability of new, promising immunotherapeutic approaches for ovarian cancer.

Dysplastic cervical epithelial cells also present an attractive target for immune enhancement-based prophylactic (vaccine) and immune-based therapeutic strategies, because these cells reflect infection with an oncogenic virus (HPV).  In addition to this, effective vaccines for the prevention of HPV infection have been developed.

Enhancement of Specific Immune Responses to Human Papillomavirus-Infected Cervical Epithelial Cells

HPV, more specifically a subset of HPV subtypes (HPV 16, 18, 31, and 45), has been implicated as the central etiologic agent in the majority of cervical cancers. HPV-infected dysplastic and cancerous cervical epithelial cells consistently retain and express two of the viral genes, E6 and E7, which respectively interact with, and disrupt the function of, the p53 and RB tumor suppressor gene products. Because an infectious agent (HPV) plays a clear and central role in the pathogenesis of this gynecologic cancer, opportunities exist for treatment and prevention strategies based on the enhancement of antiviral immune function, especially because it appears that factors other than infection with HPV, such as cellular immune function, play an important role in determining whether the infection of cervical epithelial cells will regress or progress to cancer. This has led to the development of prophylactic and/or therapeutic vaccines to HPV, as well as treatment approaches based on the enhancement of host immune function.^{70, 71, 72, 73} In fact, an HPV vaccine has been shown to have an exceptional level of efficacy in a recent clinical trial: an HPV-16 and HPV-18 virus-like particle vaccine was seen to result in 100% efficacy in the reduction of the incidence of persistent HPV-16 and HPV-18.⁷³ All cases of HPV-16 and HPV-18-related cervical intraepithelial neoplasia seen in this study occurred among the placebo recipients. Therefore, administration of this HPV-16/HPV-18 vaccine clearly reduced the incidence of both HPV-16-18 infection and HPV-16-18-related cervical intraepithelial neoplasia. These findings suggest that immunization of HPV-negative women with similar vaccines, especially polyvalent vaccines providing immunity to several oncogenic HPV subtypes, will markedly reduce the incidence of cervical cancer in the future, especially in developing countries where clinical screening for cervical dysplasia is not widely available.⁷⁴

Immunotherapy with Biologic Response Modifiers

Most early experimental immunotherapies for metastatic ovarian cancer involved biologic response modifiers, such as Corynebacterium parvum (a heat-killed, Gram-negative anaerobic bacillus), bacillus Calmette-Guérin (BCG), Freund's complete adjuvant, or modifications of these agents.⁷⁵ Although initial studies with nonspecific immunotherapy, such as BCG and C. parvum, demonstrated tumor regression in some studies, these therapies did not prolong survival when combined with cytotoxic chemotherapy in randomized trials.⁶⁵

Monoclonal Antibodies and Immunotherapy

Monoclonal antibodies have played an important role in the detection of tumor markers; OC-125, a monoclonal antibody that is reactive with a molecule produced by epithelial ovarian carcinoma cells, is used widely to monitor blood CA-125 antigen levels.⁷⁶ Monoclonal antibodies can affect antitumor responses in various ways: by complement activation and tumor cell lysis, by directly inducing antiproliferative effects, by enhancing the activity of phagocytic cells, or by mediating ADCC. Also, monoclonal antibody-based therapeutic strategies can target or block the biologic function of cell surface signaling molecules.⁶⁶

Anti-HER2 Antibody

Because the HER-2/neu oncogene (HER-2) is overexpressed in many cancers, it has been suggested that the HER-2 antigen, a transmembrane protein tyrosine kinase that is homologous to the human epidermal growth factor receptor, be used as the target for immunotherapy.⁷⁷ In an experimental animal model system, a monoclonal antibody directed to HER-2 was seen to enhance human tumor cell susceptibility to TNF and cisplatin, suggesting that anti-HER-2 monoclonal antibodies could modulate the in vivo effects of HER-2/neu.⁷⁸ In fact, anti-HER-2 monoclonal antibodies, specifically, the humanized monoclonal antibody Herceptin (Genentech, South San Francisco, CA, USA), has been tested in several clinical trials and found to be an effective adjuvant therapy for HER-2-positive breast and ovarian cancer patients.^{79, 80} This monoclonal antibody has established itself as an active agent, in conjunction with paclitaxel chemotherapy, in relapsed breast cancer patients who express the HER2/neu protein product. HER-2 is believed to play an important role in the pathogenesis of ovarian cancer; elevated levels of this oncogene were seen in approximately one third of ovarian cancers.⁸¹ The Gynecologic Oncology Group (GOG) has conducted phase 1/2 trials of Herceptin in ovarian cancer.⁸² Among patients with recurrent ovarian cancer, the rate of HER2/neu overexpression was 11% (i.e. lower than expected). In a trial of 41 patients, the overall response rate is only 2% (GOG data, Society of Gynecologic Oncologists, 2000). The drug is undergoing testing in combination with platinum and taxane-based chemotherapy to determine its activity in that setting. Clearly, not all patients respond to Herceptin therapy, and many patients who initially respond have resistance within 1 year of treatment. This has led some workers to suggest that vaccination strategies that generate T cell responses to HER-2 be developed.⁸⁰

