Molecular Modalities in the Treatment of Lung Cancer

ABSTRACT: Despite recent advances in the treatment of lung cancer, long-term survival remains rare. As more information pertaining to the biology of lung cancer is understood, it is hoped that improvements in outcome can be realized with the use of molecularly based therapies. The identification of gene mutations in lung cancer has led to the development of inhibitory therapies, including antisense oligonucleotides and direct injection of tumor-suppressor genes, such as wild-type p53. Other therapeutic approaches are targeted at inhibiting angiogenesis by blocking endogenous growth factors with antibodies or administering natural antiangiogenic substances. Recognition of the dendritic cell as one of the primary cells responsible for antitumor immunity has encouraged studies of immunotherapy for patients with lung cancer. In addition, studies have shown that dendritic cell function is defective in tumor-bearing animals. Research continues to explore the effect of tumor on immune cell function and ways to overcome such defects. Rationally derived therapies based on these biological findings may advance the treatment, as well as early detection and prevention, of lung cancer, thereby improving patient outcomes. [ONCOLOGY 13(Suppl 5):142-147, 1999]

Introduction

In spite of many complex and aggressive approaches to therapy and great strides in understanding the biology and etiology of lung cancer, corresponding improvements in outcome are not yet apparent. It is hoped that in the future, advances in our knowledge of the molecular biology of lung cancer will provide the foundation for real improvement in outcomes. Emerging molecularly based modalities that may soon be combined with chemotherapy, radiation therapy, and surgery to improve the effectiveness of lung cancer treatment are discussed.

Knowledge of Lung Cancer Precursors May Allow Prevention or Early Detection

It is quite clear that lung cancer is caused by an accumulation of genetic damage in the bronchial epithelium. Exposure to inhaled carcinogens, such as the polycyclic aromatic hydrocarbons and the nitrosamines from cigarette smoke, can directly damage the DNA of bronchial epithelial cells. These agents covalently modify DNA, causing misreplication and mutation or loss of genetic material. The bound carcinogens can be directly detected in the DNA of smokers, and, disturbingly, also in the DNA of infants born to smoking mothers.[1] Using modern molecular detection techniques, loss of genetic material from large regions of chromosomes or point mutations in dominant or recessive oncogenes can be found in the bronchial epithelium of smokers, even those without microscopically visible histological changes.[2] Prolonged exposure results in visible hyperplasia and metaplasia, which often, but not always, precede frank malignancy. It is likely that detectable genetic abnormalities always precede the development of invasive cancer. New molecular markers of loss of growth control, such as loss of expression of the retinoic acid receptor beta (RAR-b), have been found to be strongly associated with malignant progression and are being tested as molecular intermediate markers in chemoprevention and early detection studies. These genetic premalignant changes are widespread throughout the respiratory epithelium, suggesting that a field effect is induced by the carcinogens,[3] explaining the high incidence of second malignancies in those cured of lung cancer or head and neck cancer.

Lung Cancer Biology: Gene Therapy and Receptor-Targeted Therapeutics

It is clinically useful to categorize bronchogenic cancers into two groups that reflect their biology and management: small-cell lung cancer and non–small-cell lung cancer. Small-cell lung cancer is highly responsive to chemotherapy, but only very infrequently curable, as it rapidly relapses and metastasizes. Non–small-cell lung cancer is often less dramatically responsive to chemotherapy, but is more often cured by surgery or combined-modality therapy. Each of these categories is divided into subtypes, but as mentioned above, in reality, these categories often blend into each other or coexist with each other. Data on cellular and molecular biology, as well as ultrastructural studies, can help refine these groupings, and more importantly, perhaps guide therapy in the future.

