NONSPECIFIC IMMUNOTHERAPY IN A RAT MODEL OF MALIGNANT GLIOMA By CHRISTOPHER LEONARD MARIANI A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2006
Copyright 2006 by Christopher Leonard Mariani
To my parents, Brian and Sharon Mariani, for all that theyâ€™ve done to make it possible for me to reach this point.
iv ACKNOWLEDGMENTS I thank my mentor, Dr. Jake Streit for taking me on as a graduate student and allowing me to do this work in his laborat ory. I thank my committee members; Drs. Steve Blackband, Laurence Morel, and Dietma r Siemann, for constructive criticism, helpful discussions and for keeping me on tr ack to graduate within a reasonable time frame. I would also like to thank my colleagues in the Streit Lab; Jessica Conde, Josh Stopek, Barry Flanary, Sarah Fendrick, Kelly M iller, Kryslaine Lopes, and eMalick Njie for help with techniques, scientific disc ussions, and commiserative discourse over an occasional frosty beverage. I thank Sean Kearns and Chris Futtner for friendship throughout my time in the IDP and help with in termittent scientific side-excursions. I thank my former mentor, Cheryl Chrisman, fo r continuing career s upport and for always providing helpful advice. Finall y, I would like to thank my wife , Rita Hanel, her parents, Harry and Jean Hanel, and my family for their love and support.
v TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES.............................................................................................................ix LIST OF FIGURES.............................................................................................................x ABSTRACT......................................................................................................................xii CHAPTER 1 BACKGROUND AND LI TERATURE REVIEW......................................................1 Malignant Gliomas: An Overview...............................................................................1 The Immune System: A General Overview.................................................................2 The Innate versus the Adaptive Immune System..................................................2 The Major Histocompatability Comp lex and Antigen Presenting Cells...............3 Cross-Presentation.................................................................................................4 Dendritic Cell Migration and Maturation..............................................................5 Lymphocyte Populations.......................................................................................5 The T cells and B cells...................................................................................5 Natural killer (NK) cells.................................................................................6 Gamma delta ( ) T cells...............................................................................7 Regulatory T cells..........................................................................................7 Toll-like Receptors................................................................................................8 Immunologic Considerations in the Central Nervous System....................................10 Immune Privilege and Antigen Drainage............................................................10 Microglia as Immunocompetent Central Nervous System Cells........................11 Tumor-Immune System Interactions..........................................................................13 Tumor Immunosurveillance and Immunoediting................................................13 Tumor-Associated Immune Cells........................................................................14 Escape of Tumors from the Immune System......................................................15 Immunotherapeutic Strategies for Cancer..................................................................16 Passive (Adoptive) Immunotherapy for Gliomas................................................16 Active Specific Immunotherapy for Gliomas......................................................17 Active Nonspecific Immunotherapy for Gliomas...............................................18 Historical use of nonspecifi c immunotherapy for cancer.............................19 Microglial responses to nonspecific immunotherapy...................................21
vi 2 IMMUNE SYSTEM REJECTION OF ESTABLISHED BRAIN TUMORS: A PROOF OF PRINCIPLE FOR IMMUNOTHERAPY...............................................23 Introduction.................................................................................................................23 Materials and Methods...............................................................................................24 Animals................................................................................................................24 Cell Lines, Culture and Tumor Implantation......................................................25 Animal Observation and Clinical Examination...................................................25 Magnetic Resonance Imaging.............................................................................26 Histology and Immuno histochemistry................................................................27 Statistical Analysis..............................................................................................28 Results........................................................................................................................ .28 Observation and Clinical Examination................................................................28 Magnetic Resonance Imaging.............................................................................29 Histology and Immuno histochemistry................................................................29 Long-Term Survival Studies...............................................................................32 Discussion...................................................................................................................32 3 NONSPECIFIC IMMUNOTHERAPY IN SUBCUTANEOUS RG-2 GLIOMAS...45 Introduction.................................................................................................................45 Lipopolysaccharide..............................................................................................46 Zymosan A..........................................................................................................46 Granulocyte-Macrophage Co lony Stimulating Factor........................................46 Interferon-gamma................................................................................................47 Routes of Administration: Intratum oral versus Systemic Treatment.................48 Materials and Methods...............................................................................................49 Animals................................................................................................................49 Cell Lines, Culture and Tumor Implantation......................................................49 Animal Observation and Tumor Measurement...................................................49 Tumor Treatments...............................................................................................50 Histology and Immuno histochemistry................................................................50 Statistical Analysis..............................................................................................52 Results........................................................................................................................ .52 Lipopolysaccharide..............................................................................................52 Zymosan A..........................................................................................................54 Granulocyte-Macrophage Co lony Stimulating Factor........................................54 Lipopolysaccharide Combin ed with Interferon................................................54 Morbidity and Mortality Unrelated to Treatment................................................55 Athymic Nude Rat Studies..................................................................................55 Rechallenge Studies.............................................................................................56 Histology and Immuno histochemistry................................................................56 Discussion...................................................................................................................58
vii 4 COMBINATION RADIATION AND IMMUNOTHERAPY IN SUBCUTANEOUS RG-2 GLIOMAS..................................................................75 Introduction.................................................................................................................75 Materials and Methods...............................................................................................77 Tumor Treatments...............................................................................................77 Radiation Dose-Response Study.........................................................................78 Combination Treatment Study............................................................................78 Histology and Immuno histochemistry................................................................78 Statistical Analysis..............................................................................................79 Results........................................................................................................................ .79 Radiation Dose-Response Study.........................................................................79 Combination Treatment Study............................................................................79 Rechallenge Studies.............................................................................................80 Morbidity and Mortality Unrelated to Treatment................................................80 Histology and Immuno histochemistry................................................................80 Discussion...................................................................................................................81 5 NONSPECIFIC IMMUNOTHERAPY WITH OR WITHOUT RADIATION IN INTRACRANIAL RG-2 GLIOMAS.........................................................................85 Introduction.................................................................................................................85 Materials and Methods...............................................................................................86 Animals................................................................................................................86 Tumor Implantation.............................................................................................86 Tumor Treatments...............................................................................................88 Radiation Treatment and Combination Therapy.................................................88 Vaccination with Irradiated Tumor Cells............................................................89 Rechallenge Studies.............................................................................................89 Magnetic Resonance Imaging.............................................................................90 Histology and Immuno histochemistry................................................................90 Survival Calculation and Statistical Analysis......................................................90 Results........................................................................................................................ .91 Treatment of Intracranial Tumors with Intratumoral Nonspecific Immunotherapy Alone.....................................................................................91 Lipopolysaccharide and Interferon............................................................91 Lipopolysaccharide......................................................................................91 Radiation Dose-Response Experiment for Intracranial RG-2 Tumors................92 Combination Lipopolysaccharide and Radi ation Therapy for Intracranial RG-2 Tumors...................................................................................................93 Treatment of tumors on Day 9.....................................................................93 Treatment of tumors on Day 3.....................................................................94 Treatment of Intracranial RG-2 Glio mas with Subcutaneous Administration of Irradiated Tumor Cells and Lipopolysaccharide.........................................95 Rechallenge Studies.............................................................................................95 Histology and Immuno histochemistry................................................................96 Discussion...................................................................................................................97
viii 6 CONCLUSIONS......................................................................................................115 LIST OF REFERENCES.................................................................................................119 BIOGRAPHICAL SKETCH...........................................................................................150
ix LIST OF TABLES Table page 3-1 Intratumoral lipopolysaccharide for s ubcutaneous RG-2 gliomas: Treatment outcomes...................................................................................................................74 3-2 Intratumoral zymosan A for subcutane ous RG-2 gliomas: Treatment outcomes...74 3-3 Intratumoral granulocyte-macr ophage colony stimulating factor for subcutaneous RG-2 gliomas: Treatment outcomes.................................................74 4-1 Single-dose external beam radiation therapy for subcutaneous RG-2 gliomas: Treatment outcomes.................................................................................................83 4-2 Combination single-dose external beam radiation therapy and intratumoral lipopolysaccharide for subcutaneous RG -2 gliomas : Treatment outcomes...........84 5-1 Combination single-dose external beam radiation therapy and intratumoral lipopolysaccharide for intr acranial RG-2 gliomas: Treatment outcome...............105 5-2 Intracranial rechallenge e xperiment of rats with reje ction of subcutaneous RG-2 tumors.....................................................................................................................109