Anti-CA-125 monoclonal antibodies

OvaRex (B43.13, anti-CA125 monAb)

A recent innovative approach to immunologic cancer therapy involves administration of monoclonal antibodies that target tumor-specific antigens circulating in the bloodstream, in addition to antigens on tumor cells themselves.⁸³ Oregovomab (B43.13) is a murine monoclonal antibody to CA-125, which is seen in more than 90% of patients with late-stage ovarian cancer.⁸⁴ This monoclonal antibody binds to circulating CA-125 antigen, resulting in the formation of immune complexes (antibody:antigen complexes) that are recognized as foreign, because these complexes include foreign (murine) antibody. These immune complexes are believed to be taken-up by antigen presenting cells, allowing the processing of the autologous CA-125 antigen, leading to induction of CA-125-specific antibodies, helper T cells, and cytolytic T cells. The antigen processing of the autologous antigen-xenotypic antibody complex is altered, relative to the processing of either component alone, and cellular and humoral immune responses directed to the tumor antigen have been demonstrated after treatment. Administered via 20-minute intravenous infusion, low-dose monoclonal antibodies such as B43.13 are better-tolerated compared with previous high-dose murine antibody approaches.   The possible therapeutic value of this low-dose monoclonal antibody approach was serendipitously discovered in a diagnostic study with B43.13, which was first developed as a tumor-imaging agent, because of its high affinity for CA-125. In a study using technetium (Tc) 99m-radiolabeled B43.13 for the immunoscintigraphic detection of recurrent ovarian cancer, it was noted that many patients exhibited unexpectedly long survival times.^{83, 85, 86, 87} Efforts are underway to corroborate and extend these findings by assessing the therapeutic efficacy of nonradiolabeled B43.13 in a series of prospective placebo-controlled and open-label studies in patients with ovarian cancer. The therapeutic objectives are to extend the period of disease stabilization while maintaining quality of life and to supplement surgical and chemotherapeutic approaches so that greater survival with extended quality of life is achieved.  Furthermore, research characterizing the immune-altering properties of the B43.13-CA-125 murine monoclonal antibody:tumor antigen complex continues to refine the understanding of the immunobiological mechanisms involved in this therapeutic approach.^{88, 89, 90} B43.13 has been seen to modify antigen processing of CA-125 by dendritic cells in vitro.^{91, 92}

Bookman and associates explored B43.13 treatment in a randomized, double-blind, placebo-controlled study of 55 patients with no clinical or radiographic evidence of disease but with elevated CA-125 levels (<gt>35 U/mL) after front-line therapy.⁹³ Patients relapsed rapidly in this clinical population, limiting the value of the study, because treatment was discontinued on clinical relapse. In a subpopulation of patients who had time to mount an immune response and had small-diameter residual disease, a trend similar to that observed in the study by Ehlen and associates⁹⁴ was noted, with a 6-month progression-free survival of 75% for the B43.13 treated group and 35% for the placebo group (p = 0.1).  Method and associates studied the impact of B43.13 dose frequency and schedule in patients with stage III/IV ovarian cancer who had completed front-line therapy with no evidence of disease.⁹⁵ This 102-patient study involved a randomization of patients into three dosing groups. While the study is ongoing, preliminary results suggest that generation of an optimal immune response requires more than two doses. However, increasing the dosing intensity to six monthly injections at initiation did not boost the frequency or magnitude of induced immunity. Several phase 2 studies have been completed in patients with known recurrent ovarian cancer. Ehlen and associates conducted an open-label trial of 13 patients with recurrent ovarian cancer with CA-125 levels exceeding 35 U/mL, multiple previous regimens of chemotherapy, and poor prognosis.⁹⁶ This treatment was seen to induce T cell immune responses, assessed using an IFNγ ELISPOT assay. A second study,⁹⁷ which enrolled 20 patients with recurrent ovarian cancer, assessed the clinical and immunologic effects of immunotherapy, before and concurrent with salvage chemotherapy. B43.13 treatment was well tolerated, and evidence for induction of immune responsiveness was seen, with CA-125-specific T-cell immunity seen in seven of 18 patients. Subsequent chemotherapy did not abrogate the induced immune responses, and interestingly, T-cell responses to autologous tumor increased in three patients and remained stable in the other two patients who were treated with combined chemoimmunotherapy.⁹⁸ Subjects who demonstrated T cell responses to CA-125 and/or autologous tumor showed a highly significant benefit in time to progression and survival, compared with nonresponders (p <lt> .01). The third study is a long-term follow-up of patients who underwent imaging with radiolabeled B43.13 monoclonal antibody.⁹⁹ Data from 44 patients with suspected or established recurrent disease who underwent initial antibody exposure between 1989 and 1996 were examined to assess survival times and immunologic correlates of therapeutic effects, demonstrating that B43.13 exposure was associated with prolonged survival times, which are correlated with induction of humoral responses, including human antimurine antibody responses, anti-B43.13-idiotype, and anti-CA-125.86 Together, these findings suggest that treatment with the B43.13 monoclonal antibody is safe and may have activity in the recurrent disease setting.