Small-cell lung cancer tumors and a subset of non–small-cell lung cancer tumors express many neuroendocrine markers. Neuroendocrine cells are present in small numbers in many tissues and share many properties with neural cells, hence the term. The primary function of these neuroendocrine cells is to produce, package, and secrete small peptide or amine hormones. Lung tumors, especially small-cell lung cancer, may secrete factors that stimulate their own growth (autocrine secretion). Individual tumors may secrete up to 10 discrete hormones, which may contribute to the paraneoplastic syndromes often associated with small-cell lung cancer. Cross-reactive antigens, such as the HuD gene, may also lead to autoimmune paraneoplastic syndromes.

Small-cell lung cancer is strongly associated with cigarette smoking, and nearly always demonstrates loss of genetic material on chromosomes,[4] including the gene for RAR-b and FHIT (fragile histidine triad), but the important genes are not fully identified. Mutations in the ras oncogene are rare, but mutations in p53 and overexpression of bcl-2 are nearly universal. These abnormalities form the basis for several new therapeutic approaches. One of these is the inhibition of expression of bcl-2 in small-cell lung cancer using antisense oligonucleotides[4] or other approaches. The antisense oligonucleotides have been found to be highly effective in cell lines, and if nontoxic methods can be developed to inhibit bcl-2 expression in patients, this will be a promising new modality.

Non–small-cell lung cancer is a morphologically diverse group that includes the squamous (epidermoid) carcinoma, adenocarcinoma, and large-cell carcinoma. The squamous phenotype used to be the predominant form of lung cancer worldwide, although its relative and absolute incidence in the United States (and other parts of the world such as East Asia) has dramatically declined within the last two decades.[5] Squamous carcinomas are strongly associated with cigarette smoking, and this explains their frequent association with metaplastic and dysplastic changes in adjacent epithelium. Adenocarcinomas have become the most common form of lung cancer in the United States. In general, they tend to arise in the peripheral airways and may possess distinctive intracellular mucin granules as part of their acinar/glandular differentiation.

Mutations in p53 are observed in about half of non–small-cell lung cancers,[6] occurring somewhat more frequently in squamous cell carcinomas, whereas ras mutations are found in about 20% of adenocarcinomas[7] and less frequently in squamous carcinomas. Several approaches are being clinically tested that are based on the tumor-suppressive properties of p53. One of these is the delivery of a normal p53 in a recombinant adenovirus to cause high-level expression of this tumor-suppressor gene. Overexpression of p53 has been found to be selectively toxic to tumor cells and not normal ones. The normal p53 is delivered by direct injection into tumor masses either alone or in combination with radiation or chemotherapy.[8] It is hoped that this combination will allow improved local control and palliation of unresectable tumors. An Eastern Cooperative Oncology Group study is evaluating adenovirus-p53 delivered by bronchoalveolar lavage directly to entire lobes of the lung with bronchoalveolar carcinoma. This approach should allow excellent access of the gene therapeutic vector to the tumor cells that cause the main respiratory symptoms of this disease; that is, those lining and involving the alveoli and small airways.

Overexpression of several growth factor receptors, such as insulin-like growth factor 1 receptor (IGF-1r) and epidermal growth factor receptor (EGFr), as well as HER-2/neu, has also been observed and may be correlated with the biology of lung cancer, and, thus, important therapeutic targets. Several companies are developing small-molecule antagonists of tyrosine kinase receptors, such as EGFr and IGF-1r, for clinical application. Gene-based therapeutic approaches to block these receptors have been effective in animal models.[9] Similarly, antibodies against HER-2/neu, found to be useful in breast cancer, are now being tested in non–small-cell lung cancer.

Tumor Vasculature as a Therapeutic Target

Successful tumors require a blood supply to grow beyond microscopic size and have subverted a variety of normal cellular pathways to achieve this end.[10] A number of factors, including thrombospondin and vascular endothelial growth factor (VEGF) produced by tumors, act upon defined cell-surface receptors in capillary endothelia to promote their ingrowth into solid tumors. Stage I non–small-cell lung cancer tumors with high microvessel density are associated with an adverse prognosis and a higher incidence of metastases.

