- Email: [email protected]
GENE THERAPY FOR BREAST CANCER
~~
0889-8588/98 $8.00
GENE THERAPY
+ .OO
GENE THERAPY FOR BREAST CANCER Heike K. E. Boxhorn, PhD, and Stephen L. Eck, MD, PhD
Surgery, chemotherapy, hormonal therapy, and radiation therapy are the main forms of treatment of breast cancer. These approaches reduce the risk of death and have been curative in the majority of patients, but many tumors will recur as metastatic lesions. Gene therapy offers a potentially useful approach for the treatment of breast cancer inasmuch as a variety of molecular processes can be introduced by gene transfer, which can in principle arrest tumor growth.46Several major obstacles, however, need to be overcome for these approaches to be successful. First, our understanding of the molecular processes that lead to breast cancer is incomplete. Thus, therapies directed at the molecular events known to promote tumor growth may be limited by an incomplete understanding of the underlying molecular and cellular processes. Second, gene delivery into breast cancer tissue is made difficult by the slow growth rate of most breast cancers, which may make it difficult to use some vectors such as retroviruses. Finally, for those gene therapies that are not immediately cytotoxic, the often indolent nature of breast cancer cell growth may require a longer term of gene expression than would be otherwise required. Despite these apparent problems, there has been significant progress in developing gene therapy approaches for breast cancer. The approaches can be divided into two general strategies: (1)approaches that alter the metabolic or signaling pathways within the breast cancer cell; and (2) approaches designed to enhance the immune This work was supported in part by grants UO1 CA65805 from the National Institutes of Health (SLE) and by DAMD17-96-1-6287from the Department of Defense (HKEB).
From the Division of Hematology and Oncology, Department of Medicine, The University of Pennsylvania Medical Center, Philadelphia, Pennsylvania
HEMATOLOGY/ONCOLOGY CLINICS OF NORTH AMERICA VOLUME 12 * NUMBER 3 * JUNE 1998
665
666
BOXHORN & ECK
response to the tumor cells. The latter group has achieved the most attention because immunologic approaches avoid many of the problems identified previously. These approaches are not unique to breast cancer and may be proven more successful in other tumors first. THERAPEUTIC GENES THAT ALTER THE METABOLIC OR SIGNALING PATHWAYS IN BREAST CANCER CELLS All breast cancer cells contain molecular genetic abnormalities that contribute to tumor cell growth. Many of these genetic changes occur sporadically or at such low frequency that they are not useful targets for designing gene-based therapies. A few molecular alterations occur with sufficient frequency that they could become useful targets in a significant portion of breast cancer patients. Of these HER2/neu, p53, and c-myc have received the most attention as therapeutic targets because of their relatively high frequency of alteration in breast cancer. The HER2/neu oncogene encoding a transmembrane receptor protein is over-expressed in 30% of breast cancer patients.27The overexpression of HER2/neu is associated with a worse prognosis27by contributing to increased metastases and decreased sensitivity to chemotherapy. Several therapeutic approaches are being developed to genetically target HER2/neu. One approach is to synthesize an antibody to HER2/neu within the tumor cell itself. Curie1 and have used an adenovirus vector that was modified to encode a single-chain monoclonal antibody that would bind HER2/neu. Human breast cancers that over-expressed HER2/neu were growth-arrested in tissue culture experiments, whereas breast cancers cells that expressed little HER2/neu were much less affected by this treatment. Another approach to down-regulate transmembrane growth factor receptors such as HER2/neu is to express a transdominant-negative inhibitor. Such inhibitors function by dimerizing with the growth factor receptor and thereby prevent the growth factor receptor from dimerizing with itself, as it normally d0es.4~Alternatively, rather than inhibiting the HER2/neu protein function after it is made, it is possible to block the synthesis of this protein altogether. The adenovirus E1A transcription factor (necessary for adenovirus replication) has been shown to block HER2/neu expression by inhibiting the expression of the HER2/neu gene in breast cancer cells infected with adeno~irus.~ This in turn sensitized the cells to the cytotoxic effects of paclitaxel. This both confirms the previous association of the HER2/neu over-expression with chemoresistance and provides for the possibility that novel therapies can be based on the combined use of gene therapy and chemotherapy. Cell cycle regulation is altered in many tumors including breast cancer. Mutations in the p53 gene are the most common molecular abnormality in cancer. Restoring normal p53 function in breast cancer cells has been shown to decrease tumorigenicity by leading to cell cycle
GENE THERAPY FOR BREAST CANCER
667
arrest and induction of a p o p t ~ s i s .39,~ 49 ~ ,Adenoviral vectors expressing normal p53 efficiently transduced breast cancer cells in culture and animal models. Scientists at the Schering-Plough Research Institute recently reported that intravenously administered recombinant adenovirus expressing p53 substantially reduced the number of lung metastases and total tumor burden in a murine model of metastatic breast cancer.4oIn this model the immune response to the virus and the tumor could not be assessed, because immunodeficient mice were used as hosts of the human breast cancer xenografts. This may both underestimate the toxicity induced by anti-adenovirus immune responses and obscure any immune response to the tumor. Nonetheless, this experiment convincingly demonstrates that this approach has potential for treating metastatic disease.4oFrom this study, as well as many previous studies, it is apparent that intravenous dosing of adenovirus vectors (the current vector of choice for p53) induces significant liver injury. Recent work by Wilson's group45indicates that the newer-generation adenovirus vectors that have deletions of the E4 adenovirus genes (in addition to E l gene deletion) have much less hepatotoxicity than do the adenovirus vector employed by the Schering-Plough group. Application of these new adenovirus vectors to p53-based therapies could likely offer substantial improvements to therapeutic strategies. In addition to the direct effects on cell cycle, overexpression of the normal p53 gene has been shown to restore sensitivity to doxorubicin (Adriamy~in).~~ The expression of normal p53 within the tumor cells also alters the tumor microenvironment through a paracrine effect. It has been shown that over-expression of p53 leads to an overall reduction in tumor blood vessels.6zThis effectively extends the influence of the gene transfer far beyond the confines of the tumor cell that expresses the therapeutic gene. The enforced expression of p53 has been shown to up-regulate thrombospondin synthesis, which in turn is a negative regulator of tumor angiogenesis.' The contribution of this antiangiogenic effect to the overall therapy may turn out to be important, because with the current technology it is not possible to effect gene transfer in all tumor cells. Systemic antitumor effects may be achieved with tumor suppressor gene therapy by co-administering a cytokine gene. Graham's group4 compared the effectiveness of p53 and IL-2 gene therapies either alone or in combination using adenovirus vectors. They found that a single injection of a combination of the two vectors resulted in tumor regressions in 65% of the treated mice. In this immune competent transgenic mouse mammary adenocarcinoma model, half of'the treated mice remained tumor-free. These mice developed systemic immunity to the tumors as demonstrated by rechallenge experiments and in vitro measures of tumor-specific cytolytic T-lymphocyte activity. Adenovirus delivery of p53 or IL-2 alone produced only transient delays in tumor growth.4 The heterogeneous nature of genetic mutations in breast cancers has encouraged the development of strategies that are less dependent on the genetic alterations of the tumor cell. Introduction of prodrug activating
668
BOXHORN & ECK
genes has been extensively studied in brain tumors, and it may be applicable to breast cancer therapy. In this strategy the tumor cell is genetically modified to express an enzyme capable of activating an otherwise inactive prodrug. In the case of breast cancer, the cytosine deaminase gene has been used for this purpose. Cytosine deaminase (a bacterial enzyme) can activate 5-fluorocytidine to 5-fluorouracil, a chemotherapy agent with known activity and widespread use in breast cancer. The advantage of the gene therapy approach is that sustained high levels of the drug can be achieved within the tumor, whereas the systemic exposure to the drug is low. Seth and colleagues49have demonstrated that an adenovirus vector carrying the cytosine deaminase gene can render breast cancer cells 1000 fold more sensitive to 5-flU01-0cytidine than are mock transfected controls. Furthermore, only 10% of the cells needed to express the cytosine deaminase gene in order to achieve arrest of tumor cell growth. This bystander effect arises from the high rate of diffusion of 5-fluorouracil out of the transfected cells into nearby cells that lack the cytosine deaminase gene.35Application of this approach to human disease will require the development of genetic vectors capable of reaching widespread metastases. GENETIC ENHANCEMENT OF THE IMMUNE RESPONSE TO BREAST CANCER
Immunotherapy offers a promising alternative to conventional therapy because the systemic nature of the immune response can access disseminated disease, and its specificity may limit undesirable side effects. A variety of immune cells, including natural killer (NK) cells, lymphokine activated killers (LAK), macrophages, B cells, and T cells mediate antitumor immune effects. Among these cells, T cells possess unique characteristics that render them attractive targets for the development of immunotherapy. T cells are highly specific in recognizing antigenic peptides. This is ideal for an anticancer reagent that is expected to recognize and destroy tumor cells, but not normal cells. Once activated, T cells differentiate into long-lived memory cells that recirculate continuously. This enables T cells to detect and destroy metastases that arise at distant sites at an early stage. Although new monoclonal antibody therapies are entering into clinical practice, the enhancement of T cell functions has become the major goal of immunotherapy against breast cancer. Prior experience with immunotherapies highlights both the potential and the problems associated with gene transfer-based immunotherapy. Tumor infiltrating T cells (TILs) can be expanded ex vivo prior to readministration of the enlarged population of T cells. This form of passive (and transient) immunization in combination with IL-2 immunotherapy has been applied to patients with a variety of cancers including breast cancer.54Rather than readministrating tumor-reactive T cells, active immunization can be used to enhance T-cell functions and potentially provide a longer period of therapy. Different vaccination strategies