B43.13 has been studied as maintenance of clinical remission after front-line surgery and chemotherapy, with the aim of providing immune stimulation at a time when patient disease burden is minimal. Three protocols involving more than 500 patients have been conducted in this consolidation period. Ehlen and associates recently reported on the initial results of their 342-patient randomized, placebo-controlled, multicenter study of patients with stage III/IV disease and no evaluable disease after front-line treatment.⁹⁴ A 2-mg, 20-minute B43.13 infusion, or placebo control, was provided for three monthly treatments, and then was provided quarterly until relapse. More than half of the treated patients generated a vigorous immune response, and these immune responses are significantly associated with favorable clinical outcome. These immune responses are direct humoral antibody responses to the administered antibody, including responses to the constant regions of this murine antibody (IgG1 isotype), as well as to the variable antigen binding region of these antibodies. Neither of these responses is related to the CA-125 level. For the population as a whole, a significant difference was not seen between B43.13 treatment and placebo. In a randomized, placebo-controlled study of oregovomab for consolidation of clinical remission in patients with advanced ovarian cancer showed that consolidation therapy with oregovomab did not significantly improve time to relapse (TTR), overall.¹⁰⁰ In this study, 145 patients were treated with oregovomab (n = 73) or placebo (n = 72).  For the population overall, median TTR was not different between treatments (13.3 months for oregovomab and 10.3 months for placebo).  Immune responses were induced in most actively treated patients, and this was seen to be associated with prolonged TTR.  Quality of life was not adversely impacted by treatment.  Adverse events were reported with similar frequency in oregovomab and placebo groups.  Further evaluation identified a subpopulation of patients who had favorable prognostic indicators (successful front-line therapy -SFLT).  For this SFLT population, TTR was 24.0 months in the oregovomab group compared with 10.8 months for placebo, with a hazard ratio bordering on statistical significance, suggesting that oregovomab therapy may be of use in selected subsets of patients.  Based on these hypothesis generating data, two phase III placebo-controlled randomized trials in patients following initial clinical remission of B43.13 as maintenence therapy (IMPACT I and IMPACT II) were conducted. Unfortunately, no improvement in disease-progresion-free interval was demonstrated.

The use of B43.13 is now being studied in conjunction with front-line therapy. At first thought, it may seem that the immunosuppressive properties of front-line therapy may ablate the immunostimulatory effects of the antibody treatment. However, chemotherapy may beneficially alter the characteristics of an immune response, perhaps through selective elimination of immune suppressive lymphocyte populations.¹⁰¹ The results of phase 2 studies of B43.13 in patients with recurrent disease and concurrent chemoimmunotherapy suggest that study of front-line chemoimmunotherapy should be considered.¹⁰² Induction of CTL responses have been suggested with concomitant therapy and combination therapy did not alter neither the safety profile of B43.13 treatment, nor that of the chemotherapeutic agents tested. Therefore, study of antibody-based immunostimulation with B43.13 in conjunction with front-line treatment may be warranted.