x LIST OF FIGURES Figure page 1-1 Immune system overview.........................................................................................22 2-1 An RG-2 glioma in a F344 rat (MR images)...........................................................37 2-2 An RG-2 glioma in a Wistar rat (MR images).........................................................38 2-3 Cresyl violet stai ning of RG-2 gliomas....................................................................39 2-4 Immunohistochemical staining of RG -2 gliomas for CD11b, MHC I and MHC II in F344 and Wistar rats............................................................................................40 2-5 Immunohistochemical staining of RG-2 gliomas for CD6, CD8, and CD4 in F344 and Wistar rats................................................................................................42 2-6 Quantification of intratumoral and peritumoral inflammatory cells in RG-2 gliomas.....................................................................................................................43 3-1 Differences in RG-2 glioma behavior in male versus female rats...........................66 3-2 Preliminary high dose LPS experiment in male rats with subcutaneous RG-2 gliomas.....................................................................................................................66 3-3 Lipopolysaccharide dose response experime nt in male rats with subcutaneous RG-2 gliomas...........................................................................................................67 3-4 Zymosan A treatment of s ubcutaneous RG-2 gliomas.............................................67 3-5 Granulocyte-macrophage colony stimulati ng factor treatment of subcutaneous RG-2 gliomas...........................................................................................................68 3-6 Combination LPS and IFNtreatment of subcutaneous RG-2 gliomas.................68 3-7 Treatment of subcutaneous RG-2 glioma s in male athymic nude rats (rnu/rnu).....69 3-8 Immunohistochemical evaluation of a s ubcutaneous RG-2 glioma (low power)....70 3-9 Immunohistochemical staining in a subc utaneous RG-2 glioma (higher power)....71 3-10 Immunohistochemical staining in a subc utaneous RG-2 glioma (higher power)....72
xi 3-11 Immunostaining for MHC II in a subcutaneous RG-2 glioma.................................73 3-12 Immunostaining for CD86 in a subcutaneous RG-2 glioma....................................73 4-1 Dose-response experiment of subcutane ous RG-2 gliomas treated with external beam radiation therapy (EBRT)...............................................................................83 4-2 Combination therapy with EBRT and LPS in rats with subcutaneous RG-2 gliomas.....................................................................................................................84 5-1 An intracranial RG-2 glioma treated with IT LPS and IFN(MR images)..........101 5-2 Intratumoral LPS and IFNcombination treatment of intracranial RG-2 gliomas...................................................................................................................102 5-3 Intratumoral LPS treatment of intracranial RG-2 gliomas.....................................102 5-4 Dose-response experiment of intracrani al RG-2 gliomas treated with radiation therapy....................................................................................................................103 5-5 Intratumoral radiation therapy and L PS combination treatment of intracranial RG-2 gliomas.........................................................................................................104 5-6 Combination radiation therapy and IT LPS treatme nt of intracranial RG-2 gliomas...................................................................................................................106 5-7 Intracranial RG-2 gliomas treated with LPS and radiation versus radiation alone (MR images)...........................................................................................................107 5-8 Treatment of intracranial RG-2 glioma s with subcutaneous LPS and irradiated tumor cells..............................................................................................................108 5-9 Intracranial rechallenge experiment in rats rejecti ng previous subcutaneous RG2 gliomas................................................................................................................111 5-10 Immunohistochemical evaluation of an intracranial RG-2 glioma (low power)...112 5-11 Immunohistochemical evaluation of an intracranial RG-2 glioma (higher power).....................................................................................................................113 5-12 Immunohistochemical evaluation of an intracranial RG-2 glioma (higher power).....................................................................................................................114
xii Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy NONSPECIFIC IMMUNOTHERAPY IN A RAT MODEL OF MALIGNANT GLIOMA By Christopher Leonard Mariani December, 2006 Chair: Wolfgang J. Streit. Major Department: Medical Sciencesâ€”Neuroscience Malignant gliomas are the most common primary brain tumors in humans and unfortunately, also the most aggressive. Current thera py prolongs survival, but the response is temporary, and these tumors inevita bly recur within a relatively short period of time. As a result, new therapies for thes e devastating cancers are desperately needed. One such therapy is nonspecific immunotherapy (NSI), which attempts to encourage cells of the immune system to attack and eliminate tumor cells. We evaluated NSI in subcutaneous and intr acranial rat models of malignant glioma, using the RG-2 cell line, which is syngeneic to the Fisher 344 rat. Subcutaneous implantation of RG-2 tumors was used initially as a screening proce dure in the evaluation of four different NSI protocols delivered in tratumorally (IT): li popolysaccharide (LPS), zymosan A (ZymA), granulocyte-macrophage colony stimulating factor (GM-CSF), and LPS in combination with interferon. In this model, IT LPS and ZymA prolonged survival compared to that of control rats. Multiple doses were supe rior to a single dose
xiii and led to complete tumor regression in mo st rats. The GM-CSF showed no anti-tumor effects. The subcutaneous model was then used to assess a combination of radiation treatment with IT LPS. Tumors showed a dos e-dependent delay in growth with radiation therapy and combinations of ra diation and IT LPS worked s ynergistically to delay tumor growth and to prolong surviv al in this model. Finally, the intracranial RG-2 mode l was used to evaluate several immunotherapeutic protocols, in cluding intracranial treatment with IT LPS, intracranial treatment with a combination of radiation a nd IT LPS, and treatment with subcutaneous administration of irradiated RG-2 cells a nd LPS. The protocols used did not show prolongations of survival, and the administrati on of intracranial LPS was associated with considerable toxicity. These findings show that IT thera py with certain microbially-derived immunostimulants may have anti-tumor eff ects, and may work synergistically with radiotherapy. However, methods of circumve nting the toxicity s een with intracranial administration must be developed before thes e therapies may be successful in clinical patients.
1 CHAPTER 1 BACKGROUND AND LITERATURE REVIEW Malignant Gliomas: An Overview Gliomas are tumors arising from cells of g lial origin, and are classified according to their presumed cell of origin and apparent aggressiveness. The most frequently recognized tumors are oligodendrogliomas and astrocytomas, and a grading system delineates grade I tumors (reserved for the benign pilocytic astrocytoma) from grade II (low-grade) and grade III (anaplastic) tumors. The grade IV designation is reserved for glioblastoma multiforme (GBM), the most aggressive form of astrocytoma, which is also unfortunately the most common primary brain tumor diagnosed in human patients. The current therapeutic standard of care consists of surgical excision followed by fractionated radiation therapy. Many patient s also receive adju nctive chemotherapy either at the time of diagnosis or at tumor recurrence. Despite advances in these treatment modalities, the prognosis for intracranial malignant gliomas, and GBM in particular, remains dismal (13). The median survival time is approximate ly 12 months for patients with GBM, and 36 months for patients with an aplastic astrocytoma. Metastasis of malignant gliomas is a rare event, and treatment failure typically occurs in these patients because of local r ecurrence. Glioma cells have the ability to invade surrounding neural tissue, and histol ogical studies have shown neoplastic cells within seemingly normal brain tissue up to 4 cm from the main tumor mass (3). These satellite cells are inaccessible to current surgical and radiation treatments without causing profound side effects, and eventually result in regrowth of the tumor.
2 As a result of these therapeutic shortcom ings, novel treatments for these devastating tumors are desperately needed. One nove l category of trea tment is known as immunotherapy. Immunotherapy may be broadl y defined as manipula tion of the immune system in order to retard or halt the growth of tumors, or ideally, lead to their eradication. Immunotherapy holds enormous promise for patie nts with glioma, because if the immune system can be convinced to recognize the tumo r as foreign (or dangerous), then it has the potential to seek out and eliminate cells re mote from the main tumor mass. Different strategies for immunotherapeu tic treatment of malignant gl iomas are discussed below. The Immune System: A General Overview The Innate versus the Adaptive Immune System The immune system can be divided into innate and adaptive components. The innate system is considered to be evolu tionarily primitive, and comprises cells and immune molecules prepared to defend the body against invasion without prior sensitization. It consists of cells such as neutrophils, natural killer (NK) cells, local tissue macrophages, and the protein complement. The adaptive immune system is composed mainly of lymphocytes and their products, which are able to recognize a multitude of foreign antigens and generate a powerful, fo cused response against these antigens, if properly sensitized. Lymphocytes have the ab ility to produce long-lived memory cells, which can recognize previously processed anti gen long after it has been cleared from the body, and produce a robust immune response on re-exposure to these antigens. Dendritic cells (DC) and other antigen presenting cells (APC) function to connect the innate and adaptive immune systems through the capture of foreign antigen and presentation to cells of the adaptive immune system.
3 The Major Histocompatability Complex and Antigen Presenting Cells Major histocompatability complex (MHC) pr oteins are an integral part of the immune system. They are cell-surface molecule s that are associated with fragments of protein from various sources, and are typically designated as MHC I and MHC II. Most cells of the body (with the notable exception of cells within the br ain at rest) express MHC I on their surface. Associated with th ese MHC I molecules are peptide fragments produced from endogenous sources within these cel ls. MHC II, in contrast, is typically only expressed on cells that are specialized to process antigens and influence decisions on immune acceptance or rejection. Antigen presenting cells are cells of the immune system that can obtain antigen from the surrounding environment, process this antigen, and then present it to T cells to bring about an effective immune respons e. Antigen may be obtained from the environment through various mechanisms, incl uding phagocytosis of dead or dying cells, macropinocytosis, and adsorptive endocytosis (4, 5). Antigenic proteins are then broken down within endosomes and lysosomes, conj ugated onto MHC molecules, and moved to the plasma membrane for interaction with th e appropriate T cells (4) . Although several cell types have been shown to be able to present antigen to T cells, including macrophages and B cells, DC are considered to be the most efficient at this process (4). Presentation of antigen in the context of MHC is not sufficient for induction of an immune response. In fact, presentation in this situation without a second stimulatory signal leads to T cell anergy, a state of func tional paralysis agains t the antigen. The second signal is provided by costimulatory molecules on the surface of the APC, which include CD80 and CD86 (5). These interact with CD28 on the surf ace of the T cell to
4 lead to full T cell activation. Upregulati on of these costimulatory molecules occurs during the maturation process in dendritic cells (see below). Classical immunologic theory holds that endogenous-source peptides in the cytoplasm are moved into the endoplasmic reticulum, conjugated to MHC I molecules and presented on the surface of most cells of the body; while exogenous antigen (e.g., from microbes or other cells) is taken up by APC, conjugated to MH C II molecules in the endosomal-lysosomal compartment, and then moved to the plasma membrane (4). The MHC II complex on APC binds to CD4, present on the surface of â€œhelperâ€ T cells, which then help to generate an immune response (eit her stimulation or tole rance). In contrast, MHC I binds to CD8, typically on the surface of cytotoxic killer cells. Foreign antigen present in the context of MHC I (such as in a virally-infect ed cell) would elicit a CD8 T cell response, and subsequent destru ction of the infected cell. Cross-Presentation The classical view of endogenous versus exogenous antigen processing is not absolute, however. It is now known that exogenous antigen captu red by APC can enter the cytoplasm and then follow endogenous an tigen pathways, and be presented in the context of MHC I (6). This process (know n as cross-presentation) (7, 8), has been documented to occur in DC (9-13) and macr ophages (14, 15), and may result in either activation of the immune system against the antigen in question (c ross-priming), or the development of tolerance agains t this antigen (cross-tolerance) (7). This result depends largely on the presence of costimulatory molecules on the surface of the APC, including CD80 (B7.1) and CD86 (B7.2). Cross-priming ha s been documented to occur for antigen from a number of sources, including viruses and tumor cells (8, 16-19).