GENE THERAPY FOR BREAST CANCER
669
using DNA (encoding for a tumor antigen), purified protein, peptide, whole tumor cells, or tumor lysates have been developed for the treatment of breast cancer and other tumors.17 In these early studies there was no survival advantage for patients treated. Genetically modified tumor cells have been used to induce a cellular antitumor immune The rationale of using whole tumor cells as a vaccine is that they express the entire repertoire of antigens that would not be present when a defined tumor antigen was used. Thus, the knowledge of specific antigens is not required for the development of a vaccination strategy. Furthermore, recent improvements in our understanding of activation of immune functions may allow one to improve T-cell recognition and activation by genetically modifying tumor cells. For example, tumor cells engineered to express IL-2 were rejected in animal studies and induced protective immunityI3 under conditions in which the unmodified tumor cells grew progressively. A growing number of cytokines and related soluble immune stimulatory factors have been systemically administered to bolster the immune response. The principal problems encountered in the cytokine clinical trials have arisen from the systemic toxicity. Gene transfer strategies may overcome this by the local expression of cytokines within the tumor microenvironment. One approach is to isolate human skin cells (fibroblasts) and genetically alter them so that they secrete one of a number of cytokines.'* The gene-modified cells can then be introduced with the tumor cells as an irradiated vaccine.'* This approach allows the technique of gene delivery to be standardized in fibroblasts and is more reproducible than gene modification of tumor cells. Nonetheless, isolation and ex vivo modification of fibroblasts are cumbersome and expensive procedures and not amenable to use in most clinical practices. In vivo gene transfer overcomes this problem and can be readily implemented using adenovirus vectors.59For example, an adenovirus expressing tumor necrosis factor-alpha (TNFa) has been used in murine models of breast cancer to induce complete regression of spontaneous mammary tumors.36Although the vector was injected into the tumor, systemic levels of TNFa were achieved and resulted in significant toxicity. This was overcome, without loss of antitumor efficacy, by altering the TNFa so that it would remain bound to the tumor cells and not be released into the circulation. Not only was tumor regression observed, but these animals developed tumor specific immunity? Similarly, a recombinant adenovirus expressing a human interferon consensus gene has been shown to induce regressions of human breast cancers grown in immunodeficient mice.63This effect is likely the result of several independent processes, including the production of interferon and direct tumor cell lysis by the adenovirus. The oncolytic effects of viruses are well known and have been studied clinically in small patient studies.47Thus, the use of viral oncolysis to release tumor antigens and immune stimulation by local production of a cytokine may warrant further study in clinical trials. Stewart and colleagues5*may achieve such a result in their recently described plans for a phase 1/11 study using a recombinant adenovirus
670
BOXHORN & ECK
expressing IL-2. A replication-defective adenovirus carrying the IL-2 gene will be injected into metastatic breast cancer tumor sites. This is anticipated to result in very high intratumor concentrations of IL-2 and lesser systemic levels. The adenovirus vector itself will undoubtedly elicit an immune response and result in local destruction of tumor cells as a result of pre-existing antiadenovirus immunity. Prior to their destruction, however, the tumor cells will make IL-2, which is intended to augment the T cell-mediated antitumor response and result in a systemic immune response to the breast cancer. The IL-2-secreting cells will be eliminated within a few days, so that the IL-2 exposure will be brief, with high levels of IL-2 confined to the tumor site. T-cell activation has been shown to require two distinct signals: an 23, 42 The antigenantigen-specific and an antigen-nonspecific specific signal (signal 1) is provided by the interaction of the T-cell receptor (TCR) with peptides presented on the cell surface in the major histocompatability complex proteins (MHC). A secondary signal (signal 2) is generated by a co-stimulatory molecule (e.g., B7-l), which interacts with its cognate ligand (CD28) on T cells. Normally both signals are provided by antigen-presenting cells that process tumor antigens acquired from the local environment. The interaction of B7-1 with CD28 results in enhanced IL-2 production by T cells and proliferation and generation of cytotoxic T cells.29,48 In the absence of signal 1, T cells do not proliferate but remain capable to do so upon antigen presentation. The absence of a co-stimulatory signal (as occurs when tumor cells In this stage the T cell is present antigen to T cells) results in anergy.10,50 unresponsive to specific antigens and unable to respond to a subsequent optimal stimulation with signal 1and 2.15Thus, co-stimulatory molecules alone do not initiate, but rather enable the generation and amplification 58 B7-1 is not of antigen-specific T-cell responses and effector functions.23* expressed by most tumor cells, including breast cancer cells.I5Based on this concept, gene transfer of the B7-1 gene to breast cancer cells can be used to drive the expression of B7-1 on the tumor cells that already express a variety of tumor antigens in MHC. Expression of B7-1 on a breast cancer cell vaccine has been shown to induce protective immunity in animal studies.l' This effect could be further enhanced by the coadministration of IL-12." These responses were shown to be mediated by T cells, indicating the ability of modified tumor cells to induce a cellular immune response. This strategy is currently being tested in a phase I study of patients with refractory breast cancer and melanoma. Many investigators have shown that the expression of B7-1 on murine tumors results in tumor rejection8,*l, 16, 55 and the induction of protective immunity8,11, 34, 37, 38, 55; however, the effectiveness of B7-mediated antitumor immunity is highly dependent on the tumor model chosen and the immunogenicity of the tumor. Tumor cells are generally termed immunogenic when they express antigens that result in the rejection of tumor cells by syngeneic animals previously immunized with the irradiated parental tumor cells.9*32 Accordingly, non-immunogenic tumors are not rejected when similarly tested. Enhanced protective
GENE THERAPY FOR BREAST CANCER
671
antitumor immunity and increased specific cytotoxic T-lymphocyte (CTL) activity was achieved by B7-1 transduced immunogenic tumors but not by B7-1 transduced non-immunogenic tumor^.^ In immunogenic tumor systems, 87-1 and 87-2 have similar effects in tumor rejection and induction of a protective immunity,16whereas B7-2 was found superior to B7-1 in inducing a curative immunity on poorly immunogenic tumor~.~~ At this stage it is not clear how well the animal studies will predict immune responses in humans, because human tumors are in general far less immunogenic than murine tumors. Immunosuppression that is commonly seen in tumor microenvironment may result from tumorinduced down-regulation of T-cell responses.20One mechanism by which tumor cells down-regulate T-cell responses is the release of tumor-derived immunosuppressive factors such as prostaglandin E, (PGE2).28 PGE, is produced by oxidative metabolism of arachidonic acid (AA), which itself is derived from linoleic acid. Cyclooxygenase catalyzes the conversion from AA to PGE, and is known to be produced by a variety of tumors, including human breast cancer cells.14,53, 57, 63 PGE, levels are elevated in advanced malignant breast and it is reported to play a role in breast cancer turnorigene~is,’~, 41 meta~tasis,’~ and tumorinduced immune suppression?, 28, 30, 51, 56 Our recent studies have shown that tumor-derived immune suppressive factors may play an important role in limiting the immune response to gene-modified tumor cells.4 The true impact of PGE, on breast cancer progression is not known; however, it appears to play a variety of roles in breast cancer biology. Diets rich in omega-3 polyunsaturated fatty acids (PUFA) result in the decrease of membrane arachidonic acid levels.6,31 Furthermore, studies in rats fed with diets rich in linoleic acid showed increased prostaglandin levels2 associated with chemically induced mammary tum~rigenesis.~ The rise in breast cancer incidence in Japanese and Eskimo women since 1960 has been associated with qualitative and quantitative changes in dietary fat intake.61Japanese and Eskimo women previously consumed large amounts of fish, rich in omega-3 PUFA, but there is now a trend to a higher consumption of saturated fat and omega-6 PUFA characteristic of the western diets. These changes in the diet of Japanese and Eskimo women have been suggested to account for the increased breast cancer incidence in Japanese and Eskimo women. The inverse relationship between the incidence of breast cancer and the level of fish oil consumption indicates a potential role of omega-3 PUFA in protecting against breast cancer. In the absence of omega-3 PUFA, arachidonic acid is metabolized to PGE,. Animal studies confirmed that diets rich in omega-3 PUFA resulted in the inhibition of the development or progression of mammary carcinomas,” whereas diets rich in the omega-6 PUFA linoleic acid promoted mammary carcinoma growthz1,26 and metastasis of mammary carcinoma.18Thus, both the composition of dietary fat and the caloric intake may influence tumorigenesis. Whether metabolized dietary fat plays a role in tumor-derived immunosuppression has yet to be convincingly demonstrated.