Anti-idiotype anti-CA125 MonAb

Abagovomab is an anti-idiotype vaccine for patients with advanced ovarian carcinomas with the murine monoclonal anti-idiotype antibody Mab ACA125 (Anti-ID OC 125).  Phase I/II studies^{103, 104} demonstrated specific immune response induced by Abagovomab and tolerability of repeated doses. A multicenter study was performed to evaluate the safety and immunogenicity of repeated doses of Abagovomab in 119 patients with recurrent ovarian, fallopian tube, and peritoneal carcinomas. Patients were eligible for treatment if they had positive history of debulking surgery and previous platinum-based chemotherapy. No serious allergic reactions could be detected within a follow up period of up to 56 months. A specific anti-anti-idiotypic antibody (Ab3) was induced in 68.1% of patients; antiCA125 antibodies (Ab1’) were induced in 50.4% of patients and an antibodydependent cell-mediated cytotoxicity (ADCC) was observed in the serum of 26.9% of patients. In terms of the scheduled doses, the results of the study suggested that a median of 4 induction bi-weekly doses was sufficient to induce detectable Ab3 titers. The median survival of all patients was 19.4 months (range 0.5-56.1 months). Notably, patients with an Ab3 response (Ab3+) showed a significantly longer survival (median 23.4 months) compared with patients without an immune response (Ab3-) (median 4.9 months) (p<0.001). Ab3 responders received more Abagovomab doses compared with the Ab3 non-responders (mean 12.3+9.5 vs. 4.3+2.1), with the ‘responding’ patients being in the study for a longer time. In another Phase I trial of the monoclonal anti-idiotype antibody ACA125 in patients with epithelial ovarian, fallopian tube, or peritoneal cancer, 33 patients reached the fourth dose and underwent immunologic tests: in all patients, regardless of the treatment group, Ab3 titer was positive at the first evaluation point (week 10), with a further increase in titer in the subsequent follow-up samples collected up to 6 weeks after the last administered. There was no effect of the route of administration or of the administered dose on Ab3 or HAMA titer. The best overall response obtained in Ab3+ patients (n=33) was stable disease (12 patients, 36.4%). Another Phase I trial of ACA125 in patients with recurrent epithelial ovarian fallopian tube or peritoneal cancer was an open-label study that collected additional data on the Ab3 response following 6 doses versus 9 doses. Ab3 resulted positive in all evaluable patients (100%), with a further increase in the Ab3 titer in the treatment arm exposed to 9 doses of Abagovomab. No severe drug-related toxicity was recorded in either treatment group. The most frequent adverse events (mainly grade 1-2) were: anemia, leucopenia, neutropenia, injection site reactions, fatigue, and nausea. Overall, the results of this second Phase I/II study confirmed the good tolerability of repeated administrations of Abagovomab and provided additional supporting evidence on the induction of an active immune response using the 2 mg dose and SC route of administration selected for the Phase III trial.  Therefore, Abagovomab proved to be effective in generating a specific immune response as expressed as Ab3 against CA125 positive cells. Antibody responses were associated with increased survival. Treatment with Abagovomab was well tolerated independently from the duration of treatment and route of administration (SC vs IM). These studies are the basis for a large ongoing randomized, double blind, multicenter trial in which Abagovomab is being compared with placebo as maintenance therapy in stage III-IV ovarian cancer patients after completion of successful frontline therapy with surgery and standard platinum-taxane chemotherapy.

Other MonAbs

Several other monoclonal antibody-based products, directed against tumor-specific antigens, are being explored in preclinical and clinical studies. There is great potential for the development of experimental treatment strategies for the treatment of ovarian cancer with multiple antibodies specific to different tumor antigens.

Anti-EpCAM × Anti-CD3 Trifunctional Antibody EpCAM

This MonAb targets EpCAM, a 'tumor-associated' antigen^{105, 106} and investigations have led to the development of EpCAM-specific antibodies to arrest tumor growth. The investigational compound, catumaxomab, is a hybrid-hybridoma derived, trifunctional bi-specific antibody with anti-EpCAM × anti-CD3 specificity. Catumaxomab binds with its two different binding arms specifically to the CD3 antigen on T-lymphocytes and to the EpCAM antigen, expressed on the epithelial tumor cells. As a heterologous antibody, configured of the two potent subclasses of mouse IgG2a and rat IgG2b, the Fc region of catumaxomab simultaneously binds to Fcγ-receptor type I and III positive accessory cells (e.g. macrophages, dendritic cells, and natural killer cells) of the immune system. Therefore, catumaxomab is constructed to work in synergy with the immune system, and as a result of this concerted attack of different immune cells, tumor cells are expected to be eliminated with high efficiency. In the 'tri-cell complex', an important 'crosstalk' between T cells and accessory cells can occur, which includes co-stimulatory signals necessary for a physiological T cell activation cascade.¹⁰⁷  The simultaneous activation of different immune cells at the tumor site results in efficient killing of tumor cells by several complementary mechanisms (e.g. release of cytokines, perforin-mediated lysis, and phagocytosis).¹⁰⁸ The subsequent antibody-mediated phagocytosis of tumor cells by accessory cells (macrophages and dendritic cells) is believed to result in the processing of tumor antigens and presentation on the surface of these cells. Since this can result in polyclonal humoral and cellular immune responses, a T-cell response even against unknown, tumor-associated peptides may be induced.¹⁰⁹ 