Inhibitors of angiogenesis cause dramatic inhibition of metastases and tumor growth in animal models; clinical trials using antiangiogenic natural substances or blocking antibodies are planned or under way. These include antibodies that block VEGF binding and others that are already in clinical testing, including naturally derived substances (eg, angiostatin and endostatin) and designed VEGF receptor tyrosine kinase inhibitors. These agents have a significant potential usefulness in the treatment of all forms of cancer, including lung cancer.

Immunotherapy

Intracellular or intranuclear proteins expressed by tumors can be recognized by cytotoxic T cells, resulting in tumor-cell killing. The discovery of mutated or overexpressed proteins in tumor cells relative to normal ones may allow for targeted immunotherapeutic approaches. Responses to both overexpressed normal and mutant p53 and ras can be observed in patients with cancer. Trials are under way that attempt to induce specific anti-p53 immunity to either wild-type or mutant portions of the molecule.

In a recently completed pilot clinical trial, we have successfully induced p53 and ras-specific T-cell responses by vaccination with peptides that were custom designed to match the mutations in the patients’ tumors (manuscript in preparation). A total of 34 patients were treated and immunized up to 19 times. Detectable responses were observed in 21 of these patients. Some of these patients had no evidence of tumor (they were treated in the adjuvant setting), and some of those with evident tumor had unexpected stable disease. We are therefore planning second-generation trials to give us an indication of whether this approach has any clinical effectiveness.

The above study immunized patients with synthetic peptide pulsed onto autologous peripheral blood mononuclear cells. The primary cell in this preparation responsible for immune induction is the dendritic cell. We have shown in animal studies that peptide-loaded dendritic cells are much more effective at induction of antitumor immunity than bulk peripheral blood mononuclear cells. Our second-generation clinical trials will use similar peptides pulsed onto autologous dendritic cells. Another possible approach is to immunize patients with small-cell lung cancer or stage III non–small-cell lung cancer after best standard therapy, as shown in Figure 1.

Mechanisms of Immune Escape

The immune cells responsible for tumor-cell killing appear to be primarily major histocompatibility complex (MHC)-restricted cytotoxic T lymphocytes. Cytotoxic T-lymphocytes detect target cells for killing by recognizing short peptide fragments of endogenous proteins, which are presented to them by class I MHC molecules on the surface of the target cell. There are now specific examples in both rodent models and patients of tumor-cell immunity generated by MHC-restricted cytotoxic T cells detecting endogenous cytoplasmic peptide antigens.[11-19] The identification of clear tumor antigens naturally leads to the question of why these antigens failed to prevent the tumor from developing initially. Several mechanisms may contribute to the failure of immune control of tumor growth, including downregulation of MHC expression and lack of costimulation. However, recent evidence indicates that these factors may not be dominant.[20] It becomes obvious that understanding the mechanisms of tumor escape requires the knowledge of the tumor-bearing host’s immune system.

In the 1970s, several groups studied whether patients with cancer could be effectively immunized against common antigens. Eilber and Morton have shown that only 60% of patients with localized, potentially resectable neoplasms developed delayed cutaneous hypersensitivity after sensitization with 2,4-dinitrochlorbenzene. More than 95% of healthy individuals and 100% of patients with benign tumors developed delayed cutaneous hypersensitivity.[21] Similar observations were reported by several other groups.[22, 23] Stiver and Weinerman immunized patients with localized (not metastatic) cancer with influenza vaccine and found that almost 93% of healthy individuals but only 36% of cancer patients developed fourfold or greater increases in antibody titers.[24] During the last decade, various defects in the function of T-lymphocytes were identified in patients with cancer and in tumor-bearing animals.[25] However, much less is known about the function of antigen-presenting cells in cancer patients. These cells play a crucial role in antitumor immune responses. Moreover, antigen-presenting cells, and not the tumor cells, are now thought to be responsible for the induction of antitumor immunity in a tumor-bearing host.[26,27]

Induction of cytotoxic T-lymphocyte responses requires effective cooperation among at least three elements of the immune system: Antigen-presenting cells, T-helper cells, and cytotoxic T cells. Antigen-presenting cells provide effective presentation of antigen and delivery of costimulatory signals to CD4+ and CD8+ cells, whereas activated CD4+ cells together with monocytes and macrophages provide cytokine support for antigen-specific cytotoxic T-lymphocyte clonal outgrowth. Thus, effective responses to tumor-specific antigens may be possible if both CD4+ and CD8+ lymphocytes are engaged.