672
BOXHORN & ECK
CONCLUSIONS
A variety of gene therapy strategies have been employed for the treatment of breast cancer. Some of these approaches target the tumor cells, whereas others aim at modulating the function of immune cells. Molecular changes such as activation of oncogenes or mutation of suppressor genes in breast cancer cells have been proposed as targets for gene therapy. Several investigators have used gene transfer to interfere with oncogene products or restore suppressor gene functions. Gene therapy approaches that aim at enhancing the immune responses against breast cancer cells have the perceived advantage of not having to introduce the gene into all metastatic sites. These therapies include passive and active immunization, introduction of cytokines, and expression of T-cell co-stimulatory molecules. Gene therapy clinical trials are underway for a variety of other cancers; however, given the magnitude of this clinical problem, comparatively fewer gene therapy clinical trials have been initiated for breast cancer. The heterogeneity of breast cancer biology and the secretion of immunosuppressive factors are two examples of the difficulties faced in breast cancer gene therapy. Nevertheless, recent improvements in our understanding of the cellular and molecular biology of breast cancer have revealed several potentially clinically useful gene therapy approaches. Thus, gene therapy of breast cancer may offer an alternative form of treatment or may be useful in combination with conventional therapies. References 1. Albo D, Berger DH, Wang TN, et al: Thrombospondin-1 and transforming growth factor-beta 1promote breast tumor cell invasion through up-regulation of the plasmirogen/plasmin system. Surgery 122493, 1997 2. Aylsworth CF, Jone C, Trosko JE, et al: Promotion of 7,12-dimethylbenz[a]anthraceneinduced mammary tumorigenesis by high dietary fat in the r a t Possible role of intercellular communication. J Natl Cancer Inst 72637, 1984 3. Bourguignon LY, Zhu H, Chu A, et al: Interaction between the adhesion receptor, CD44, and the oncogene product, p185HER2, promotes human ovarian tumor cell activation. J Biol Chem 272:27913, 1997 4. Boxhorn HKE, Smith JG, Chang Y, et al: Adenoviral transduction of melanoma cells with B7-1: Anti-tumor immunity and immunosuppressive factors. Cancer Immunol Immunother 46:(in press), 1998 5. Brunda MJ, Herberman RB, Holden HT Inhibition of murine natural killer cell activity by prostaglandins. J Immunol 124:2682, 1980 6. Calder PC, Costa-Rosa LF, Curi R Effects of feeding lipids of different fatty acid compositions upon rat lymphocyte proliferation. Life Sci 56:455, 1995 7. Carter CA, Milholland RJ, Shea W, et al: Effect of the prostaglandin synthetase inhibitor indomethacin on 7,12-dimethylbenz(a)anthracene-inducedmammary tumorigenesis in rats fed different levels of fat. Cancer Res 433559, 1983 8. Chen L, Ashe S, Brady WA, Hellstrom I, et a1 Costimulation of antitumor immunity by the B7 counterreceptor for the T lymphocyte molecules CD28 and CTLA-4. Cell 71:1093, 1992 9. Chen L, McGowan P, Ashe S, et al: Tumor immunogenicity determines the effect of 87 costimulation on T cell-mediated tumor immunity. J Exp Med 179:523, 1994
GENE THERAPY FOR BREAST CANCER
673
10. Couez D, Pages F, Ragueneau M, et al: Functional expression of human CD28 in murine T cell hybridomas. Mol Immunol 31:47, 1994 11. Coughlin CM, Wysocka M, Kurzawa HL, et al: 87-1 and interleukin 12 synergistically induce effective antitumor immunity. Cancer Res 55:4980, 1995 12. Elder EM, Lotze MT, Whiteside TL Successful culture and selection of cytokine genemodified human dermal fibroblasts for the biologic therapy of patients with cancer. Hum Gene Ther 7479, 1996 13. Fearon ER, Pardoll DM, Itaya T, et al: Interleukin-2 production by tumor cells bypasses T helper function in the generation of an antitumor response. Cell 60:397, 1990 14. Fulton A, Zhang S, Chong Y: Role of the prostaglandin E2 receptor in mammary tumor metastasis. Cancer Res 512017, 1991 15. Cuinan EC, Gribben JG, Boussiotis VA, et al: Pivotal role of the B7CD28 pathway in transplantation tolerance and tumor immunity. Blood 84:3261, 1994 16. Hodge JW, Abrams S, Schlom J, et al: Induction of antitumor immunity by recombinant vaccinia viruses expressing 87-1 or B7-2 costimulatory molecules. Cancer Res 54:5552, 1994 17. Hoover JC, Hanna MC Jr: Immunotherapy by active specific immunization. In DeVita VT, Hellman S, Rosenberg SA (eds): Biological Therapy of Cancer. Philadelphia, JB Lippincott, 1991, p 670 18. Hubbard NE, Erickson KL Enhancement of metastasis from a transplantation mouse mammary tumor by dietary linoleic acid. Cancer Res 476171, 1987 19. Hughes R, Timmermans P, Schrey MP: Regulation of arachidonic acid metabolism, aromatase activity and growth in human breast cancer cells by interleukin-1 beta and phorbol ester: Dissociation of a mediatory role for prostaglandin E2 in the autocrine control of cell function. Int J Cancer 67684, 1996 20. Ioannides CG, Whiteside TL: T cell recognition of human tumors: Implications for molecular immunotherapy of cancer. Clin Immunol Immunopathol 66:91, 1993 21. Ip C: Ability of dietary fat to overcome the resistance of mature female rats to 7,12dimethylbenz(a)anthracene-induced mammary tumorigenesis. Cancer Res 40:2785, 1980 22. Janeway CA Jr: Approaching the asymptote?: Evolution and revolution in immunology. Cold Spring Harbor Symposia on Quantitative Biology 54:1, 1989 23. June CH, Bluestone JA, Nadler LM, et al: The 87 and CD28 receptor families. Immunol Today 15:321, 1994 24. Karmali RA, Marsh J, Fuchs C: Effect of omega-3 fatty acids on growth of a rat mammary tumor. J Natl Cancer Inst 73:457, 1984 25. Karmali RA, Welt S, Thaler HT, et al: Prostaglandins in breast cancer: Relationship to disease stage and hormone status. Br J Cancer 48:689, 1983 26. Kitagawa H, Noguchi M Comparative effects of piroxicam and esculetin on incidence, proliferation, and cell kinetics of mammary carcinomas induced by 7,12-dimethylben~[alanthracenein rats on high- and low-fat diets. Oncology 51:401, 1994 27. Kolata G: Oncogenes give breast cancer prognosis. Science 235:160, 1987 28. Lala PK, Parhar RS, Singh I? Indomethacin therapy abrogates the prostaglandinmediated suppression of natural killer activity in tumor-bearing mice and prevents tumor metastasis. Cell Immunol 99:108, 1986 29. Lanier LL, O’Fallon S, Somoza C, et al: CD80 (87) and CD86 (B70) provide similar costimulatory signals for T cell proliferation, cytokine production, and generation of CTL. J Immunol 154:97, 1995 30. Laning JC, Isaacs CM, Hardin-Young J: Normal human keratinocytes inhibit the proliferation of unprimed T cells by TGF beta and PGE2, but not IL-10. Cell Immunol 175:16, 1997 31. Lee TH, Hoover RL, Williams JD, et al: Effect of dietary enrichment with eicosapentaenoic and docosahexaenoic acids on in vitro neutrophil and monocyte leukotriene generation and neutrophil function. N Engl J Med 3121217, 1985 32. Leong CC, Marley JV, Loh S, et al: Transfection of the gene for B7-1 but not 87-2 can induce immunity to murine malignant mesothelioma. Int J Cancer 71:476, 1997 33. Lesoon-Wood LA, Kim WH, Kleinman HK, et al: Systemic gene therapy with p53
674
BOXHORN & ECK