5 Dendritic Cell Migration and Maturation Dendritic cells are present in many orga ns of the body, including the skin, liver, kidney, respiratory tract, and in testine (4). Here they reside in an immature state, awaiting contact with foreign organisms or antigen. They exhib it a stellate morphology, with cellular processes extending into the su rrounding tissue, presumably to sample the environment for antigen or cells (5). Interaction with a vari ety of stimulatory substances leads to maturation of the dendritic cells, characterized by upregulation of MHC II and the costimulatory molecules CD80 and CD86 on the surface of the ce ll and migration to lymphoid organs such as regional lymph nodes or the spleen (4, 5). Whereas immature dendritic cells are uniquely equipped to captu re antigen from the environment through the mechanisms described above, mature DC downr egulate this ability as they travel to lymphoid areas and prepare to pres ent antigen to T cell populations. Lymphocyte Populations Lymphocytes have classically been divi ded into T cells and B cells, although several other populations have been described (NK cells, natural killer T (NKT) cells and others). In addition, certain T cell subsets may play important and distinct roles in antitumor immunity, including T cells and regulatory T cells. The T cells and B cells The B cells originate in the bone marrow; e xpress a B cell receptor on their surface; and when stimulated with antigen, differentia te into plasma cells and express large amounts of antibody. The T cells originate in the bone marrow, develop in the thymus, and express a T cell receptor (TCR) on their surface. In the thymus, they undergo a selection process to remove cells with a high affinity for self-proteins (central tolerance). Nave T cells circulate through the blood, ly mph, and lymphoid organs and generally
6 express either CD4 or CD8. The CD8 cells f unction as cytotoxic killer cells, with the capacity to destroy target cel ls in a controlled manner th rough the release of molecules such as perforin and granzyme (20). As mentioned above, CD8 binds to MHC I molecules on the surface of APCs and target cells. The CD4 positive lymphocytes are known as he lper T cells, as they help to direct the immune response in different directions. After interaction w ith MHC II molecules on the surface of an APC, helper T cells may differe ntiate into either type 1 (Th1) or type 2 (Th2) helper cells. This differentiation depe nds on the T cell-APC interaction, density of peptides presented, types of costimulatory molecules expressed, cytokine secretion, and the local milieu surrounding the cells (21). A Th2 response is promoted by interleukin (IL)-4, and supports the producti on of an antibody response and humoral immunity. Th2 clones secrete IL-4, IL-5, IL-10 and IL13. In contrast, a Th1 response supports cell-mediated immunity, and is considered important for a response to intracellular pathogens and tumors. Cytokines favoring a Th1 response include IL-12 and Th1 clones secrete IL-2 and interferon (IFN)(22, 23). Natural killer (NK) cells Natural killer cells are large lymphocytes considered to be part of the innate immune system. They do not express T or B cell receptors and do not depend on thymic processing for their development. NK cells recognize target cells us ing several different mechanisms, including both inhibitory and activating interactions. Surface receptors binding cells with MHC I generate an inhibi tory signal within the NK cell, and cells failing to express MHC I molecules are ofte n targeted for killing. In addition, the NKG2D receptor and other surface molecules are capable of recognizing aberrant proteins present on the surface of stressed, transformed or infected cells, such as the
7 MHC class I chain-related proteins A and B (MICA and MICB) (24). Similar to T cells, NK cells kill their targets th rough the release of perforins and granzymes and expression of Fas ligand (20, 25). NK cells were first iden tified based on their role in anti-tumor surveillance but also act to de stroy other abnormal cells such as virally infected cells or those present in allografts or xenografts. Gamma delta ( ) T cells The majority of T cells have a TCR comp rised of protein subunits termed the and subunits. However, a small subset has alternate subunits: and , and these are known as T cells. These cells have several unique functions in immunologic defense that are more akin to an innate rather th an an adaptive response. Unlike T cells, T cells do not recognize processed peptide antigens presented by APCs on MHC molecules. Similar to NK cells, T cells can recognize microbial and tumor antigens on the surface of infected, stressed or transformed cells and generate an immunologic response against these cells without prior antigen exposure or priming (26, 27). T cells are commonly found in the skin and other epithelial tissu es, where they exhibit a dendritic morphology, but have been documented in pathologic lesi ons of the central nervous system, including multiple sclerosis (28). Regulatory T cells The existence of a population of T cells that downregulate the immune response was proposed decades ago and has generated c onsiderable interest in recent years. Previously known as suppressive T cells, th ese cells are now generally referred to as regulatory T cells. This subset is CD4 and CD25 positive, and its development is controlled by the forkhead box P3 (FOXP3) transcription factor (29, 30). Classic CD25+CD4+FOXP3+ regulatory T cells are thymus de rived, occur naturally and comprise
8 5 to 10% of peripheral CD4 positive T cells in normal mice and humans (31). However, a number of other regulatory T cell populations are now believ ed to exist, which may be induced to form in the periphery, and have been termed adaptive regulatory T cells. These cells may have varying expression of di fferent surface markers; for example, some are CD8+ (31). Regulatory T cells appear to utilize multiple mechanisms in suppressing the immune response, including those employing direct cell-ce ll contact as well as the secretion of soluble factors ( 31, 32). Several different targ et cells might be involved, including effector T cells and APCs. As a result, regulatory T cells appear to be important in the induction of peripheral tolerance to self-antigens. As many tumor-associated antigens are a lterations of self-antigens, it was hypothesized that these cells may play a role in tumor immune syst em evasion, and recently, regulatory T cells have been shown to play important role s in contributing to the immunosuppressive environment of a variety of different tumors (31). Regulatory T cells are present in increased numbers in the circulation of patients with a variety of different cancers (31, 33, 34), including those with GBM (35) and comprise the majority of CD4+ lymphocytes in tumor samples. Se lective depletion of these cells has been shown to increase surviv al in a number of mouse tumor models (36) and their presence in some human cancers has been correlated with redu ced survival (37). Toll-like Receptors Toll-like receptors (TLRs) are pattern-r ecognition receptors present on the cell surface or within lysosomal and endosomal co mpartments of immune cells that trigger immune responses. At least 11 TLRs have been identified in mammals (38). They bind to a variety of different microbial products , including peptidoglycan from gram-positive
9 bacteria (TLR2), zymosan A from yeas t (TLR2), double-stra nded RNA (TLR3), lipopolysaccharide (TLR4), fl agellin (TLR5), single-s tranded RNA (TLR7,8) and bacterial DNA with unmethylated CpG motif s (TLR9) (39). In addition, some endogenous molecules, such as heat shock prot eins, may be able to activate TLRs as well (38, 39). The TLRs are associated with several di fferent adapter molecules, and ligation initiates a signal transduction cascade that t ypically leads to the upr egulation of a number of genes promoting inflammatory and immunologi c responses. Portions of this activation pathway are shared among different TLRs and ot her pro-inflammatory receptors, such as the IL-1 and TNFreceptors (38, 40). A common result of ligation of these receptors is activation of NFB in the cytoplasm, translocation to the nucleus and upregulation of a variety of different inflammatory gene produc ts (38, 40). However, discrimination of the response of the cell does occur based on wh ich TLR is activated, and a number of pathways and signaling molecules have been identified. For example, the ligation of TLR4 with lipopolysaccharide (LPS) can activate NFB through a signaling molecule known as myeloid differentiation primary -response protein 88 (MyD88), and this mechanism is shared by several other TLRs, including TLR2 and TLR9, as well as IL-1 (41). Macrophages from MyD88-deficient mice show no NFB activation immediately after lipoprotein or CpG administration but de monstrate a delayed activation after LPS as a result of an alternate pathway (41-43). This MyD88-independent pathway is also utilized by TLR3 signaling (38). The response to TLR ligation is controlled at a number of different levels. Some TLRs can form heterodimers that recognize specific microbial products (e.g., TLR 1/2
10 and TLR 2/6). Further specificity of the resulting immune response is provided by differential location of TLRs within the cel l. While TLRs 1, 2, and 4 are located on the surface of the cell, th e receptors recognizing nucleic-aci d-like structures, TLRs 3, 7 and 9, are located within endosomal or lysosomal compartments (38). In addition, different immune cell populations express different TLRs, adding another layer of complexity to the system (44). Finally, ligation of th e same TLR on two different cell types may produce distinct outcomes and downstream cellular responses (44). In addition to the production of infla mmatory gene products, TLR ligation on dendritic cells can lead to maturation of these cells with upregul ation of CD40, CD80 and CD86. Chemokine receptor expression also changes, encouraging migration of the dendritic cell to draining lymph nodes in prepar ation for stimulation of nave T cells (44). This has been noted with both TLR2 and TLR4 ligands (38). Thus, TLR ligands provide critical signals in the connection of the innate and adaptive immune systems Immunologic Considerations in the Central Nervous System Immune Privilege and Antigen Drainage The brain has traditionally been regarded as an â€œimmunologically privilegedâ€ site, with a limited ability to mount an immunol ogic response to antigens present within the central nervous system (CNS) parenchyma. Th is view has been developed based mainly on reports describing long-term survival of allogeneic or xenogeneic tumors or extraneural tissue after intracranial implantation (45-47). Potential mechanisms contributing to this immunologi c privilege include tight j unctions between endothelial cells within the brain comprising the â€œbloodbrain barrierâ€, lack of a CNS lymphatic system, absence of MHC expression by cells w ithin the brain, exclusion of T cells from the brain parenchyma and possibly the expression of Fas ligand (FasL) within the brain.