The most potent antigen-presenting cells known are dendritic cells. They have several distinct features making them ideal candidates as vehicles for presenting tumor-specific antigens for immunotherapy. Dendritic cells express on their surface a high level of both MHC class I and II molecules, as well as a variety of adhesion and costimulatory molecules.[28] Dendritic cells are extremely effective in the stimulation of secondary immune responses (100 times more potent than B cells or macrophages). Dendritic cells are the only cells capable of stimulating primary immune responses, including cytotoxic ones.[28] They have a high potential to acquire, process, and present soluble antigens on their surface. Dendritic cells arise from bone marrow, can be isolated from peripheral blood, lymph nodes, spleen, and skin, and can be generated in vitro from progenitors. When isolated from the peripheral blood, they are capable of taking up soluble proteins, processing them, and presenting them on their cell surface. High levels of MHC class II expression and antigen presentation, however, appear to require a further step of differentiation, which can be accomplished by overnight growth in vitro.

Murine spleen dendritic cells and their cutaneous counterpart, Langerhans cells, have been reported to induce specific cytotoxic immune responses against tumor antigens.[29,30] Tumor protection has been induced when dendritic cells or Langerhans cells were used for induction of immune responses.[29] Dendritic cells were about 100 times more effective in the induction of anti-influenza virus or anti-human immunodeficiency virus (HIV) specific cytotoxic T-lymphocyte responses than unseparated spleen cells.[31,32] Dendritic cells thus appear to be the ideal vehicle for cellular immunization.

All of these studies, however, have been performed with dendritic cells from normal animals. Direct translation of these studies to the use of autologous dendritic cells in human trials presumes that dendritic cell function in tumor-bearing mice and patients with cancer is normal. However, the functional ability of dendritic cells from animals or patients with cancer to present antigen has not been systematically studied until recently. Even if the antigen-pulsed dendritic cells used for immunization function normally after in vitro growth, it is possible that the lack of effective endogenous dendritic cells can blunt the response.[33] A recent study by Zitvogel et al showed suppression of growth but not eradication of weakly immunogenic tumors (MCA205 and TS/A) after treatment with antigen-pulsed dendritic cells. When the same approach was applied to the immunogenic C3 tumor, complete elimination was observed.[34]

Defective dendritic cell function in tumor-bearing hosts has been described by several groups and could be a major factor responsible for defective immune function in cancer.[35-38] We found that peptide immunization was much less effective at induction of cytotoxic T-lymphocytes in mice even only a few days after tumor challenge than it was in nontumor-bearing animals. This defect was not tumor-epitope specific, since the same effect was observed after immunization with an irrelevant peptide, p 18IIIB, which is known to induce high levels of cytotoxic T-lymphocyte responses in control BALB/c mice.[39] To identify the potential role of dendritic cells and T cells in this process, these cells were separately purified. We found a reduced ability of dendritic cells from tumor-bearing mice to stimulate allogeneic control T cells, a reduced ability to stimulate specific cytotoxic T-lymphocyte responses after incubation with T cells from control immune mice, and a reduced ability to induce cytotoxic T-lymphocyte responses in control mice. These defects were associated with a substantial decrease in the expression of class II and class I MHC on the dendritic cell surface and decreased expression of B7-1.[33] Restimulation of CD4+ and CD8+ cells from tumor-bearing mice with peptide-pulsed control dendritic cells produced near control levels of cytotoxic T-lymphocyte responses.[33] This was confirmed in direct experiments when splenocytes from immunized tumor-bearing mice were restimulated with peptide-pulsed control dendritic cells. These data suggest that defects in antitumor immune responses can be overcome if sufficient antigen presentation is provided.