reduces growth and metastases of a malignant human breast cancer in nude mice. Hum Gene Ther 6395, 1995 34. Li Y, Hellstrom KE, Newby SA, et al: Costimulation by CD48 and B7-1 induces immunity against poorly immunogenic tumors. J Exp Med 183:639,1996 35. Li Z, Shanmugam N, Katayose D, et a1 Enzyme/prodrug gene therapy approach for breast cancer using a recombinant adenovirus expressing Escherichiu coli cytosine deaminase. Cancer Gene Ther 4:113, 1997 36. Marr RA, Addison CL, Snider D, et al: Tumour immunotherapy using an adenoviral vector expressing a membrane-bound mutant of murine TNF alpha. Gene Ther 4:1181, 1997 37. Martin-Fontecha A, Cavallo F, Bellone M, et al: Heterogeneous effects of B7-1 and B72 in the induction of both protective and therapeutic anti-tumor immunity against different mouse tumors. Eur J Immunol 261851, 1996 38. Matulonis U, Dosiou C, Freeman G, et al: B7-1 is superior to B7-2 costimulation in the induction and maintenance of T cell-mediated antileukemia immunity. Further evidence that B7-1 and 87-2 are functionally distinct. J Immunol 1561126, 1996 39. Nielsen LL, Dell J, Maxwell E, et al: Efficacy of p53 adenovirus-mediated gene therapy against human breast cancer xenografts. Cancer Gene Ther 4129,1997 40. Nielsen LL, Gumani M, Syed J, et a1 Recombinant El-deleted adenovirus-mediated gene therapy for cancer: Efficacy studies with p53 tumor suppressor gene and liver histology in tumor xenograft models. Hum Gene Ther 9:681, 1998 41. Noguchi M, Rose DP, Earashi M, et a1 The role of fatty acids and eicosanoid synthesis inhibitors in breast carcinoma. Oncology 52:265, 1995 42. Nossal GJV: Immunological tolerance: Collaboration between antigen and lymphokines. Science 245:147, 1989 43. O’Rourke DM, Qian X, Zhang HT, et al: Trans receptor inhibition of human glioblastoma cells by erbB family ectodomains. Proc Natl Acad Sci USA 943250, 1997 44. Putzer BM, Bramson JL, Addison CL, et al: Combination therapy with interleukin-2 and wild-type p53 expressed by adenoviral vectors potentiates tumor regression in a murine model of breast cancer. Hum Gene Ther 9:707, 1998 45. Raper SE, Haskal ZJ, Ye X, et al: Selective gene transfer into the liver of non-human primates with E l deleted, E2A-defective, or El-E4 deleted recombinant adenoviruses. Hum Gen Ther 9:671, 1998 46. Ruppert JM, Wright M, Rosenfeld M, et al: Gene therapy strategies for carcinoma of the breast. Breast Cancer Res Treat 44:93, 1997 47. Russell SJ: Replicating vectors for gene therapy of cancer: Risks, limitations and prospects. Eur J Cancer 30:1165, 1994 48. Schwartz RH: Costimulation of T lymphocytes: The role of CD28, CTLA-4, and B7/ BB1 in interleukin-2 production and immunotherapy. Cell 71:1065, 1992 49. Seth P, Katayose D, Li Z, et al: A recombinant adenovirus expressing wild type p53 induces apoptosis in drug-resistant human breast cancer cells: A gene therapy approach for drug-resistant cancers. Cancer Gene Ther 4383, 1997 50. Shi YF, Sahai BM, Green DR Cyclosporin A inhibits activation-induced cell death in T-cell hybridomas and thymocytes. Nature 339625, 1989 51. Skibinski G, Kelly R, Harrison C, et al: Relative immunosuppressive activity of human seminal prostaglandins. J Reprod Lmmunol22185, 1992 52. Stewart AK, Lassam NJ, Graham FL, et a1 A phase I study of adenovirus mediated gene transfer of interleukin 2 cDNA into metastatic breast cancer or melanoma. Hum Gene Ther 8:1403, 1997 53. Stoll KE, Ottino P, Duncan JR Interrelationship of ascorbate, arachidonic acid and prostaglandin E2 in 816 melanoma cells. Prostaglandins Leukot Essent Fatty Acids 50:123, 1994 54. Topalian SL, Solomon D, Avis FP, et al: Immunotherapy of patients with advanced cancer using tumor-infiltrating lymphocytes and recombinant interleukin-2: A pilot study. J Clin Oncol6:839, 1988 55. Townsend SE, Allison JF: Tumor rejection after direct costimulation of CD8+ T cells by B7-transfected melanoma cells. Science 259368, 1993
GENE THERAPY FOR BREAST CANCER
675
56. Valitutti S, Castellino F, Aiello FB, et a1 The role of arachidonic acid metabolite PGE2 on T cell proliferative response. J Clin Lab Immunol 29:167, 1989 57. Watson J, Chuah SY: Technique for the primary culture of human breast cancer cells and measurement of their prostaglandin secretion. Clin Sci 83347, 1992 58. Weaver CT, Unanue E R The costimulatory function of antigen-presenting cells. Immuno1 Today 11:49, 1990 59. Weitzman MD, Wilson JM, Eck S L Adenovirus vectors in cancer gene therapy. In Sobol RE, Scanlon KJ (eds): The Internet Book of Gene Therapy: Cancer 1995. Stamford, CT, Appleton & Lange, 1995, p 17 60. Wright M, Grim J, Deshane J, et a 1 An intracellular anti-erbB-2 single-chain antibody is specifically cytotoxic to human breast carcinoma cells overexpressing erbB-2. Gene Ther 4:317, 1997 61. X u M, Kumar D, Srinivas S, et a1 Parenteral gene therapy with p53 inhibits human breast tumors in vivo through a bystander mechanism without evidence of toxicity. Hum Gene Ther 8:177,1997 62. Yang CY, Meng C L Regulation of PG synthase by EGF and PDGF in human oral, breast, stomach, and fibrosarcoma cancer cell lines. J Dental Res 73:1407, 1994 63. Zhang JF, Hu C, Geng Y, et al: Treatment of a human breast cancer xenograft with an adenovirus vector containing an interferon gene results in rapid regression due to viral oncolysis and gene therapy. Proc Natl Acad Sci USA 93:4513, 1996
Address reprint requests to Stephen L. Eck, MD, PhD Division of Hematology and Oncology Department of Medicine Room 409 Stellar-Chance Laboratories 422 Curie Blvd. Philadelphia, PA 19104-6100
0889-8588/98 $8.00
GENE THERAPY
+ .OO
GENE THERAPY FOR BREAST CANCER Heike K. E. Boxhorn, PhD, and Stephen L. Eck, MD, PhD
Surgery, chemotherapy, hormonal therapy, and radiation therapy are the main forms of treatment of breast cancer. These approaches reduce the risk of death and have been curative in the majority of patients, but many tumors will recur as metastatic lesions. Gene therapy offers a potentially useful approach for the treatment of breast cancer inasmuch as a variety of molecular processes can be introduced by gene transfer, which can in principle arrest tumor growth.46Several major obstacles, however, need to be overcome for these approaches to be successful. First, our understanding of the molecular processes that lead to breast cancer is incomplete. Thus, therapies directed at the molecular events known to promote tumor growth may be limited by an incomplete understanding of the underlying molecular and cellular processes. Second, gene delivery into breast cancer tissue is made difficult by the slow growth rate of most breast cancers, which may make it difficult to use some vectors such as retroviruses. Finally, for those gene therapies that are not immediately cytotoxic, the often indolent nature of breast cancer cell growth may require a longer term of gene expression than would be otherwise required. Despite these apparent problems, there has been significant progress in developing gene therapy approaches for breast cancer. The approaches can be divided into two general strategies: (1)approaches that alter the metabolic or signaling pathways within the breast cancer cell; and (2) approaches designed to enhance the immune This work was supported in part by grants UO1 CA65805 from the National Institutes of Health (SLE) and by DAMD17-96-1-6287from the Department of Defense (HKEB).