11 Dendritic cells, although present within the meninges and choroid plexuses, are absent from the brain parenchyma (48). However, despite these factors, it is clear that the immune privilege of the brain is far from complete (49). Although MHC expression is absent from th e normal brain in a quiescent state, MHC molecules can be upregulat ed on various cells of the br ain, including astrocytes and microglia, and microglia have the ability to perform many of the functions of DC and macrophages present in peripheral tissues (see below). Although it lacks a separate lymphatic system, antigen within the brain can drain/move through perivascular spaces and collect in the nasal lymphatics (50). In this way, CNS antigen can generate a peripheral immunologic response. The superf icial and deep cervical lymph nodes have been shown to be an important site for this response (51, 52). In addition, although nave T cells are typically excluded, T cells that have been activated in the periphery are able to efficiently enter and patrol the brain (53). Microglia as Immunocompetent Central Nervous System Cells In addition to neurons, astrocytes and oligodendrocytes, several immune cell populations of mesodermal origin are present within the CNS. The meninges and choroid plexus contain populations of macrophages a nd DC (48). There is a population of perivascular macrophages (or pe rivascular microglia) within the basal lamina of brain capillaries. Finally, parenchymal microglia are found throughout the brain parenchyma, where they comprise 5-20% of the non-neurona l cell population (54). These cells exist in a highly ramified state in the normal brain at rest, likely reflecting their ability to monitor and sample the local environment. The orig ins and differentiation of these cell types have been the subject of cons iderable controversy in the past (55), although most now
12 agree that microglia are derived from CD45+ bone marrow derived myeloid precursors that invade and colonize the fetal brain (56). The perivascular and parenc hymal populations of microglia appear to differ in a number of respects (57). Whereas the form er express relatively high levels of CD45, parenchymal microglia are CD45low (58). Perivascular cells stain positive with the RMG-1, and ED2 antibodies, while parenchymal microglia do not (59) and perivascular cells are more likely to express MHC II and costimulatory molecules (56, 60). Chimeric studies in irradiated adult rats show th at the perivascular population is regularly replenished by bone marrow precursors, while only a very small number of parenchymal cells appear to be derived th is way (61, 62). Parenchymal mi croglia have the ability to divide and renew their p opulations within the CNS compartment (63). Although classically considered as a local macrophage popul ation akin to resident macrophage populations in other organs, micr oglia appear to possess more functional plasticity than these other populations, and ma y be capable of differentiation into DC-like or macrophage-like phenotypes, depending on the environment (56). They are capable of performing many of the functions of DC a nd macrophages found in other organs, albeit with some notable differences. Microglia are capable of secr eting a variety of pro-inflammatory cytokines with appropria te stimulation, including IL-1, IL-6, IL-12, and TNF(64-67). They can express molecules important in antigen presentation such as MHC I and II and the costimulatory molecules CD40, CD80 and CD86 (54, 68-70). In addition, activated microglia can prime allor eactive T cell responses and stimulate T cell lines to proliferate and produce cytokine s (71-74) and function as the primary antigen-presenting cells of th e CNS (75-79). However, they do not appear to be as
13 efficient at antigen presentation or the pr oduction of a primary T cell response as DC, although their ability to restimul ate effector T cells was show n to be equivalent (75, 80, 81). Tumor-Immune System Interactions Tumor Immunosurveillance and Immunoediting The original hypothesis that the immune sy stem could detect and eliminate tumor cells was put forth by Burnet and Thomas in the 1950s (82-84). They postulated that tumor cells arising in an organism may possess new antigenic variants that will be recognized by the immune system, leading to de struction of the tumo r, often without any â€œclinical hint of its existenceâ€ (82). This theory was extensively tested but ultimately abandoned after studies in nude mice failed to show an increased susceptibility to spontaneous or carcinogen-i nduced tumors (85). After several decades, the theory has been revived and expanded, with the realization that nude mice are not completely immunodeficien t and the development of a variety of selectively immunodeficient anim als through the knockout of certain genes, such as RAG-2-/-, perforin-/-, IFN-/and IL-12-/mice, which do have an increased susceptibility to tumors (85). In additi on, the identification of additional lymphocyte subtypes such as NK cells and T cells provided other possible candidates for surveillance duty. Finally, studies in humans have shown that patients immunosuppressed after organ transplantation or secondary to HIV infection are at increased risk for the developmen t of certain cancers (85). However, tumors form in immunocom petent individuals despite tumor immunosurveillance. Newer work provides ev idence that as tumors develop and grow, they are â€œsculptedâ€ by the immune syst em, which selects for the growth of
14 non-immunogenic tumor variants. Thus, the te rm â€œimmunoeditingâ€ has been proposed in lieu of immunosurveillance to reflect this complex tumor-immune system dynamic (85). In this theory, tumor cells are first identif ied and eliminated by the immune system (surveillance); followed by a period of equilibrium, during which immunoresistant cells are being selected for. Finally, there is pr eferential growth of th ese cells and escape of the tumor from immune attack. Tumor-Associated Immune Cells Evidence for a potential anti-tumor immune response can be seen in brain tumors, as a variety of inflammatory cells have b een shown to invade these tumors, including neutrophils, lymphocytes, macrophages a nd microglia (86-91) . Microglia and macrophages infiltrate glial tumors produced experimentally in laboratory animals (89, 92-94) and occurring naturally in humans ( 91, 95-99) and may comprise up to 20 to 50% of all cells found in glioma biopsy samples (96) . In addition, microgl ia also proliferate within and around gliomas (89, 100). However, whether or not this cellular inf iltration represents an immunologic attack on the tumor is debatable. Glioma cells produce a number of molecules that may be responsible for recruiting microglia and macr ophages to the site of the tumor, including macrophage chemoattractant protei n-1 (MCP-1) (86, 101), granulocyte colony-stimulating factor (G-CSF) (102) and hepatocyte growth factor/scatter factor (103). In addition, there is evidence to sugge st that microglia themselves may perform pro-glioma functions. For example, microglia appear to be responsib le for the production of a large portion of the IL-10 produced by gliomas (104, 105) and also secrete TGF(106, 107) which not only contribute to th e immunosuppressive milieu but may play a
15 role in promoting the invasion of gliomas (104, 108). Microglia may also have a role in causing apoptosis in infiltrati ng T cells through Fas-Fas ligand interactions (89, 109). Several authors have correlated the pres ence of large numbers of intratumoral lymphocytes with a favorable prognosis (110 -113), although this re mains controversial (114). Thus, it seems that glioma antigen may elicit a systemic immune response, likely concentrated in the cervical lymph nodes, wh ich can generate activat ed lymphocytes that enter the CNS parenchyma and infiltrate th e tumor tissue. However, despite this response, regression of gliomas, at least once identif ied in clinical patients, are extremely rare events. The reasons for this discrepa ncy are likely multifactorial, and center around the immunosuppressive environment present within the tumor. Escape of Tumors from the Immune System Despite the infiltration of large numbers of immune cells into malignant gliomas, these cells are not effective in controlli ng these malignancies (86, 90, 92, 95, 115). This is likely attributable to the immunosuppressive environment created by the glioma (116). Glioma cells express low levels of MHC I (116) and secrete a variety of immunosuppressive factors, including TGF, IL-10, and prostaglandin E2 (104, 116121). Glioma culture supernatant can dow nregulate MHC II expr ession on microglial cells and despite large numbers of macrophage s and microglia, higher grade astrocytomas (i.e., more aggressive) have fewer number s of MHC II positive cells (70, 90). TGFsuppresses the activation and proliferation of microglia in vitro , and IL-10 can inhibit MHC II expression by microglia (122, 123). Peripheral blood leukocytes have been documented to have reduced functional activity in patients with glioma (124).