In a Phase I/II study of i.p. catumaxomab (removab®, anti-EpCAM x anti-CD3) conducted in Europe, the tolerability and efficacy in patients with ascites due to ovarian carcinoma was analyzed.¹¹⁰ Twenty-three patients with malignant ascites due to ovarian cancer received a regimen consisting of up to 5 increasing intraperitoneal doses of catumaxomab infusions (from 5 mcg to 200 mcg) within 10–13 days. The patients had an improvement of ascites-related symptoms (dyspnea, anorexia, and insomnia) due to a significant reduction in ascites. The mean ascites flow rate decreased from 156 mL/h at baseline (n=20) to 45 mL/h one day after the 4th infusion (n=18). By the end of observation (day 37) only 1 patient had required paracentesis. The most common adverse events were fever, nausea, vomiting, and abdominal pain (cytokine release syndrome). Development of human anti-mouse and anti-rat antibodies (HAMA and HARA) were observed in all but one patient tested, but did not elicit additional adverse events. In a Phase II/III trial in malignant ascites (IP-REM-AC-01), 129 patients with ascites due to ovarian cancer were randomized 2:1.¹¹¹ Eighty patients received paracentesis plus catumaxomab and 44 received only paracentesis as needed for the relief of ascites symptoms. Compared to the control, this study demonstrated that treatment with catumaxomab administered as a sequence of 4 IP infusions of 10, 20, 50 and 150 mcgs resulted in a statistically significant superiority in puncture-free survival – a more than 4-fold prolongation: 52 versus 11 days (median; p>0.0001) and in puncture-free time – a more than 6-fold prolongation: 71 versus 11 days (median; p>0.0001). The second European study (AGO-OVAR 2.10), a randomized, 2-dose-level, open-label, phase IIa study of catumaxomab to select the better dose level in platinum-refractory epithelial ovarian cancer patients. Women with platinum-refractory (progressing during or less than 6 mos. after the last platinum containing regimen) epithelial ovarian cancer and measurable recurrent disease were randomized to receive either 10 -10 -10 - 10 mcgs or 10-20-50-100 mcgs of catumaxomab over 6 hours i.p on days 0, 3, 7 and 10. Forty-five patients were entered (22 high dose (HD)-arm, 23 low dose (LD)-arm). These data provide evidence that catumaxomab can reduce the need for paracenteses in patients with malignant ascites.  Based on these data, there are two ongoing trials being conducted in the USA, one in women with malignant ascites and another as maintenance therapy in women in clinical remission after their inital platinum-taxane-based chemotherapy.

Anti-HMFG1

Radioimmunotherapy with antipolymorphic epithelial mucin monoclonal antibody HMFG1 is a murine monoclonal antibody (IgG1 isotype) that is specific for an epitope of polymorphic epithelial mucin (PEM), which is expressed on more than 90% of epithelial ovarian cancers and many other carcinomas.¹¹² Phase 1/2 trials of HMFG1 monoclonal antibody radiolabeled with yttrium (Yt) 90 have shown that this agent is well tolerated when administered intraperitoneally.¹¹³ The survival of patients with microscopic residual disease after induction chemotherapy, who received radiolabeled HMFG1, appears to be prolonged compared with historical controls.¹¹⁴ Antisoma has conducted a multicenter, randomized, prospective phase 3 trial of ⁹⁰yttrium-labeled HMFG1 in women with no evidence of disease used as consolidation therapy at the completion of their primary chemotherapy. The trial showed no impact on survival in the patients treated with the monoclonal antibody.

IgE-based monoclonal antibodies

Recent pre-clinical work suggests that the efficacy of immunotherapy of solid tumors, in particular ovarian carcinoma, may be improved by the use of IgE Abs in place of the conventional IgG.  In an ovarian cancer xenograft-bearing mice, treatment with an IgE antibody (MOv18 IgE) directed to an ovarian cancer associated antigen (folate binding protein), in combination with human PBMC, was seen to result in improved survival, when compared to IgG-based monoclonal antibodies.^{115, 116} Monocytes were seen to be essential in mediating this enhanced survival of tumor-bearing mice treated with MOv18 IgE and human PBMC, mediating tumor cell killing by cytotoxicity and phagocytosis.  The IgE Fc receptors FcepsilonRI and CD23 (low affinity IgE Fc receptor) were seeen to be involved in these processes. These preliminary results suggest that IgE-based monoclonal antibodies merit further study for the treatment of ovarian cancer.

Intraperitoneal Cytokine Therapy

The use of recombinant DNA technology has made possible the production of large quantities of cytokines. Several of these agents have been examined in phase 1 and 2 clinical trials, including recombinant human IFN-α, IFN-γ, TNF-α, and IL-2.

INTERFERON-α

IFN-α and IFN-γ have been shown to be active against ovarian cancer both in vitro and in vivo.^{117, 118, 119, 120} IFN-α, which is capable of augmenting the cytotoxicity of autologous PBMC to human ovarian carcinoma cells,¹²¹ is well-tolerated locally but has significant systemic side effects, making it an attractive candidate for intraperitoneal immunotherapy. The earliest clinical trials of IFN-α administered intraperitoneally were conducted in women with persistent ovarian cancer at second-look laparotomy.^{122, 123, 124} Intraperitoneal IFN-α can augment NK cytotoxicity, and this was associated with tumor rejection.^{125, 126} However, augmentation of NK activity was not invariably associated with clinical response. The dominant mechanism that is responsible for killing tumor cells in the peritoneal cavity may involve the direct effect of IFN on cancer cells, as is seen with cytotoxic chemotherapeutic agents, rather than the enhancement of antitumor immune responsiveness.⁶⁰ In several confirmatory second-line trials in women with minimal residual disease, the surgically documented response rates to IFN-α were 30–50%.^{127, 128, 129, 130} This outcome was shown to be similar in women treated with intraperitoneal IFN-γ as second-line therapy,^{131, 132, 133, 134} with surgically documented responses in the range of 30–45% in patients with minimal residual disease. Although there is in vitro evidence that IFN-γ can increase the sensitivity of cancer cells to cisplatin,¹¹⁸ second-line intraperitoneal IFN-α in combination with cisplatin does not appear to offer any advantage to interferon alone in these patients.^{127, 128} Adverse effects, such as lethargy, fatigue, and flu-like symptoms, are common with the administration of the interferons.