From the Division of Hematology and Oncology, Department of Medicine, The University of Pennsylvania Medical Center, Philadelphia, Pennsylvania
HEMATOLOGY/ONCOLOGY CLINICS OF NORTH AMERICA VOLUME 12 * NUMBER 3 * JUNE 1998
665
666
BOXHORN & ECK
response to the tumor cells. The latter group has achieved the most attention because immunologic approaches avoid many of the problems identified previously. These approaches are not unique to breast cancer and may be proven more successful in other tumors first. THERAPEUTIC GENES THAT ALTER THE METABOLIC OR SIGNALING PATHWAYS IN BREAST CANCER CELLS All breast cancer cells contain molecular genetic abnormalities that contribute to tumor cell growth. Many of these genetic changes occur sporadically or at such low frequency that they are not useful targets for designing gene-based therapies. A few molecular alterations occur with sufficient frequency that they could become useful targets in a significant portion of breast cancer patients. Of these HER2/neu, p53, and c-myc have received the most attention as therapeutic targets because of their relatively high frequency of alteration in breast cancer. The HER2/neu oncogene encoding a transmembrane receptor protein is over-expressed in 30% of breast cancer patients.27The overexpression of HER2/neu is associated with a worse prognosis27by contributing to increased metastases and decreased sensitivity to chemotherapy. Several therapeutic approaches are being developed to genetically target HER2/neu. One approach is to synthesize an antibody to HER2/neu within the tumor cell itself. Curie1 and have used an adenovirus vector that was modified to encode a single-chain monoclonal antibody that would bind HER2/neu. Human breast cancers that over-expressed HER2/neu were growth-arrested in tissue culture experiments, whereas breast cancers cells that expressed little HER2/neu were much less affected by this treatment. Another approach to down-regulate transmembrane growth factor receptors such as HER2/neu is to express a transdominant-negative inhibitor. Such inhibitors function by dimerizing with the growth factor receptor and thereby prevent the growth factor receptor from dimerizing with itself, as it normally d0es.4~Alternatively, rather than inhibiting the HER2/neu protein function after it is made, it is possible to block the synthesis of this protein altogether. The adenovirus E1A transcription factor (necessary for adenovirus replication) has been shown to block HER2/neu expression by inhibiting the expression of the HER2/neu gene in breast cancer cells infected with adeno~irus.~ This in turn sensitized the cells to the cytotoxic effects of paclitaxel. This both confirms the previous association of the HER2/neu over-expression with chemoresistance and provides for the possibility that novel therapies can be based on the combined use of gene therapy and chemotherapy. Cell cycle regulation is altered in many tumors including breast cancer. Mutations in the p53 gene are the most common molecular abnormality in cancer. Restoring normal p53 function in breast cancer cells has been shown to decrease tumorigenicity by leading to cell cycle
GENE THERAPY FOR BREAST CANCER
667
arrest and induction of a p o p t ~ s i s .39,~ 49 ~ ,Adenoviral vectors expressing normal p53 efficiently transduced breast cancer cells in culture and animal models. Scientists at the Schering-Plough Research Institute recently reported that intravenously administered recombinant adenovirus expressing p53 substantially reduced the number of lung metastases and total tumor burden in a murine model of metastatic breast cancer.4oIn this model the immune response to the virus and the tumor could not be assessed, because immunodeficient mice were used as hosts of the human breast cancer xenografts. This may both underestimate the toxicity induced by anti-adenovirus immune responses and obscure any immune response to the tumor. Nonetheless, this experiment convincingly demonstrates that this approach has potential for treating metastatic disease.4oFrom this study, as well as many previous studies, it is apparent that intravenous dosing of adenovirus vectors (the current vector of choice for p53) induces significant liver injury. Recent work by Wilson's group45indicates that the newer-generation adenovirus vectors that have deletions of the E4 adenovirus genes (in addition to E l gene deletion) have much less hepatotoxicity than do the adenovirus vector employed by the Schering-Plough group. Application of these new adenovirus vectors to p53-based therapies could likely offer substantial improvements to therapeutic strategies. In addition to the direct effects on cell cycle, overexpression of the normal p53 gene has been shown to restore sensitivity to doxorubicin (Adriamy~in).~~ The expression of normal p53 within the tumor cells also alters the tumor microenvironment through a paracrine effect. It has been shown that over-expression of p53 leads to an overall reduction in tumor blood vessels.6zThis effectively extends the influence of the gene transfer far beyond the confines of the tumor cell that expresses the therapeutic gene. The enforced expression of p53 has been shown to up-regulate thrombospondin synthesis, which in turn is a negative regulator of tumor angiogenesis.' The contribution of this antiangiogenic effect to the overall therapy may turn out to be important, because with the current technology it is not possible to effect gene transfer in all tumor cells. Systemic antitumor effects may be achieved with tumor suppressor gene therapy by co-administering a cytokine gene. Graham's group4 compared the effectiveness of p53 and IL-2 gene therapies either alone or in combination using adenovirus vectors. They found that a single injection of a combination of the two vectors resulted in tumor regressions in 65% of the treated mice. In this immune competent transgenic mouse mammary adenocarcinoma model, half of'the treated mice remained tumor-free. These mice developed systemic immunity to the tumors as demonstrated by rechallenge experiments and in vitro measures of tumor-specific cytolytic T-lymphocyte activity. Adenovirus delivery of p53 or IL-2 alone produced only transient delays in tumor growth.4 The heterogeneous nature of genetic mutations in breast cancers has encouraged the development of strategies that are less dependent on the genetic alterations of the tumor cell. Introduction of prodrug activating
668
BOXHORN & ECK
genes has been extensively studied in brain tumors, and it may be applicable to breast cancer therapy. In this strategy the tumor cell is genetically modified to express an enzyme capable of activating an otherwise inactive prodrug. In the case of breast cancer, the cytosine deaminase gene has been used for this purpose. Cytosine deaminase (a bacterial enzyme) can activate 5-fluorocytidine to 5-fluorouracil, a chemotherapy agent with known activity and widespread use in breast cancer. The advantage of the gene therapy approach is that sustained high levels of the drug can be achieved within the tumor, whereas the systemic exposure to the drug is low. Seth and colleagues49have demonstrated that an adenovirus vector carrying the cytosine deaminase gene can render breast cancer cells 1000 fold more sensitive to 5-flU01-0cytidine than are mock transfected controls. Furthermore, only 10% of the cells needed to express the cytosine deaminase gene in order to achieve arrest of tumor cell growth. This bystander effect arises from the high rate of diffusion of 5-fluorouracil out of the transfected cells into nearby cells that lack the cytosine deaminase gene.35Application of this approach to human disease will require the development of genetic vectors capable of reaching widespread metastases. GENETIC ENHANCEMENT OF THE IMMUNE RESPONSE TO BREAST CANCER
Immunotherapy offers a promising alternative to conventional therapy because the systemic nature of the immune response can access disseminated disease, and its specificity may limit undesirable side effects. A variety of immune cells, including natural killer (NK) cells, lymphokine activated killers (LAK), macrophages, B cells, and T cells mediate antitumor immune effects. Among these cells, T cells possess unique characteristics that render them attractive targets for the development of immunotherapy. T cells are highly specific in recognizing antigenic peptides. This is ideal for an anticancer reagent that is expected to recognize and destroy tumor cells, but not normal cells. Once activated, T cells differentiate into long-lived memory cells that recirculate continuously. This enables T cells to detect and destroy metastases that arise at distant sites at an early stage. Although new monoclonal antibody therapies are entering into clinical practice, the enhancement of T cell functions has become the major goal of immunotherapy against breast cancer. Prior experience with immunotherapies highlights both the potential and the problems associated with gene transfer-based immunotherapy. Tumor infiltrating T cells (TILs) can be expanded ex vivo prior to readministration of the enlarged population of T cells. This form of passive (and transient) immunization in combination with IL-2 immunotherapy has been applied to patients with a variety of cancers including breast cancer.54Rather than readministrating tumor-reactive T cells, active immunization can be used to enhance T-cell functions and potentially provide a longer period of therapy. Different vaccination strategies
GENE THERAPY FOR BREAST CANCER
669