16 The expression of death receptor ligands on the surface of tumor cells may be another mechanism potentially employed by gliomas to elude the immune system (116, 125, 126). The Fas-Fas ligand (FasL) system is utilized in the regul ation of lymphocyte populations, and some have postulated that upregulation of FasL on tumor (127, 128) or microglia (109) cell surfaces may kill invading lymphocytes (the so-called â€œtumor counterattackâ€). Another possible countera ttack mechanism involves CD70 expression by glioma cells leading to apoptosis of CD27-expressi ng lymphocytes(126). Immunotherapeutic Strategies for Cancer Immunotherapy may be categorized as passi ve or active. Passive or adoptive immunotherapy involves the administration of antibodies or various immune effector cells (e.g., T cells, lymphokine activated killer [LAK] cells) from a donor into a recipient in order to elicit an anti-tumor effect. Active immunothe rapy involves stimulation of the patientâ€™s own immune system in order to genera te similar effector cells or molecules, and may be further categorized as specific or nonspecific. In specific immunotherapy, the intended immune response is focused on a part icular molecule or protein of interest, typically one that is expressed on the surface of the tumor cells. Antibodies or cells (e.g., CD8+ cytotoxic T cells) are then generated ag ainst this molecule, ideally leading to a specific anti-tumor effect. Nonspecific i mmunotherapy (NSI) involves the administration of cytokines or microbial products or analogs that have a stimulatory effect of certain immune cells without targeting a sp ecific tumor antigen or molecule. Passive (Adoptive) Immunotherapy for Gliomas The administration of antibodies or effector cells has been evaluated in a number of preclinical (129-131) and clinical trials (132 ). The infusion of in vitro activated T cells, in particular, has been successful in media ting tumor regression in animal models of
17 cancer and objective clinical responses in clinical patient s with metastatic melanoma (132, 133). Results in human patients with glioma have been ge nerally disappointing, although responses to therapy have been observed in some cases (134-136). Disadvantages of adoptive immunotherapy includ e its temporary nature, as these cells or antibodies will eventually be eliminated, and re-administration of the therapy will be required. This therapy is very labor-intens ive due to the nature of these techniques, which require extensive in vitro or ex vivo manipulations, and complicates translation into the clinical setting. Active Specific Immunotherapy for Gliomas There has been extensive investigatio n of active specific immunotherapy for a number of different tumors. Results from precl inical trials have often been seen as very promising, and have generated considerable in terest in translating these therapies into humans. A review in 2003 cited 98 publis hed trials treating over 1000 human patients with dendritic cell-based vaccines (137) and a nother reference from the same year listed 216 ongoing clinical trials (138). Rosenberg et al (133) described an additional 440 individuals treated with a va riety of cancer vaccines (mainly for metastatic melanoma), and also pointed out the importance of appropr iate endpoint measures. In this groupâ€™s opinion, these large numbers of human patients were treated despite a lack of convincing evidence of efficacy in animal models w ith invasive, well-vascularized tumors. Reviewing their data and the results of numerous other va ccination trials, Rosenbergâ€™s group calculated a combined overall objective tumor response rate of 3.3% (133). The results in glioma models have been similarly promising, and some human trials have shown differences in certain outcome m easures with treatment, such as immune cell infiltration into tumors. However, objective tumor responses resul ting in regression or
18 increased survival are infrequent (139, 140) . Some reasons for this low efficacy may include low levels of circulating immune ce lls, low avidity for specific tumor antigens, inability of immune cells to infiltrate the tumor mass a nd to become appropriately activated at the site of interest and the activity of immunosuppressive and tolerance-inducing mechanisms within the tu mor microenvironment (133). In addition, even if a specific tumor antigen is successfully targeted, the genetic instability inherent in tumor cells may make it possible, or even likely, that a tumor clone with a mutated version of the antigen develops , and then expands unchecked. Thus, it appears that active specific immunotherapy, at leas t in the context of systemic vaccination paradigms, is unlikely to be an effective therapy when used alone for the majority of cancer patients. Active Nonspecific Immunotherapy for Gliomas Potential substances useful for nonsp ecific immunotherapy (NSI) of gliomas include cytokines and various products derived from different microbes. Cytokines may be intended to expand certain populations of immune cells (lymphocytes, NK cells, macrophages, microglia), upregulate certain immune molecules on the surface of specific cells (MHC molecules, costimulatory molecu les, adhesion molecules), or increase the production of additional immunomodulatory or in flammatory cytokines from one or more cells of interest. Microbes stimulate an immune response thr ough the binding of specific microbial products to the surface of certain cells of the immune system. These cells include dendritic cells, monocytes, tissue macropha ges and microglia. As discussed above, pattern recognition proteins on the cell su rface, such as TLRs are responsible for identifying these microbial products. Genera lly, the results of this interaction are stimulation of immune cells and promotion of a pro-inflammatory environment, with
19 upregulation of additional cel l surface receptors and the secretion of inflammatory cytokines, such as IL-1 and TNF. Many of these effects are mediated through NFB, which translocates into the nucleus to increa se expression of inflammatory gene products. The advantages of NSI over adoptive a nd active specific immunotherapy include ease of implementation, as no in vitro manipulation of cells is required, rapidity of translation to clinical patients and creation of a pro-inflammatory environment, which may be effective in creating a â€œdangerâ€ re sponse and breaking the peripheral tolerance associated with established neoplastic lesions. Historical use of nonspecific immunotherapy for cancer Nonspecific immunotherapy with microbial products has been used for the treatment of human cancer since the 1890s, wh en William Coley publis hed the results of his experience in treating patients with ex tracts of Erysipelas (141) and later, combinations of this organism and Bacillus prodigiosus ( Serratia marcescens ), to patients with a variety of cancers (142, 143). As a result, a commercially available bacterial preparation, known as â€œColeyâ€™s to xinsâ€ became available for this purpose, and was used for several decades, with some success. Later groups discovered that administration of a filtrate of the cultures of various bacteria could produce similar effects in experimental tumors in animals, specifically hemorrhagic necrosis of the tumor, and complete regression in some cases ( 144-146). Shear and colleagues isolated the active component of these culture filtrates, which turned out to be LPS (147). At the time, it was theorized that the action of LPS was upon tumor blood vessels, resulting in ischemia and subsequent necrosis of the tumor.
20 The primary structure of LPS was determ ined in 1966 and eventually lipid A was shown to be the active component (148, 149). LPS has been evaluated for its anti-tumor effects in a number of experimental tumor models (147, 150). The hemorrhagic necrosis of the tumor center thought to be related to effects on blood vesse ls (151), was later attributed to the production of TNFby activated macrophages after LPS stimulation (152, 153). Northâ€™s group demonstrated that the hemorrhagic necrosis of tumors after LPS treatment was separate from tumor regr ession, which required T cells to eliminate the surviving outer rim of tumo r tissue (150). Previous imm une sensitivity to the tumor and priming of macrophages was also require d in this model (154). Treatment with TNFwas not as effective as LPS in cau sing tumor regression, and demonstrated considerable toxicity at e ffective doses (152, 155). The use of LPS has also been evaluated in human patients with cancer with marginal success (156-158). While some treat ment responses were documented, toxicity was a limiting factor in these studies. LPS was usually administered intravenously, and one group found that intradermal administration reduced toxicity considerably (157). Another limiting factor of these trials was the enrollment of patients with advanced and usually metastatic cancers that had previously failed other treatments. Lipid A was first synthesized in 1985, and since then a number of different lipid A analogues have been developed (159). Many of these appear to retain the immunostimulatory effects of LPS while causi ng less toxic side effects (160, 161). These compounds have shown efficacy in a variety of experimental tumor models (160-163), and their effects have also been evaluated in human cancer patients (164).
21 A number of other compounds have been evaluated for nonspecific immunotherapy of brain tumors. Mice with chronic Toxoplasma gondii infection showed delayed growth of implanted brain tumors (165). Corynebacterim parvum delayed brain tumor growth when injected intracranially (166-168) and levamisole and Bacillus Calmette-Guerin (BCG) have both been evaluated in experime ntal glioma models as well as in human patients with malignant g liomas (169-171). Interferondid not show be neficial antitumor effects after intravenous or intratumoral administration in patients with malignant gliomas (172, 173). More recently, synthetic CpG oligonucleotides have been effective in the treatment of gliomas in a rat model when injected intratumorally (174) and are currently being evaluated in human patients with glioma (175). Microglial responses to nonspecific immunotherapy Microglia express a variet y of TLRs, including TLR2, TLR4 and TLR9, and as a result, are capable of responding to a variety of microbial products (176, 177). Both LPS and ZymA can lead to micr oglial activation, ch aracterized by increased phagocytic activity and the secretion of inflammatory cytokines (TNF, IL-1) and nitric oxide (178183). Cultured microglia treated with LPS and IFNexpress MHC II and costimulatory molecules and produce IL-12 (66, 67, 69, 72). In vitro studies have demonstrated the potenti al tumor cytotoxicity of microglia (72, 184, 185), which appears to be mediated in part by nitric oxide and/or TNF(184, 186). Microglia have the ability to pha gocytose apoptotic glioma cells (94, 187), although the result of this activity and the ability to produce effective antigen presentation remains uncertain. Microglia may act as effector cells in non-phagocytic tumor cell killing through the release of TNF(72, 188). Compounds that stimulate
22 microglial or macrophage recruitment and activation, phagocytosis or TNFrelease may be useful in the thera py of malignant gliomas. APC CD8 CD4 Th1 Th2MHC II MHC I Cell-Mediated Immunity Promotes CellMediated Immunity Promotes AntibodyMediated Immunity IL-12 IL-4 IFNIL-4 NKCD80/86 Treg Figure 1-1. Immune system overview. Antig en presenting cells (APC) control immunity through the presentation of antigen to lymphocytes and the secretion of cytokines. Antigen is either pr oduced endogenously within the cell and presented in the context of MHC I to CD8 positive lymphocytes (green circles) or exogenous antigen is cap tured , processed in lysosomes and presented in the context of MHC II to CD4 positive lymphocytes (red circles). Cross-presentation occurs wh en exogenous antigen is shuttled into the endogenous pathway and presented with MHC I (dotted line). For an effective response, MHC presentation requires the presence of costimulatory molecules (yellow cylinders). Under th e influence of IL-12, a Th1 response favoring cell-mediated immunity is produ ced. Secretion of IL-4 favors the production of a Th2 response, which promotes humoral immunity. The CD8 positive lymphocytes are the eff ectors of cell-mediated immunity. Other cells influencing the immune response include NK cells and Treg cells.