The mechanism of defective dendritic cell function is not clear. It has been suggested that tumor cells release factors affecting dendritic cell function.[36,40,41] We found that dendritic cells generated from precursors, but not mature dendritic cells from tumor-bearing mice, were able to stimulate effective cytotoxic T-lymphocyte responses in control animals. These data were confirmed in the treatment of tumor-bearing mice.[33,42] Immunization of tumor-bearing mice with peptide-pulsed dendritic cells generated from bone marrow precursors, but not those isolated from spleen, were capable of slowing tumor growth. The results of these experiments indicate that dendritic cells acquire defects during their maturation in tumor-bearing mice. This suggests that tumor cells may release factors affecting dendritic cell differentiation. To test this hypothesis, supernatants were collected from the tumor cells used in this study. Dendritic cells isolated from spleen or generated from bone marrow of control animals were treated with this supernatant for different periods of time. Dendritic cell function was assessed by the ability of the cells to stimulate allogeneic control T cells or peptide-specific cytotoxic T lymphocytes. Tumor-cell supernatants did not affect the function of dendritic cells isolated from spleens of control mice even if the cells were cultured for 5 days in the presence of granulocyte-macrophage colony-stimulating factor (GM-CSF). At the same time, they significantly affected the function of cells generated from bone marrow precursors. Cells generated in the presence of tumor-cell supernatants also had significantly lower levels of MHC class II expression. Thus, from these experiments, we concluded that a soluble factor or factors released by tumor cells may play an important role in dendritic cell dysfunction in cancer.

We then asked whether the dendritic cell dysfunction seen in tumor-bearing mice is observed in patients with cancer. For this purpose, we studied 32 patients with breast cancer. Patients with advanced stages of cancer had a significantly decreased ability to respond to stimulation with influenza virus or tetanus toxoid, two common antigens with prevalent immune responses in the general population. The level of tetanus toxoid-dependent proliferation was decreased more than threefold in breast cancer patients as compared to a control group of healthy individuals; influenza virus-specific cytotoxic T-lymphocyte response was detected in only one of six patients with advanced breast cancer (stages III and IV) and in four of five patients in the control group.[43]

Dendritic cells from patients with advanced-stage breast cancer had a significantly reduced ability to stimulate control allogeneic T cells, whereas the ability of T cells from those patients to respond to allogeneic stimulation was not affected.[43] Dendritic cells generated from precursors in peripheral blood using GM-CSF and interleukin 4 (IL-4) demonstrated normal levels of functional activity. They were able to significantly increase tetanus toxoid-dependent T-cell proliferation and to increase influenza virus-specific cytotoxic T-lymphocyte responses. Thus, these results were in agreement with the results from the animal experiments, indicating defective dendritic cell function in cancer patients, and the ability to overcome, at least in part, such defects using dendritic cells generated from precursors.

In the human system, we tried to reproduce our data concerning the effects of tumor-cell supernatants on dendritic cell function in animal systems, and then to identify the possible mechanisms involved. Twelve human breast cancer cell lines were used to generate conditioned tumor-cell supernatants. The addition of tumor-cell supernatants to control mature dendritic cells isolated from peripheral blood of control individuals had no effect. However, when tumor-cell supernatants were added to cultures of hematopoietic stem cells (CD34+ cells) isolated from umbilical cord blood and grown with GM-CSF and tumor-necrosis factor alpha (TNFa) or IL-4, they dramatically affected normal dendritic cell differentiation.[44] Tumor-cell supernatants were then size- fractionated on fast protein liquid chromatography, and the active factors were shown to migrate in the 30- to 60-kDa range.