using DNA (encoding for a tumor antigen), purified protein, peptide, whole tumor cells, or tumor lysates have been developed for the treatment of breast cancer and other tumors.17 In these early studies there was no survival advantage for patients treated. Genetically modified tumor cells have been used to induce a cellular antitumor immune The rationale of using whole tumor cells as a vaccine is that they express the entire repertoire of antigens that would not be present when a defined tumor antigen was used. Thus, the knowledge of specific antigens is not required for the development of a vaccination strategy. Furthermore, recent improvements in our understanding of activation of immune functions may allow one to improve T-cell recognition and activation by genetically modifying tumor cells. For example, tumor cells engineered to express IL-2 were rejected in animal studies and induced protective immunityI3 under conditions in which the unmodified tumor cells grew progressively. A growing number of cytokines and related soluble immune stimulatory factors have been systemically administered to bolster the immune response. The principal problems encountered in the cytokine clinical trials have arisen from the systemic toxicity. Gene transfer strategies may overcome this by the local expression of cytokines within the tumor microenvironment. One approach is to isolate human skin cells (fibroblasts) and genetically alter them so that they secrete one of a number of cytokines.'* The gene-modified cells can then be introduced with the tumor cells as an irradiated vaccine.'* This approach allows the technique of gene delivery to be standardized in fibroblasts and is more reproducible than gene modification of tumor cells. Nonetheless, isolation and ex vivo modification of fibroblasts are cumbersome and expensive procedures and not amenable to use in most clinical practices. In vivo gene transfer overcomes this problem and can be readily implemented using adenovirus vectors.59For example, an adenovirus expressing tumor necrosis factor-alpha (TNFa) has been used in murine models of breast cancer to induce complete regression of spontaneous mammary tumors.36Although the vector was injected into the tumor, systemic levels of TNFa were achieved and resulted in significant toxicity. This was overcome, without loss of antitumor efficacy, by altering the TNFa so that it would remain bound to the tumor cells and not be released into the circulation. Not only was tumor regression observed, but these animals developed tumor specific immunity? Similarly, a recombinant adenovirus expressing a human interferon consensus gene has been shown to induce regressions of human breast cancers grown in immunodeficient mice.63This effect is likely the result of several independent processes, including the production of interferon and direct tumor cell lysis by the adenovirus. The oncolytic effects of viruses are well known and have been studied clinically in small patient studies.47Thus, the use of viral oncolysis to release tumor antigens and immune stimulation by local production of a cytokine may warrant further study in clinical trials. Stewart and colleagues5*may achieve such a result in their recently described plans for a phase 1/11 study using a recombinant adenovirus
670
BOXHORN & ECK
expressing IL-2. A replication-defective adenovirus carrying the IL-2 gene will be injected into metastatic breast cancer tumor sites. This is anticipated to result in very high intratumor concentrations of IL-2 and lesser systemic levels. The adenovirus vector itself will undoubtedly elicit an immune response and result in local destruction of tumor cells as a result of pre-existing antiadenovirus immunity. Prior to their destruction, however, the tumor cells will make IL-2, which is intended to augment the T cell-mediated antitumor response and result in a systemic immune response to the breast cancer. The IL-2-secreting cells will be eliminated within a few days, so that the IL-2 exposure will be brief, with high levels of IL-2 confined to the tumor site. T-cell activation has been shown to require two distinct signals: an 23, 42 The antigenantigen-specific and an antigen-nonspecific specific signal (signal 1) is provided by the interaction of the T-cell receptor (TCR) with peptides presented on the cell surface in the major histocompatability complex proteins (MHC). A secondary signal (signal 2) is generated by a co-stimulatory molecule (e.g., B7-l), which interacts with its cognate ligand (CD28) on T cells. Normally both signals are provided by antigen-presenting cells that process tumor antigens acquired from the local environment. The interaction of B7-1 with CD28 results in enhanced IL-2 production by T cells and proliferation and generation of cytotoxic T cells.29,48 In the absence of signal 1, T cells do not proliferate but remain capable to do so upon antigen presentation. The absence of a co-stimulatory signal (as occurs when tumor cells In this stage the T cell is present antigen to T cells) results in anergy.10,50 unresponsive to specific antigens and unable to respond to a subsequent optimal stimulation with signal 1and 2.15Thus, co-stimulatory molecules alone do not initiate, but rather enable the generation and amplification 58 B7-1 is not of antigen-specific T-cell responses and effector functions.23* expressed by most tumor cells, including breast cancer cells.I5Based on this concept, gene transfer of the B7-1 gene to breast cancer cells can be used to drive the expression of B7-1 on the tumor cells that already express a variety of tumor antigens in MHC. Expression of B7-1 on a breast cancer cell vaccine has been shown to induce protective immunity in animal studies.l' This effect could be further enhanced by the coadministration of IL-12." These responses were shown to be mediated by T cells, indicating the ability of modified tumor cells to induce a cellular immune response. This strategy is currently being tested in a phase I study of patients with refractory breast cancer and melanoma. Many investigators have shown that the expression of B7-1 on murine tumors results in tumor rejection8,*l, 16, 55 and the induction of protective immunity8,11, 34, 37, 38, 55; however, the effectiveness of B7-mediated antitumor immunity is highly dependent on the tumor model chosen and the immunogenicity of the tumor. Tumor cells are generally termed immunogenic when they express antigens that result in the rejection of tumor cells by syngeneic animals previously immunized with the irradiated parental tumor cells.9*32 Accordingly, non-immunogenic tumors are not rejected when similarly tested. Enhanced protective
GENE THERAPY FOR BREAST CANCER
671
antitumor immunity and increased specific cytotoxic T-lymphocyte (CTL) activity was achieved by B7-1 transduced immunogenic tumors but not by B7-1 transduced non-immunogenic tumor^.^ In immunogenic tumor systems, 87-1 and 87-2 have similar effects in tumor rejection and induction of a protective immunity,16whereas B7-2 was found superior to B7-1 in inducing a curative immunity on poorly immunogenic tumor~.~~ At this stage it is not clear how well the animal studies will predict immune responses in humans, because human tumors are in general far less immunogenic than murine tumors. Immunosuppression that is commonly seen in tumor microenvironment may result from tumorinduced down-regulation of T-cell responses.20One mechanism by which tumor cells down-regulate T-cell responses is the release of tumor-derived immunosuppressive factors such as prostaglandin E, (PGE2).28 PGE, is produced by oxidative metabolism of arachidonic acid (AA), which itself is derived from linoleic acid. Cyclooxygenase catalyzes the conversion from AA to PGE, and is known to be produced by a variety of tumors, including human breast cancer cells.14,53, 57, 63 PGE, levels are elevated in advanced malignant breast and it is reported to play a role in breast cancer turnorigene~is,’~, 41 meta~tasis,’~ and tumorinduced immune suppression?, 28, 30, 51, 56 Our recent studies have shown that tumor-derived immune suppressive factors may play an important role in limiting the immune response to gene-modified tumor cells.4 The true impact of PGE, on breast cancer progression is not known; however, it appears to play a variety of roles in breast cancer biology. Diets rich in omega-3 polyunsaturated fatty acids (PUFA) result in the decrease of membrane arachidonic acid levels.6,31 Furthermore, studies in rats fed with diets rich in linoleic acid showed increased prostaglandin levels2 associated with chemically induced mammary tum~rigenesis.~ The rise in breast cancer incidence in Japanese and Eskimo women since 1960 has been associated with qualitative and quantitative changes in dietary fat intake.61Japanese and Eskimo women previously consumed large amounts of fish, rich in omega-3 PUFA, but there is now a trend to a higher consumption of saturated fat and omega-6 PUFA characteristic of the western diets. These changes in the diet of Japanese and Eskimo women have been suggested to account for the increased breast cancer incidence in Japanese and Eskimo women. The inverse relationship between the incidence of breast cancer and the level of fish oil consumption indicates a potential role of omega-3 PUFA in protecting against breast cancer. In the absence of omega-3 PUFA, arachidonic acid is metabolized to PGE,. Animal studies confirmed that diets rich in omega-3 PUFA resulted in the inhibition of the development or progression of mammary carcinomas,” whereas diets rich in the omega-6 PUFA linoleic acid promoted mammary carcinoma growthz1,26 and metastasis of mammary carcinoma.18Thus, both the composition of dietary fat and the caloric intake may influence tumorigenesis. Whether metabolized dietary fat plays a role in tumor-derived immunosuppression has yet to be convincingly demonstrated.