23 CHAPTER 2 IMMUNE SYSTEM REJECTION OF ESTABLISHED BRAIN TUMORS: A PROOF OF PRINCIPLE FOR IMMUNOTHERAPY Introduction The brain has traditionally been seen as an immunologically privileged organ, with attenuated immune responses and little capacity for the rejection of foreign tissue. This view has been fueled by various findings, in cluding a lack of lymphatic drainage from the central nervous system (CNS) and observations that allografts or xenografts survive for extended periods of time when engrafted in to the brain (45-47, 189). However, more recent studies have shown that antigens with in the brain do indeed reach the systemic circulation and peripheral lymph nodes (50, 190) and that activated T cells are capable of entering and routinely patrolli ng the CNS (53). In addition, microglia and perivascular macrophages within the CNS are capable of pr ocessing and presenting antigen to T cells and may possess direct tumoricidal activity th rough the phagocytosis of tumor cells and the secretion of cytokines such as TNF(61, 71, 75, 79, 191, 192). Rodent cell implantation paradigms are the most frequently used models to study brain tumor therapy. There is considerable variability in th e inherent immunogenicity of the rat glioma cell lines that are available, when implanted into their syngeneic hosts (113, 193). Indeed, it has become clear that one widely used tumor cell line (C6) does not have a clear syngeneic host (193, 194). The RG-2 cell line was crea ted through culture of a glioma that developed in the offspring of a Fisher rat after the administration of N-ethyl-N-nitrosurea during gest ation (193, 195). When inocul ated intracranially into
24 Fisher 344 (F344) rats, there is reliable grow th of a weakly immunogenic tumor, which is considered to more accurately model ma lignant human gliomas (113, 193). Although cells from these tumor lines are generally implanted into their syngeneic hosts, the implantation of 9L glioma cells into alloge neic rats has been advocated by number of authors as an acceptable model (196, 197) and implantation of C6 cells into Wistar or Sprague-Daley rats is still widely practiced. A number of studies have demonstrated rejection of allogeneic tumor cells when im planted into the brains of experimental animals (198-201), although it has been sugge sted that the mechanism of immune rejection may be different from what is seen elsewhere in the body because of a lack of lymphocytic infiltrate (198, 200). During previous studies in our laborator y utilizing RG-2 cell implantation into allogeneic Wistar rats, a proportion of rats did not have tumors on histological examination of the brain after sacrifice at predetermined time points. The purpose of this study was to examine the responses of syngeneic (F344) and allogeneic (Wistar) rats after intracranial implantation of RG-2 gliomas. Materials and Methods Animals Male, F344 and Wistar rats weighing 175-200 grams were obtained from Harlan (Indianapolis, IN). Animals were housed in a climate and humidity controlled environment and provided free access to a standard rodent diet (Harlan Teklad, Madison, WI) and water. All experimental animal pr ocedures were approved by the Institutional Animal Care and Use Committee at the University of Florida.
25 Cell Lines, Culture and Tumor Implantation The RG-2 cell line was originally obtained by our laboratory as a gift from Dr. Keith Black (University of California, Los Angeles). Cells were grown in Dulbeccoâ€™s Modified Eagle Medium (Gibco-Invitrogen, Ca rlsbad, CA) supplemented with 10% fetal bovine serum (Gibco-Invitrogen), penici llin (100 U/ml, Gibco-Invitrogen) and streptomycin (100 g/ml, Gibco-Invitrogen) and incubated with 8% CO2 at 37C in a humidified chamber. Immediately before im plantation, the cells were trypsinized with 0.25% trypsin (Mediatech, Herndon, VA), centr ifuged, counted with a hemacytometer, and resuspended in sterile phosphate buffered saline (PBS). A total of 8 F344 and 16 Wistar rats were used for these experiments. For tumor implantation, rats were anesthetized with isoflurane (Phoenix Pharmaceutical, St. Joseph, MO) in oxygen, placed in a stereotactic head frame and a burr hole was made in the skull 3 mm to the right of bregma. Tumor cells (10,000 cells in 10 l) were injected with a Hamilton syringe (Hamilton, Reno, NV) 5 mm ve ntral to the dura over a period of 2 minutes, after which the syringe was slowly withdrawn. The burr hole was filled with bone wax (Surgical Specialtie s Corporation, Reading, PA) a nd the skin incision closed with autoclips (Becton Dickinson and Co mpany, Sparks, MD). Buprenorphine (Buprenex, 0.03 mg/kg subcutaneously, Reck itt Benckiser Pharmaceuticals, Richmond, VA) was administered for post-operative analgesia. Animal Observation and Clinical Examination Animals were observed daily for changes in mentation, behavior and mobility. Animals were periodically removed from their cages to assess conscious proprioception, evaluated by turning the pa ws of the pelvic and thoracic limbs over onto the dorsal surface and observing for replacement to a normal position (202). Animals that displayed
26 significant lethargy, obtundation or reluctance to move, which, ba sed on previous studies, typically preceded death by approximately 1 day, were euthanized with pentobarbital (Beuthanasia-D, 200-400 mg/kg intraperit oneally, Schering-Plough Animal Health, Union, NJ). Magnetic Resonance Imaging Magnetic resonance imaging (MRI) was performed at the McKnight Brain Instituteâ€™s Advanced Magnetic Resonance Imaging and Spectroscopy (AMRIS) Facility at the University of Florida. Rats were an esthetized with isoflu rane in oxygen and an intravenous catheter was placed in a tail vein. Brain imaging was performed with a 4.7 T Oxford Magnet (Oxford Instruments, Carter et, NJ) using a Bruker Avance console and ParaVision software (Bruker Biospin MRI, In c., Billerica, MA). A custom-built 3.5 cm inside-diameter quadrature birdcage volume coil was used to acquire the images. Following pilot images, dorsal T2-weighted images were acquired using a multi-slice multi-echo (MSME) rare sequence with the following parameters: Field of view (FOV) 5.0 x 3.0 cm, matrix of 256 x 128, TR = 2800 ms ec, TE = 26 msec, 8 averages, slice thickness 0.85 mm. Transverse T2-weighted MSME images were then obtained (FOV 3.2 x 3.2 cm, matrix 256 x 128, TR = 2500 ms ec, TE = 19 msec, 8 averages, slice thickness 1.0 mm) followed by transverse T1-weighted MSME images (FOV 3.2 x 3.2 cm, matrix 128 x 128, TR= 614.3 msec, TE = 10 .8 msec, 4 averages, slice thickness 1.0 mm). After the administration of an intr avenous contrast agent (gadodiamide, 172 mg/kg, Amersham Health, Arlington Heights, IL) an additional set of transverse T1-weighted MSME images were obtained.