INTERFERON-γ

Despite these results for second-line administration of interferon in which response rates comparable with cytotoxic chemotherapeutic agents were reported, the use of interferon in ovarian cancer has been limited to clinical trials. However, in a randomized phase 3 trial, Windbichler and colleagues¹³⁵ demonstrated that the use of subcutaneous IFN-γ in women receiving first-line platinum-based chemotherapy in ovarian cancer was well-tolerated. As one would expect with the administration of an interferon, a higher proportion of patients had fever and flu-like syndrome. However, there was no appreciable increase in gastrointestinal, neurological, or hematological toxic effects. Importantly, the results demonstrated a higher response rate and longer disease progression-free survival in women receiving IFN-γ plus chemotherapy than women who were treated with chemotherapy alone. The improved progression-free survival was seen in both optimally resected patients (as defined with residual maximum tumor dimension of less than 2 centimeters) and those with bulky residual disease (greater than 2 centimeters). No statistically significant improvement in overall survival was observed. In those patients, the benefit only appeared to exist in those who had minimal residual disease (i.e. patients whose residual disease was no greater than 5 mm in maximum diameter). In the randomized, prospective trial, the IFN-γ was administered as a primary treatment, presumably before the development of drug resistance. Although this represents the first trial of its kind, one of the limitations of the trial is that the chemotherapy used was a combination of cisplatin and cyclophosphamide. These drugs have now been supplanted by drugs that appear to be more efficacious and less toxic (e.g. carboplatin and paclitaxel).

Although a mechanism of action of the effects mediated by IFN-γ in patients with ovarian cancer is unknown, it has been speculated that this cytokine may inhibit the expression of dysregulated oncogenes (e.g. HER-2/neu), improving the response of cisplatin-resistant cells. Overall, evidence for this is mixed.^{136, 137, 138} Cytokine administration intraperitoneally appears to be necessary to augment local cytotoxic effector cells, which are located where most of the residual cancer occurs. Presumably, direct contact with the cytokine through regional administration is necessary to induce cytostatic and/or cytotoxic effects in the peritoneal cavity. However, in the randomized trial by Windbichler and associates, systemic administration conferred the survival benefit. Most likely, one should attribute this to the stimulatory effects of IFN-γ on cells of the immune system, including stimulation of NK cells and macrophages, both of which are known to have the ability to exert antitumor effects.¹²⁴ The stimulation of macrophages and other cytotoxic effector cells may have a nonspecific immunomodulatory effect in these patients, which favors responsiveness to chemotherapy or exert direct antitumor effects.

Intriguingly, the interferons, including IFN-γ, have been shown to have not only an antiproliferative effect but also antiangiogenic effects. Theoretically, these cytokines have their highest probability of success in women who are also subjected to chemotherapy for their disease. The combination of agents, such as IFN-γ, thalidomide or anti-VEGF, and cytotoxic therapy presents a new opportunity for the development of innovative pharmacologic agents in solid tumors.

TUMOR NECROSIS FACTOR

In various preclinical studies, TNF-α showed significant antineoplastic activity to a variety of malignant cell lines.^{139, 140, 141} However, in phase 1 trials of TNF-α delivered systemically, there was limited clinical activity with considerable systemic toxicity, especially fevers, rigors, and hypotension.^{142, 143, 144, 145} As with other cytokines and biologic response modifiers, it had been hoped that intraperitoneal TNF would produce an increased antitumor response and fewer systemic side effects. In a phase 1 trial, recombinant human TNF-α was administered safely when given intraperitoneally, with a marked pharmacokinetic advantage for intraperitoneal administration compared with systemic administration.¹⁴⁶ Although TNF-α was not detectable in plasma, patients experienced mild emesis, temperature elevations, and chills, even with intraperitoneal TNF-α doses as low as 10 mg/m². Only one patient had a hypotensive episode (blood pressure 80/50 mmHg), but treatment-related abdominal pain was common. No clinical responses were observed in this phase 1 trial.¹⁴⁶

In another study, intraperitoneal recombinant human TNF-α was used to control the formation of malignant ascites, and was shown to have a potential role in the management of this process.¹⁴⁷ Thirty-two patients with symptomatic malignant ascites were treated with a weekly infusion of TNF-α (80 mg/m²). Twenty patients had ovarian cancer; the remaining 12 had various nonovarian malignancies. Of the 31 evaluable patients, 17 (55%) experienced complete resolution of ascites 30 days after the initiation of therapy, and 14 patients exhibited partial control of malignant ascites. Interestingly, only one patient had a relapse during the follow-up period of approximately 8 months. The most significant side effects were fever, chills, abdominal pain, and emesis.¹⁴⁷ On the basis of this report, this treatment may be an option for patients with malignant ascites when alternative therapies have been ineffective. Further studies of intraperitoneal TNF-α for the control of malignant ascites would be needed to confirm this.