672
BOXHORN & ECK
CONCLUSIONS
A variety of gene therapy strategies have been employed for the treatment of breast cancer. Some of these approaches target the tumor cells, whereas others aim at modulating the function of immune cells. Molecular changes such as activation of oncogenes or mutation of suppressor genes in breast cancer cells have been proposed as targets for gene therapy. Several investigators have used gene transfer to interfere with oncogene products or restore suppressor gene functions. Gene therapy approaches that aim at enhancing the immune responses against breast cancer cells have the perceived advantage of not having to introduce the gene into all metastatic sites. These therapies include passive and active immunization, introduction of cytokines, and expression of T-cell co-stimulatory molecules. Gene therapy clinical trials are underway for a variety of other cancers; however, given the magnitude of this clinical problem, comparatively fewer gene therapy clinical trials have been initiated for breast cancer. The heterogeneity of breast cancer biology and the secretion of immunosuppressive factors are two examples of the difficulties faced in breast cancer gene therapy. Nevertheless, recent improvements in our understanding of the cellular and molecular biology of breast cancer have revealed several potentially clinically useful gene therapy approaches. Thus, gene therapy of breast cancer may offer an alternative form of treatment or may be useful in combination with conventional therapies. References 1. Albo D, Berger DH, Wang TN, et al: Thrombospondin-1 and transforming growth factor-beta 1promote breast tumor cell invasion through up-regulation of the plasmirogen/plasmin system. Surgery 122493, 1997 2. Aylsworth CF, Jone C, Trosko JE, et al: Promotion of 7,12-dimethylbenz[a]anthraceneinduced mammary tumorigenesis by high dietary fat in the r a t Possible role of intercellular communication. J Natl Cancer Inst 72637, 1984 3. Bourguignon LY, Zhu H, Chu A, et al: Interaction between the adhesion receptor, CD44, and the oncogene product, p185HER2, promotes human ovarian tumor cell activation. J Biol Chem 272:27913, 1997 4. Boxhorn HKE, Smith JG, Chang Y, et al: Adenoviral transduction of melanoma cells with B7-1: Anti-tumor immunity and immunosuppressive factors. Cancer Immunol Immunother 46:(in press), 1998 5. Brunda MJ, Herberman RB, Holden HT Inhibition of murine natural killer cell activity by prostaglandins. J Immunol 124:2682, 1980 6. Calder PC, Costa-Rosa LF, Curi R Effects of feeding lipids of different fatty acid compositions upon rat lymphocyte proliferation. Life Sci 56:455, 1995 7. Carter CA, Milholland RJ, Shea W, et al: Effect of the prostaglandin synthetase inhibitor indomethacin on 7,12-dimethylbenz(a)anthracene-inducedmammary tumorigenesis in rats fed different levels of fat. Cancer Res 433559, 1983 8. Chen L, Ashe S, Brady WA, Hellstrom I, et a1 Costimulation of antitumor immunity by the B7 counterreceptor for the T lymphocyte molecules CD28 and CTLA-4. Cell 71:1093, 1992 9. Chen L, McGowan P, Ashe S, et al: Tumor immunogenicity determines the effect of 87 costimulation on T cell-mediated tumor immunity. J Exp Med 179:523, 1994
GENE THERAPY FOR BREAST CANCER
673
10. Couez D, Pages F, Ragueneau M, et al: Functional expression of human CD28 in murine T cell hybridomas. Mol Immunol 31:47, 1994 11. Coughlin CM, Wysocka M, Kurzawa HL, et al: 87-1 and interleukin 12 synergistically induce effective antitumor immunity. Cancer Res 55:4980, 1995 12. Elder EM, Lotze MT, Whiteside TL Successful culture and selection of cytokine genemodified human dermal fibroblasts for the biologic therapy of patients with cancer. Hum Gene Ther 7479, 1996 13. Fearon ER, Pardoll DM, Itaya T, et al: Interleukin-2 production by tumor cells bypasses T helper function in the generation of an antitumor response. Cell 60:397, 1990 14. Fulton A, Zhang S, Chong Y: Role of the prostaglandin E2 receptor in mammary tumor metastasis. Cancer Res 512017, 1991 15. Cuinan EC, Gribben JG, Boussiotis VA, et al: Pivotal role of the B7CD28 pathway in transplantation tolerance and tumor immunity. Blood 84:3261, 1994 16. Hodge JW, Abrams S, Schlom J, et al: Induction of antitumor immunity by recombinant vaccinia viruses expressing 87-1 or B7-2 costimulatory molecules. Cancer Res 54:5552, 1994 17. Hoover JC, Hanna MC Jr: Immunotherapy by active specific immunization. In DeVita VT, Hellman S, Rosenberg SA (eds): Biological Therapy of Cancer. Philadelphia, JB Lippincott, 1991, p 670 18. Hubbard NE, Erickson KL Enhancement of metastasis from a transplantation mouse mammary tumor by dietary linoleic acid. Cancer Res 476171, 1987 19. Hughes R, Timmermans P, Schrey MP: Regulation of arachidonic acid metabolism, aromatase activity and growth in human breast cancer cells by interleukin-1 beta and phorbol ester: Dissociation of a mediatory role for prostaglandin E2 in the autocrine control of cell function. Int J Cancer 67684, 1996 20. Ioannides CG, Whiteside TL: T cell recognition of human tumors: Implications for molecular immunotherapy of cancer. Clin Immunol Immunopathol 66:91, 1993 21. Ip C: Ability of dietary fat to overcome the resistance of mature female rats to 7,12dimethylbenz(a)anthracene-induced mammary tumorigenesis. Cancer Res 40:2785, 1980 22. Janeway CA Jr: Approaching the asymptote?: Evolution and revolution in immunology. Cold Spring Harbor Symposia on Quantitative Biology 54:1, 1989 23. June CH, Bluestone JA, Nadler LM, et al: The 87 and CD28 receptor families. Immunol Today 15:321, 1994 24. Karmali RA, Marsh J, Fuchs C: Effect of omega-3 fatty acids on growth of a rat mammary tumor. J Natl Cancer Inst 73:457, 1984 25. Karmali RA, Welt S, Thaler HT, et al: Prostaglandins in breast cancer: Relationship to disease stage and hormone status. Br J Cancer 48:689, 1983 26. Kitagawa H, Noguchi M Comparative effects of piroxicam and esculetin on incidence, proliferation, and cell kinetics of mammary carcinomas induced by 7,12-dimethylben~[alanthracenein rats on high- and low-fat diets. Oncology 51:401, 1994 27. Kolata G: Oncogenes give breast cancer prognosis. Science 235:160, 1987 28. Lala PK, Parhar RS, Singh I? Indomethacin therapy abrogates the prostaglandinmediated suppression of natural killer activity in tumor-bearing mice and prevents tumor metastasis. Cell Immunol 99:108, 1986 29. Lanier LL, O’Fallon S, Somoza C, et al: CD80 (87) and CD86 (B70) provide similar costimulatory signals for T cell proliferation, cytokine production, and generation of CTL. J Immunol 154:97, 1995 30. Laning JC, Isaacs CM, Hardin-Young J: Normal human keratinocytes inhibit the proliferation of unprimed T cells by TGF beta and PGE2, but not IL-10. Cell Immunol 175:16, 1997 31. Lee TH, Hoover RL, Williams JD, et al: Effect of dietary enrichment with eicosapentaenoic and docosahexaenoic acids on in vitro neutrophil and monocyte leukotriene generation and neutrophil function. N Engl J Med 3121217, 1985 32. Leong CC, Marley JV, Loh S, et al: Transfection of the gene for B7-1 but not 87-2 can induce immunity to murine malignant mesothelioma. Int J Cancer 71:476, 1997 33. Lesoon-Wood LA, Kim WH, Kleinman HK, et al: Systemic gene therapy with p53