27 Histology and Immunohistochemistry Rats were euthanized if they displa yed obtundation or significant lethargy as described above. In addition, ten animals were sacrificed at specific time points after tumor inoculation (4, 7, 11, 13, 15 days) for histopathology and imm unohistochemistry. Rats were euthanized with intraperitoneal pentobarbital (200-400 mg/kg) and perfused transcardially with paraformaldehyde (4% in PBS; Ted Pella Inc., Redding, CA). Brains were removed and fixed in 4% paraformalde hyde for 1-2 hours, then washed in PBS. Tissue samples were then placed in 30% su crose for cryopreservation, frozen, cut in serial sections (14 m) on a cryostat (Micro m International GmbH, Walldorf, Germany) and mounted on slides. Routine staining was performed with cr esyl violet. For immunohistochemical evaluation, primary antibodies for microg lia/CD11b (OX-42, 1:100, Se rotec, Raleigh, NC), T cells/CD6 (OX-52, 1:100, Serotec), CD4 (W3/25, 1:100, Sero tec), CD8 (OX-8, 1:100), MHC I (OX-18, 1:100, Serotec) a nd MHC II (OX-6, 1:100, Serotec) were utilized. Prior to primary antibody applic ation, sections were quenched in 1% hydrogen peroxide in methanol for 30 minutes, permeabilized with PBS + 0.5% Triton-X for 15 minutes and blocked with 10% normal goat serum (NGS, Gibco-Invitrogen) in PBS + 0.3% Triton-X for 30-60 minutes. Primary antibodies were diluted in PBS + 0.3% Triton-X + 3% NGS and applied for 60 minutes at room temperatur e in a humidified chamber. The secondary antibody was bio tinylated goat anti-mouse (1:500, Vector Laboratories, Burlingame, CA) diluted in PBS + 0.3% Triton-X + 3% NGS and applied for 30 minutes at room temperature. Avidin -horseradish peroxida se (AV-HRP) complex (1:500, Vector Laboratories) was diluted in PBS + 0.3% Triton-X and applied for 30 minutes at room temperature. Visualiza tion was achieved with 3, 3â€™-diaminobenzidine
28 (DAB, Sigma, St. Louis, MO ) activated with hydrogen pe roxide. Sections were counterstained with cresyl violet. The secti ons were washed three times in PBS after the primary and secondary antibodies, AV-HRP, and DAB were applied. Sections were examined with brightfi eld microscopy (Carl Zeiss, Thornwood, NY) and photographed with a digital camera and SPOT software (Diagnostic Instruments, Sterling Heights, MI). The average count s of immunopositive cells within 10 randomly selected oil (100X magnification) fields bot h within the tumor mass (intratumoral) or surrounding the tumor mass (peritumoral) were used to compare OX-52, OX-8 and W3/25 positive cells between rat strains. Statistical Analysis A two-tailed Mann Whitney U test was us ed to compare cell counts between rat strains and a P value of less than 0.05 was cons idered significant. Statistical software (Instat 3.0 and Prism 4.0, Graphpad Software Inc., San Diego, CA) was used to compare groups and graph data. Results Observation and Clinical Examination F344 rats began to show obvious signs of neurologic dysfunction after 15 to 16 days, while Wistar rats typical ly had signs begin 22 to 25 days after tumor implantation. Neurologic signs consisted of a te ndency to turn or walk in wide circles towards the right side, obtundation or lethargy. Rats displayi ng these signs usually had obvious conscious proprioceptive deficits in the limbs of the left side (i.e., failure to replace the paw to a normal position). Some Wistar rats failed to develop any detectable neurologic signs, while others displayed obvious circling beha viors and conscious proprioceptive deficits, which then resolved as their tumors regres sed (see MRI results belo w). F344 rats tended
29 to show altered mentation more often than circling behaviors, which invariably progressed to a stuporous state characterized by decreased response to stimuli, reluctance or inability to move and failure to groom. All F344 rats not sacrific ed at an earlier time point reached this state, whereas all Wistar rats failed to de velop these terminal signs, and had a full recovery to ba seline neurologic function. Magnetic Resonance Imaging There was some variation in tumor grow th, although mass lesions were usually detectable by MRI at 10 days post-implantation. The tumors were generally isointense to slightly hypointense on T1-w eighted images, had a mixed intensity on T2-weighted images and displayed consistent contrast enhancement. Some tumors showed homogenous enhancement, whereas others had regions that failed to enhance, suggestive of necrosis. A heterogeneous enhancement patte rn was seen more often in F344 rats. In larger tumors, there was often considerable mass effect, leading to displacement of the normal parenchyma and ventricular system, as well as cerebral edema (Figures 2-1 and 2-2). Some tumors, particularly in the Wistar strain, were surrounded by a hypointense non-enhancing ring on both T1and T2-weight ed views. Although some Wistar rats developed tumors that grew to considerable sizes, these eventually regressed in all cases, becoming undetectable on MR imag ing (Figure 2-2). In contra st, all F344 rats followed with MRI developed extremel y large tumors with secondary edema and mass effect, which correlated with obtundation, n ecessitating euthanasia (Figure 2-1). Histology and Immunohistochemistry Tumor cells were visible along the needle tract at 4 days , and had formed a readily identifiable mass of cells by 7 days in both ra t strains. The tumor masses were composed of a heterogeneous collection of cells, and subs tantial areas of necrosis were evident in
30 both strains by 11 days (Figure 2-3C). Neoplas tic cells were fusiform or spindle-shaped, and formed large sheets and whorl-like form ations. The remaining cells were primarily inflammatory in nature, consisting of microglia, macrophages, lymphocytes and occasionally neutrophils. Neutrophils were primarily seen in F344 rats in areas with obvious tumor necrosis. Fingers of tumor tissu e were evident at the margins of the tumor mass in both strains, invading adjacent pare nchyma and infiltrating into perivascular spaces (Figure 2-3A). Pyknotic nuclei sugges tive of apoptotic cells were noted in a number of tumors (Figure 2-3D). Substantial numbers of activated microglia were noted surrounding the tumors by 7 days, and by 11 days an intense reaction with activated microglia surrounding the margins of the tumor and invading into the tumor mass itself was noted. Large, rounded, OX-42 positive cells (macrophages) were num erous within the tumor masses. Both microglia and macrophages e xpressed MHC II and there wa s no obvious difference in expression between the two strains (Figure 24G, H). In contrast, there were greater numbers of microglia surrounding the tumors in Wistar rats than in F344 rats (Figure 2-4A, B). These microglia were noted to surr ound tumor cells in the perivascular spaces, and persisted in increased numbers in some Wi star rats where the tumor appeared to have completely regressed. While OX-42 positive macrophages were distributed uniformly throughout the tumor in Wistar rats, their di stribution was patchy in the F344 rats by 13 days, with substantial portions of the tumo r mass devoid of these immune cells. The majority of cells within the tumors in both strains stained positive for MHC I, including the neoplastic RG-2 cells and invading m acrophages. In Wistar rats, peritumoral microglia were noted to stain strongly for MHC I, in contrast to the F344 rats which
31 lacked such conspicuous staining (Figure 2-4C-F). On close examination of the immediate peritumoral area in the F344 rats, very faint immunoreactiv ity was noted that suggested microglial staining. There were marked differences in T cell in filtration into the tumors between the two strains. OX-52 positive cells were present in small numbers within the tumors of F344 rats, while there were significantly greater num bers of cells in the Wistar rat tumors, noted from 7 days post-implantation onwar d (Figures 2-5A, B, and 2-6A, B). Significantly more OX-52 positive cells were al so noted within the neuropil surrounding the tumors in Wistar rats, and were often seen in clusters surrounding blood vessels. There were significantly greater numbers of CD8 positive cells both intratumorally and peritumorally in Wistar rats at most of the time points evaluated (Figures 2-5C, D and 2-6C, D). The exception to this occurred 4 days post-implantati on, where significantly greater numbers of CD8 positive cells were noted in Fisher 344 rats, all of which displayed a morphology characteristic of macr ophages (see below). Large numbers of CD8 positive cells with typical lymphocyte morphology were first observed 7 days after implantation in Wistar rats and were often clus tered perivascularly in peritumoral areas. In F344 rats, CD8 positive lymphocytes were only occasionally found within the tumor mass and were almost nonexistent in the surr ounding neuropil. Great er numbers of CD4 positive cells were also seen in Wistar rats compared to F344 rats, although this was less striking than the CD8 immunosta ining (Figures 2-5E, F and 2-6E, F) . Peritumoral CD4 positive cells could not be reliab ly counted in Wistar rats from 11 days onward due to the intense staining of microglia with this antibody (see below).
32 Interestingly, cells with typical microg lial morphology were noted to express both CD8 and CD4. CD8 positive microglia were noted as early as 4 days after RG-2 implantation, and were present in approximately equal numbers in both rat strains. These cells generally formed a rim several cells thick around the tumor. Robust CD4 staining of peritumoral microglia was seen in Wistar rats, first noted at 11 days post-implantation. Large, round to oval cells within the tumor likely to be macrophages also stained positive for CD4. In contrast, CD4 immunoreactivity was only detectable at very low levels in microglia surrounding the tumor in F344 rats pr ior to 15 days. At the 15-day time point, microglial CD4 expression was more obvious, bu t did not reach the levels noted in the Wistar rats. Intratumoral CD4 positive macrophages were present within the tumor in numbers similar to those seen in Wistar rats. Long-Term Survival Studies Some Wistar rats with substantial tumors that subsequently regressed (based on MRI) were followed for variable periods of time. These rats invariably survived without subsequent neurologic impairment or obvi ous tumor recurrence for up to 20 months. Post-mortem histologic examination in thes e animals often showed groups of what appeared to be residual tumo r cells surrounded by reactive mi croglia. Interestingly, these tumor cells were usually adjacent to the late ral ventricle on the side of the injection, particularly along the lateral wa ll and at the dorsolat eral extent of the ventricle. The striatum in the region of the injection site showed large numbers of reactive microglia and residual lymphocytes. Discussion This report describes the profound differen ce in host response when the RG-2 cell line is implanted into a syngeneic versus an allogeneic rat stra in. Rejection of the tumors