The potential of TNF-α to augment the antitumor effect of a cytotoxic chemotherapeutic agent (cisplatin) offers another potential immunotherapeutic strategy. Even low doses of TNF-α can augment the antitumor properties of drugs such as cisplatin, doxorubicin, and cyclophosphamide in vitro.¹⁴³ Therefore, the use of low-dose TNF-α therapy, administered in conjunction with cytotoxic chemotherapy, could offer a therapeutic advantage and minimize the toxicity of TNF-α immunotherapy.

INTERLEUKIN-2

IL-2 also has been used for systemic experimental immunotherapy. In an early study, recombinant human IL-2, administered intravenously to patients with progressive melanoma, renal, colon or ovarian cancer, was seen to induce lymphocytosis, increased numbers of cells expressing the IL-2 receptor, detectable circulating IL-2-activated killer (LAK) cells, and augmented NK cytotoxicity.¹⁴⁸ The intraperitoneal administration of IL-2 has been the subject of several studies.^{149, 150} The major rationale for developing intraperitoneal-based IL-2 treatment strategies is the observation that the in vitro antitumor activity of IL-2 is enhanced with increasing drug concentrations.^{148, 15} When the intraperitoneal administration of IL-2 was tested, a major pharmacokinetic advantage for intracavity IL-2 treatment was observed, as expected.^{149, 150} In a phase 1/2 study of intraperitoneal IL-2 in refractory ovarian cancer, 2 of 13 patients had a complete response.¹⁵ Systemic toxicities were mild and included fever, fatigue, myalgias, diarrhea, emesis, and abdominal pain. In a phase 1 trial, administration of IL-2 intraperitoneally resulted in a 100-fold increase in peritoneal cavity exposure, compared with systemic administration.¹⁵¹

Induction of Local Cytokine Production by Paclitaxel

Paclitaxel is an important drug in the treatment of ovarian cancer. Although paclitaxel is known to have biologic effects that can inhibit tumor cell growth directly, such as stabilizing microtubules and blocking cell mitosis, it has been seen that the effectiveness of this drug in ovarian cancer exceeds that of other antimitotic chemotherapeutic agents. This has been taken to suggest that paclitaxel may have additional, perhaps indirect, modes of action. Proinflammatory cytokine gene expression was examined in a series of cell lines and tumor explants from human ovarian cancer tissue; paclitaxel was seen to induce the secretion of IL-8, but not IL-6 or IL-1.¹⁵² These results suggest that the expression of this chemokine in vivo, induced by paclitaxel, may enhance local antitumor host immune responses, by inducing the transcription of cytokine and/or growth factor genes in ovarian cancer cells. Subsequent studies led to the identification of paclitaxel-responsive regulatory elements in the IL-8 promoter; the AP-1 and NF-kB binding sites in the IL-8 promoter were seen to be required for the induction of IL-8 gene expression, whereas a C/EBP site required for IL-8 promoter activation in other cell types was not involved.¹⁵³ Watson and associates¹⁵⁴ identified the region of paclitaxel responsible for the induction of IL-8 in ovarian cancer cells and found a direct correlation between the ability to induce IL-8 production and cytotoxicity-analogues that most markedly up-regulated IL-8 expression proved to be the most cytotoxic. However, it is not known whether any of the in vivo therapeutic effects of paclitaxel treatment are caused by the induction of inflammatory cytokine production and/or to the enhancement of antitumor immune responses by paclitaxel-induced cytokines.

Adoptive Immunotherapy

Experimental adoptive immunotherapy involving the ex vivo enhancement of antitumor immune cell responses provided an immune system-based approach for antitumor therapy.^{123, 155, 156, 157, 158, 159, 160, 161} In particular, adoptive immunotherapy involving the infusion of autologous LAK cells has been studied extensively, and has produced tumor regression in various animal and human tumors.¹⁵⁵ Exposure of PBMC to IL-2 leads to the generation of cytotoxic effect cells (LAK cells), which are cytotoxic for a variety of tumor cells, including tumor cells that are resistant to NK-cell-meditated or T-lymphocyte-mediated lysis. Experimental treatment of human subjects with autologous, ex vivo-generated LAK cells and IL-2 has yielded some responses.^{66, 155, 156, 157, 158, 159} However, the overall response rate to LAK treatment is low, and this type of adoptive immunotherapy can result in significant morbidity. Therefore, it is impractical in most medical settings.¹²³ Other, more efficient and practical applications of adoptive immunotherapy are under development. One approach involves the ex vivo generation of immune effector cells from TIL, by their activation and expansion in vitro by exposure to IL-2, and administration of ex vivo-activated TIL concurrently with IL-2.^{160, 161}