After a series of experiments, we discovered that a major component of this effect was due to VEGF, a tumor-derived growth factor responsible for tumor angiogenesis. We have also shown that the level of VEGF in tumor-cell supernatants is directly related to the magnitude of the effect, which proved that CD34+ stem cells express the flt-1 receptor for VEGF. VEGF significantly inhibited the growth of myeloid progenitor cells in the presence of cytokines and growth factors usually sufficient for the normal growth of these cells.[44] We have also demonstrated that VEGF specifically binds to flt-1 receptors on the surface of hematopoietic progenitor cells and that this leads to inhibition of NF-kB activity and to the defect in dendritic cell maturation.[45] Thus, NF-kB appears to be directly involved in the early stages of dendritic cell maturation from hematopoietic progenitor cells and represents a possible target for pharmacologic intervention.

These data indicate that tumor-derived soluble factors, including VEGF, may affect the function of antigen-presenting cells that may, in turn, result in a decreased ability of the immune system to generate effective spontaneous or induced antitumor immune responses. We now have data showing that therapeutic approaches designed to correct these defects significantly improve the natural history of tumors and the efficacy of cancer immunotherapy. The data indicate that these approaches may allow the design of truly effective cancer immunotherapies in the future.

Conclusions

Knowledge of the molecular biology of lung cancer may help in the development and application of specific therapeutics that may not only be more effective, but may be less toxic to normal tissues. Some of these approaches have been reviewed. Wild-type p53 can be effectively introduced into local tumors by recombinant adenoviruses. The function and expression of oncogenes can be directly modulated by antisense or intracellular antibody technologies. Interference of autocrine growth-factor stimulation with gene-based therapeutic approaches, antibodies, or small peptides holds promise for modulation of cancer growth. In addition, the numerous molecularly characterized genetic lesions involved in the development of lung cancer could result in the production of proteins that would be very attractive targets for immunotherapy, if the mutant form could be recognized as distinct from the normal form. We can now relate histological types of lung cancers and even clinically relevant subsets within a histologic type to certain patterns of molecular lesions. It is hoped that early detection, chemoprevention, and rationally derived therapies based on these findings will begin to decrease mortality from this disease in the near future.

References:

1. Everson RB, Randerath E, Santella RM, et al: Quantitative associations between DNA damage in human placenta and maternal smoking and birth weight. J Natl Cancer Inst 80:567-576, 1988.

2. Wistuba II, Lam S, Behrens C, et al: Molecular damage in the bronchial epithelium of current and former smokers. J Natl Cancer Inst 89:1366-1373, 1997.

3. Franklin WA, Gazdar AF, Haney J, et al: Widely dispersed p53 mutation in respiratory epithelium: A novel mechanism for field carcinogenesis. J Clin Invest 100:2133-2137, 1997.

4. Ziegler A, Luedke GH, Fabbro D, et al: Induction of apoptosis in small-cell lung cancer cells by an antisense oligodeoxynucleotide targeting the Bcl-2 coding sequence. J Natl Cancer Inst 89:1027-1036, 1997.

5. Gazdar AF, Carbone DP: The Biology and Molecular Genetics of Lung Cancer. Austin, Tex, R.G. Landes, 1994.

6. Chiba I, Takahashi T, Nau MM, et al: Mutations in the p53 gene are frequent in primary, resected non–small-cell lung cancer. Oncogene 5:1603-1610, 1990.

7. Mitsudomi T, Steinberg SM, Oie HK, et al: ras gene mutations in non–small-cell lung cancers are associated with shortened survival irrespective of treatment intent. Cancer Res 51:4999-5002, 1991.

8. Roth JA, Swisher SG, Merritt JA, et al: Gene therapy for non–small-cell lung cancer: A preliminary report of a phase I trial of adenoviral p53 gene replacement. Semin Oncol 25(3 suppl 8):33-37, 1998.

9. Lee CT, Wu S, Gabrilovich D, et al: Antitumor effects of an adenovirus expressing antisense insulin-like growth factor I receptor on human lung cancer cell lines. Cancer Res 56:3038-3041, 1996.