674
BOXHORN & ECK
reduces growth and metastases of a malignant human breast cancer in nude mice. Hum Gene Ther 6395, 1995 34. Li Y, Hellstrom KE, Newby SA, et al: Costimulation by CD48 and B7-1 induces immunity against poorly immunogenic tumors. J Exp Med 183:639,1996 35. Li Z, Shanmugam N, Katayose D, et a1 Enzyme/prodrug gene therapy approach for breast cancer using a recombinant adenovirus expressing Escherichiu coli cytosine deaminase. Cancer Gene Ther 4:113, 1997 36. Marr RA, Addison CL, Snider D, et al: Tumour immunotherapy using an adenoviral vector expressing a membrane-bound mutant of murine TNF alpha. Gene Ther 4:1181, 1997 37. Martin-Fontecha A, Cavallo F, Bellone M, et al: Heterogeneous effects of B7-1 and B72 in the induction of both protective and therapeutic anti-tumor immunity against different mouse tumors. Eur J Immunol 261851, 1996 38. Matulonis U, Dosiou C, Freeman G, et al: B7-1 is superior to B7-2 costimulation in the induction and maintenance of T cell-mediated antileukemia immunity. Further evidence that B7-1 and 87-2 are functionally distinct. J Immunol 1561126, 1996 39. Nielsen LL, Dell J, Maxwell E, et al: Efficacy of p53 adenovirus-mediated gene therapy against human breast cancer xenografts. Cancer Gene Ther 4129,1997 40. Nielsen LL, Gumani M, Syed J, et a1 Recombinant El-deleted adenovirus-mediated gene therapy for cancer: Efficacy studies with p53 tumor suppressor gene and liver histology in tumor xenograft models. Hum Gene Ther 9:681, 1998 41. Noguchi M, Rose DP, Earashi M, et a1 The role of fatty acids and eicosanoid synthesis inhibitors in breast carcinoma. Oncology 52:265, 1995 42. Nossal GJV: Immunological tolerance: Collaboration between antigen and lymphokines. Science 245:147, 1989 43. O’Rourke DM, Qian X, Zhang HT, et al: Trans receptor inhibition of human glioblastoma cells by erbB family ectodomains. Proc Natl Acad Sci USA 943250, 1997 44. Putzer BM, Bramson JL, Addison CL, et al: Combination therapy with interleukin-2 and wild-type p53 expressed by adenoviral vectors potentiates tumor regression in a murine model of breast cancer. Hum Gene Ther 9:707, 1998 45. Raper SE, Haskal ZJ, Ye X, et al: Selective gene transfer into the liver of non-human primates with E l deleted, E2A-defective, or El-E4 deleted recombinant adenoviruses. Hum Gen Ther 9:671, 1998 46. Ruppert JM, Wright M, Rosenfeld M, et al: Gene therapy strategies for carcinoma of the breast. Breast Cancer Res Treat 44:93, 1997 47. Russell SJ: Replicating vectors for gene therapy of cancer: Risks, limitations and prospects. Eur J Cancer 30:1165, 1994 48. Schwartz RH: Costimulation of T lymphocytes: The role of CD28, CTLA-4, and B7/ BB1 in interleukin-2 production and immunotherapy. Cell 71:1065, 1992 49. Seth P, Katayose D, Li Z, et al: A recombinant adenovirus expressing wild type p53 induces apoptosis in drug-resistant human breast cancer cells: A gene therapy approach for drug-resistant cancers. Cancer Gene Ther 4383, 1997 50. Shi YF, Sahai BM, Green DR Cyclosporin A inhibits activation-induced cell death in T-cell hybridomas and thymocytes. Nature 339625, 1989 51. Skibinski G, Kelly R, Harrison C, et al: Relative immunosuppressive activity of human seminal prostaglandins. J Reprod Lmmunol22185, 1992 52. Stewart AK, Lassam NJ, Graham FL, et a1 A phase I study of adenovirus mediated gene transfer of interleukin 2 cDNA into metastatic breast cancer or melanoma. Hum Gene Ther 8:1403, 1997 53. Stoll KE, Ottino P, Duncan JR Interrelationship of ascorbate, arachidonic acid and prostaglandin E2 in 816 melanoma cells. Prostaglandins Leukot Essent Fatty Acids 50:123, 1994 54. Topalian SL, Solomon D, Avis FP, et al: Immunotherapy of patients with advanced cancer using tumor-infiltrating lymphocytes and recombinant interleukin-2: A pilot study. J Clin Oncol6:839, 1988 55. Townsend SE, Allison JF: Tumor rejection after direct costimulation of CD8+ T cells by B7-transfected melanoma cells. Science 259368, 1993
GENE THERAPY FOR BREAST CANCER
675
56. Valitutti S, Castellino F, Aiello FB, et a1 The role of arachidonic acid metabolite PGE2 on T cell proliferative response. J Clin Lab Immunol 29:167, 1989 57. Watson J, Chuah SY: Technique for the primary culture of human breast cancer cells and measurement of their prostaglandin secretion. Clin Sci 83347, 1992 58. Weaver CT, Unanue E R The costimulatory function of antigen-presenting cells. Immuno1 Today 11:49, 1990 59. Weitzman MD, Wilson JM, Eck S L Adenovirus vectors in cancer gene therapy. In Sobol RE, Scanlon KJ (eds): The Internet Book of Gene Therapy: Cancer 1995. Stamford, CT, Appleton & Lange, 1995, p 17 60. Wright M, Grim J, Deshane J, et a 1 An intracellular anti-erbB-2 single-chain antibody is specifically cytotoxic to human breast carcinoma cells overexpressing erbB-2. Gene Ther 4:317, 1997 61. X u M, Kumar D, Srinivas S, et a1 Parenteral gene therapy with p53 inhibits human breast tumors in vivo through a bystander mechanism without evidence of toxicity. Hum Gene Ther 8:177,1997 62. Yang CY, Meng C L Regulation of PG synthase by EGF and PDGF in human oral, breast, stomach, and fibrosarcoma cancer cell lines. J Dental Res 73:1407, 1994 63. Zhang JF, Hu C, Geng Y, et al: Treatment of a human breast cancer xenograft with an adenovirus vector containing an interferon gene results in rapid regression due to viral oncolysis and gene therapy. Proc Natl Acad Sci USA 93:4513, 1996
Address reprint requests to Stephen L. Eck, MD, PhD Division of Hematology and Oncology Department of Medicine Room 409 Stellar-Chance Laboratories 422 Curie Blvd. Philadelphia, PA 19104-6100