Systemic IL-2 administration in various forms of human cancer, with or without LAK cells or TIL, has been the subject of intense investigation.^{123, 157, 158, 159, 160, 161} In vivo preclinical studies have shown that treatment with LAK cells plus IL-2 can prolong survival significantly.^{162, 163, 164, 165} In a phase 1 trial, IL-2 and LAK cells were administered intraperitoneally after systemic administration of IL-2.^{151, 166} While administration of IL-2 intraperitoneally resulted in a 100-fold increase in peritoneal cavity exposure, compared with systemic administration, and LAK activity was detectable in the peritoneal cavity during the period of treatment, considerable toxicity was seen, including fever, chills, emesis, hypotension, abdominal pain, fluid retention, bone marrow suppression, and liver function abnormalities. Infection also was common. Several patients had extensive peritoneal cavity fibrosis, possibly resulting from the release by IL-2-activated cells of various growth factors and cytokines capable of stimulating collagen synthesis by fibroblasts.

Tumor-infiltrating lymphocyte immunotherapy has been studied in ovarian cancer. In seven patients with advanced or recurrent epithelial ovarian cancer treated with the adoptive transfer of TIL after a single dose of cyclophosphamide, one complete response and four partial responses were seen.¹⁶⁷ Ten additional patients were treated with a cisplatin-containing chemotherapeutic regimen as well as TIL. Seven complete responses and two partial responses were seen. Four of the seven patients who had a complete response had no recurrence after 15 months of follow-up. These results suggest that TIL-based immunotherapy for ovarian cancer may achieve complete response rates without IL-2 administration. Thus, this use of TIL may prove to be a more appropriate form of adoptive immunotherapy for ovarian cancer than LAK-based regimens. Further studies are needed to examine TIL and other forms of adoptive immunotherapy in ovarian cancer.

Intraperitoneal Gene Therapy

In recent years, gene therapy-based approaches to the treatment of gynecologic malignancies have gained significant attention, as more has been learned about the molecular basis of these cancers, allowing for potential interventions at the molecular level for therapeutic purposes. Some potential gene therapy approaches involve the expression of cytokine genes, or other genes associated with the enhancement of antitumor immune responses (genetic immunopotentiation), while other approaches aim to target dysfunctional oncogenes or tumor suppressor genes (mutation compensation), or to deliver molecular chemotherapy.^{83, 168, 169, 170, 171, 172, 173} However, significant problems will need to be solved before gene therapy for cancer can be effective and practical, including limitations in the ability to deliver therapeutic genes at a sufficiently high level, and with specificity, into tumor cells. Certainly, much new information on the potential of gene therapy for gynecologic cancers will result from ongoing studies.

Various components of the immune system can exert effective antitumor responses, including direct cytotoxicity directed against tumor cells as well as the production of cytotoxic or immune-enhancing cytokines. The dysregulated expression of various genes, including overexpression of oncogenes, mutation or loss of anti-oncogenes, and overproduction of growth factor and cytokines, may play an important role in the development and progression of ovarian cancer. Modification of the host environment and host antitumor immune responses also may contribute to the development and growth of ovarian malignancies. Understanding the specific mechanisms that are involved in the development and growth of ovarian cancer will provide opportunities for the rational design of effective antitumor treatment, including immunotherapy.

Several forms of experimental immunotherapy for ovarian cancer have been examined, including biologic response modifiers and cytokines. Preliminary studies show that immune enhancement and antitumor effects that lead to tumor rejection are most likely to occur when relatively high concentrations of cytokines are brought into direct contact with tumors and when the tumor burden is minimal, such as in an adjuvant setting or when treatment is combined with cytotoxic chemotherapy. Therefore, intraperitoneal immunotherapy, in combination with chemotherapy, may provide effective new forms of treatment for ovarian cancer. Great advances have been made, in recent years, in the generation of appropriate forms of monoclonal antibodies, directed to cell-surface molecules expressed on tumor cells. This has resulted in a renaissance of monoclonal antibody-based therapies for several forms of human cancer. Adoptive immunotherapy also has created new opportunities for the use of immunotherapy in patients with ovarian cancer. Of great long-term importance, gene therapy holds the promise of novel and targeted therapeutic approaches based on an increasing understanding of the molecular and cellular biology of cancer and anticancer immune responses, although significant practical problems will need to be solved before gene therapy for cancer can be effective and widespread.

Immunologic therapy of ovarian cancer currently remains experimental. Immunological therapies appear to augment what can be accomplished with optimal chemotherapy and thus could provide an incremental improvement in patient management over current standards when results of currently ongoing studies establish definitive treatments as a standard of care. Clearly, we are in a period of rapid scientific and technological progress, which is resulting in the development and testing of various cytokine-based and immune-based therapies for gynecologic cancers.