10. Folkman J: Angiogenesis in cancer, vascular, rheumatoid, and other disease. Nat Med l:27-31, 1995.

11. De Plaen E, Lurquin C, Van Pel A, et al: Immunogenic (tum-) variants of mouse tumor P815: Cloning of the gene of tum-antigen P91A and identification of the tum-mutation. Proc Natl Acad Sci USA 85:2274-2278, 1988.

12. Lurquin C, Van Pel A, Mariamé B, et al: Structure of the gene of tum-transplantation antigen P91A: The mutated exon encodes a peptide recognized with Ld by cytolytic T cells. Cell 58:293-303, 1989.

13. Kawakami Y, Eliyahu S, Jennings C, et al: Recognition of multiple epitopes in the human melanoma antigen gp 100 by tumor-infiltrating T lymphocytes associated with in vivo tumor regression. J Immunol 154:3961-3968, 1995.

14. Cole DJ, Weil DP, Shilyansky J, et al: Characterization of the functional specificity of a cloned T-cell receptor heterodimer recognizing the MART-1 melanoma antigen. Cancer Res 55:748-752, 1995.

15. Wang RF, Robbins PF, Kawakami Y, et al: Identification of a gene encoding a melanoma tumor antigen recognized by HLA-A31-restricted tumor-infiltrating lymphocytes. [Published erratum appears in J Exp Med 181(3):1261, 1995.] J Exp Med 181:799-804, 1995.

16. Topalian SL, Rivoltini L, Mancini M, et al: Human CD4+ T cells specifically recognize a shared melanoma-associated antigen encoded by the tyrosinase gene. Proc Natl Acad Sci USA 91:9461-9465, 1994.

17. Salgaller ML, Weber JS, Koenig S, et al: Generation of specific anti-melanoma reactivity by stimulation of human tumor-infiltrating lymphocytes with MAGE-1 synthetic peptide. Cancer Immunol Immunother 39:105-116, 1994.

18. Kawakami Y, Eliyahu S, Sakaguchi K, et al: Identification of the immunodominant peptides of the MART-1 human melanoma antigen recognized by the majority of HLA-A2-restricted tumor infiltrating lymphocytes. J Exp Med 180:347-352, 1994.

19. Ciernik IF, Berzofsky JA, Carbone DP: Induction of cytotoxic T lymphocytes and antitumor immunity with DNA vaccines expressing single T cell epitopes. J Immunol 156:2369-2375, 1996.

20. Speiser DE, Miranda R, Zakarian A, et al: Self antigens expressed by solid tumors do not efficiently stimulate naive or activated T cells: Implications for immunotherapy. J Exp Med 186:645-653, 1997.

21. Eilber FR, Morton DL: Impaired immunologic reactivity and recurrence following cancer surgery. Cancer 25:362-367, 1970.

22. Hughes LE, MacKay WD: Suppression of the tuberculin response in malignant disease. Br Med J 5474:1346-1348, 1965.

23. Solowey AC, Rapaport FT: Immunologic responses in cancer patients. Surg Gynecol Obstet 121:756-760, 1965.

24. Stiver HG, Weinerman BH: Impaired serum antibody response to inactivated influenza A and B vaccine in cancer patients. Can Med Assoc J 119:733-738, 1978.

25. Ioannides CG, Whiteside TL: T cell recognition of human tumors: Implications for molecular immunotherapy of cancer. Clin Immunol Immunopathol 66:91-106, 1993.

26. Rock KL, Rothstein L, Gamble S, et al: Characterization of antigen-presenting cells that present exogenous antigens in association with class I MHC molecules. J Immunol 150:438-446, 1993.

27. Huang AY, Golumbek P, Ahmadzadeh M, et al: Role of bone marrow-derived cells in presenting MHC class I-restricted tumor antigens. Science 264:961-965, 1994.

28. Steinman RM: The dendritic cell system and its role in immunogenicity. Annu Rev Immunol 9:271-296, 1991.

29. Grabbe S, Bruvers S, Gallo RL, et al: Tumor antigen presentation by murine epidermal cells. J Immunol 146:3656-3661, 1991.

30. Cohen PJ, Cohen PA, Rosenberg SA, et al: Murine epidermal Langerhans cells and splenic dendritic cells present tumor-associated antigens to primed T cells. Eur J Immunol 24:315-319, 1994.

31. Takahashi H, Nakagawa Y, Yokomuro K, et al: Induction of CD8+ cytotoxic T lymphocytes by immunization with syngeneic irradiated HIV-1 envelope-derived, peptide-pulsed dendritic cells. Int Immunol 5:849-857, 1993.

32. Nonacs R, Humborg C, Tam JP, et al: Mechanisms of mouse spleen dendritic cell function in the generation of influenza-specific, cytolytic T lymphocytes. J Exp Med 176:519-529, 1992.

33. Gabrilovich DI, Nadaf S, Corak J, et al: Dendritic cells in antitumor immune responses. II. Dendritic cells grown from bone marrow precursors, but not mature DC from tumor-bearing mice, are effective antigen carriers in the therapy of established tumors. Cell Immunol 170:111-119, 1996.

34. Zitvogel L, Mayordomo JI, Tjandrawan T, et al: Therapy of murine tumors with tumor peptide-pulsed dendritic cells: Dependence on T cells, B7 costimulation, and T helper cell 1-associated cytokines. J Exp Med 183:87-97, 1996.

35. Tas, MP, Simons PJ, Balm FJ, et al: Depressed monocyte polarization and clustering of dendritic cells in patients with head and neck cancer: In vitro restoration of this immuosuppression by thymic hormones. Cancer Immunol Immunother 36:108-114, 1993.

36. Thurnher M, Radmayr C, Ramoner R, et al: Human renal-cell carcinoma tissue contains dendritic cells. Int J Cancer 68:1-7, 1996.

37. Chaux P, Moutet M, Faivre J, et al: Inflammatory cells infiltrating human colorectal carcinomas express HLA class II but not B7-1 and B7-2 costimulatory molecules of the T-cell activation. Lab Invest 74:975-983, 1996.

38. Nestle FO, Burg G, Fah J, et al: Human sunlight-induced basal-cell-carcinoma-associated dendritic cells are deficient in T cell costimulatory molecules and are impaired as antigen-presenting cells. Am J Pathol 150:641-651, 1997.

39. Gabrilovich DI, Ciernik IF, Carbone DP: Dendritic cells in antitumor immune responses. I. Defective antigen presentation in tumor-bearing hosts. Cell Immunol 170:101-110, 1996.

40. Kerrebijn JD, Simons PJ, Tas M, et al: In vivo effects of thymostimulin treatment on monocyte polarization, dendritic cell clustering and serum p15E-like transmembrane factors in operable head and neck squamous cell carcinoma patients. Eur Arch Otorhinolaryngol 252:409-416, 1995.

41. Buelens C, Willems F, Delvaux A, et al: Interleukin-10 differentially regulates B7-1 (CD80) and B7-2 (CD86) expression on human peripheral blood dendritic cells. Eur J Immunol 25:2668-2672, 1995.

42. Gabrilovich DI, Cunningham HT, Carbone DP: IL-12 and mutant p53 peptide-pulsed dendritic cells for the specific immunotherapy of cancer. J Immunother 19:414-418, 1997.

43. Gabrilovich DI, Corak J, Ciernik IF, et al: Decreased antigen presentation by dendritic cells in patients with breast cancer. Clin Cancer Res 3:483-490, 1997.

44. Gabrilovich DI, Chen HL, Girgis KR, et al: Production of vascular endothelial growth factor by human tumors inhibits the functional maturation of dendritic cells. Nat Med 2:1096-1103, 1996.

45. Oyama T, Ran S, Ishida T, et al: Vascular endothelial growth factor affects dendritic cell maturation through the inhibition of nuclear factor-kappa B activation in hemopoietic progenitor cells. J Immunol 160:1224-1232, 1998.