33 in Wistar rats was characterized by significantly greater numbers of CD8 positive T cells and increases in MHC I and CD4 positive microglia compared to F344 rats. Our results are consistent with other reports in which intracranial tumors were rejected after implantation into allogeneic animals (198200, 203). Although previous reports have described long-term survival of allogene ic or xenogeneic tumors or tissue after intracranial implantation and fo rwarded the view of the br ain as an immunologically privileged site (45-47, 189), it is clear that this immune privilege is far from complete (49). The RG-2 cell line displays very aggres sive growth and invasi on when transplanted into either rat strain. However, as our study illustrates, even though these tumors may reach a substantial size, leadi ng to significant neurologic deficits, they invariably regress in an allogeneic host, at least in the outbred Wistar strain. The rejection of the tumors in Wistar rats appeared to be mediated in large part by CD8 positive T cells. This is in contrast to a study of 9L tumors by Albright et al. (198), which showed allogeneic tumor rejection wi thout apparent lymphocytic infiltration. However, this study did not evaluate ly mphocytes immunohistochemically, which may have underestimated cell numbers. Other stud ies have shown modest T cell infiltration in allogeneic hosts with (203) or without (196) tumor regressio n. Tzeng et al. (113) showed a correlation between survival and CD8 pos itive lymphocyte numbers in two syngeneic tumor models (F98 and RG-2), and attempts ha ve also been made to correlate survival with lymphocytic infiltration of glioblastoma s in human patients, with varying results (110-112, 114). The role of T cells in tumo r rejection is highli ghted by the observation that allogeneic athymic nude rats (rnu/rnu) lacking T cells have rapid tumor growth
34 leading to death 17 to 20 days after implanta tion of identical numbers of RG-2 cells into the same location (data not shown). Nave T cells do not efficiently enter nonlymphatic tissue (7) and appear to be unable to cross the blood-brain barrier unde r most physiologic conditions (53, 61, 191). However, CNS antigen elicits an immune re sponse within peripheral lymphatic organs, including the spleen and deep cervical lymph nodes (52, 190). Tumor antigens may encounter antigen presenting cells in these pe ripheral organs after draining through nasal lymphatics by way of cerebral perivascular spaces (190, 204). Alternatively, it is possible that cells capture antigen within the CNS and subsequently migrate by similar pathways to these periph eral locations. Cellular candidates for antigen capture in clude microglia, perivascular macrophages and dendritic cells, which have the ability to phagocytose foreign material, organisms or cells and present antigen to T cells in the context of MHC II (61, 71, 75, 79-81, 191, 192). Substantial numbers of activated micr oglia were present surrounding and invading the tumors in both F344 and Wistar rats, a lthough the response was more pronounced in the latter strain. A similar re sponse has been noted in human patients with gliomas (192, 205). While both strains had moderate numbers of MHC II positive microglia, robust staining for MHC I on microglia was noted only in Wistar rats. Upregulation of MHC II has been reported with a number of diseases or insults affecting th e brain and appears to be somewhat nonspecific in nature (75, 192, 206, 207). It is possible that MHC I upregulation occurs as a result of secretion of interferonby a large number of invading T lymphocytes. Alternatively, it is possibl e that the upregulation of MHC I was an integral part of the immune response agains t the tumor. The presence of CD8 positive
35 lymphocytes within the brain parenchyma indi cates activation of th ese cells peripherally by antigen presenting cells in the context of MHC I. The shuttling of sampled exogenous antigen into the MHC I pathway, known as cross-presentation (7, 8), has been documented to occur within dendritic cells (9-11) and macrophages (14, 15) in immune responses to a variety of an tigenic stimuli, including tumo r antigens (8, 16-19, 208, 209). Local restimulation of invading CD8 positive T cells by MHC I positive microglia might be important in an effective immune re sponse and subsequent tumor regression. The expression of CD4 and CD8 on microglia has been reported previously (92, 210-213). As noted in our study, MHC II and CD4 expression on microglia have a similar distribution, and it has been suggest ed that CD4 may be involved in signaling between microglia (212). Similar to MHC I, significant microglial CD4 expression was only noted in Wistar rats. Whether this is a non-specific side effect of the presence of other cells or an integral part of a su ccessful immune res ponse is unclear. Similarly, the role of microglial CD8 is speculative at this time. CD8 positive macrophages and microglia have been documente d in rats with various CNS lesions, and appear to be associated with areas of necrosis or neuronal cell loss (210, 212-214). Necrotic areas within tumors from both stra ins of rat may have led to local microglial CD8 expression or the recruitment of CD8 positive macrophages. In macrophages from other organs, the ligand binding domain of CD8 appears to be distinct from that on T cells, and this molecule likely interacts with ligan ds other than MHC I (215), resulting in increased production of nitric oxide, TNFand IL-1 , which mediate a number of both proinflammatory and regenerativ e effects (212, 216). Direct microglial killing has been documented in vitro for a number of tumor cell lines (72, 184, 185), including RG-2 cells
36 in our laboratory, and TNFand nitric oxide appear to play at least a partial role in this tumoricidal activity (186). It is possible that microglia played a significant role in the direct elimination of neoplastic cells in the Wistar rats in our study through such mechanisms. Cooperative activity between microglia and T lymphocytes in terms of both sensitization of effector cells and elim ination of damaged tumor cells is likely essential to elicit a su ccessful anti-tumor response (94, 217). The rejection phenomenon reported here has se veral important implications. First, it highlights the importance of interim asse ssments of tumor growth, facilitated by advanced diagnostic imaging techniques such as MRI. Point-in-time assessments of tumor growth that are necessar ily coupled to sacrifice of the animal may not provide a complete picture of the natural history of the tumor in many cases. The increasing availability of high-field rese arch MRI facilities should impr ove this situation, and tumor development should be monitored longitudina lly whenever possible. Secondly, despite reports to the contrary (196, 197), the resu lts of studies employing allogeneic tumor models should be interpreted with caution, especially when using tumor regression or survival as endpoints (193). Studies ev aluating immunotherapy should be avoided in allogeneic models (194). Finally, our study shows that ev en in tumors of substantia l size, the immune system can effectively eradicate these masses, lead ing to resolution of clinical signs and long-term survival. This may be seen as a proof of principle fo r immunotherapy, if the immune system can be adequately sensitized to the tumor. The differences in the cellular infiltrate and surface expression of immunologic molecules between the two rat strains in our study may provide important insights into the mechanis ms of tumor rejection and
37 potential targets for future immunothera peutic studies. Our study suggests that components of both the innate (microglia and macrophages) a nd adaptive (T cells) immune systems are required for the elim ination of intracranial tumors. Figure 2-1. An RG-2 glioma in a F344 rat (MR images). A) T2weighted images 10 days after implantation. B) Post-con trast T1-weighted images 10 days after implantation. C) T2weighted images 17 days after implantation. D) Post-contrast T1-weighted images 17 da ys after implantation. Note the profound mass effect with displacement of the ventricular system, cerebral edema and heterogeneous contrast enhancement.
38 Figure 2-2. An RG-2 glioma in a Wistar rat (MR images). A,C,E, and G show T2weighted images and B,D,F, and H show post-contrast T1-weighted images. Images shown are taken at 11 days (A, B), 18 days (C, D), 25 days (E, F) and 32 days (G, H) after implantation. Note the development of substantial mass effect with displacement of the ve ntricular system, cerebral edema and heterogeneous contrast enhancement fo llowed by subsequent regression of the tumor.
39 Figure 2-3. Cresyl viol et staining of RG-2 gliomas. A) Day 7 F344 rat showing invasive growth of tumor (Bar = 200 m) and perivascular invasion along VirchowRobin space (inset, bar = 25 m). B) Day 13 F344 rat showing heterogeneous population of cells within tumor mass (Bar = 50 m). C) Day 11 Wistar rat showing extensive intratumoral necrosis (asterisks, bar = 200 m). D) Day 13 Wistar rat showing pyknotic and fragme nted nuclei within tumor (arrows) suggestive of apoptotic cell death (Bar = 12.5 m).
Figure 2-4. Immunohistochemi cal staining of RG-2 gliomas for CD11b, MHC I and MHC II in F344 and Wistar rats. CD11b (A, B), MHC I (C-F) and MHC II (G, H) immunostaining of RG-2 glioma s using horseradish peroxidase method counterstained with cresyl violet. The dotted line indicates the tumor margin in sections A-D. A) Day 11 F344 ra t showing patchy macrophage staining within tumor and modest peritumo ral microglial reaction. (Bar = 100 m). Arrows indicate perivascular tumor satellites, which is demonstrated at higher power in the inset (Bar = 25 m). B) Day 11 Wistar rat showing more robust peritumoral microglial reaction (Bar = 100 m), particularly surrounding peritumoral blood vessels (Arrows and inset, bar = 25 m). C) Day 11 F344 rat showing MHC I immunoreactivity w ithin tumor but sparse staining surrounding tumor (Bar = 100 m). D) Day 11 Wistar rat showing MHC I immunoreactivity within tumor and a more robust peritumoral immunoreactivity, due to microglial staining (Bar = 100 m). E) Higher power of peritumoral area of rat in C, showing lack of staining (Bar = 25 m). F) Higher power of peritumoral area of rat in D, showing robust MHC I staining of microglia (Bar = 25 m).G) Day 13 F344 rat showing peritumoral MHC II positive microglia (Bar = 25 m). H) Day 13 Wistar rat showing peritumoral MHC II positive microglia (Bar = 25 m).
42 Figure 2-5. Immunohistochemical staining of RG-2 gliomas for CD6, CD8, and CD4 in F344 and Wistar rats. CD6 (A, B), CD8 (C, D) and CD4 (E, H) immunostaining of RG-2 gliomas us ing horseradish peroxidase method counterstained with cresyl violet. The dotted line indicates the tumor margin in sections A-F. A) Day 11 F344 rat showing sparse T cell infiltration of tumor (arrows) and almost nonexisten t peritumoral reaction (Bar = 100 m). B) Day 11 Wistar rat showing much mo re robust T cell infiltration of tumor and peritumoral area (Arrows, bar = 100 m). C) Day 11 F344 rat showing very little CD8 immunoreactivity (Bar = 100 m). D) Day 11 Wistar rat with marked intratumoral and peritumoral CD8 positive T cell infiltrate (Bar = 100 m). E) Day 11 F344 rat showing spar se CD4 immunoreactivity (Bar = 100 m). F) Day 11 Wistar rat with robust intratumoral and peritumoral CD4 positive T cell infiltrate. Note increa sed peritumoral immunoreactivity not associated with T cells, due to CD4 positive microglia (Bar = 100 m).
Figure 2-6. Quantification of intratumoral and peritumoral inflammatory cells in RG-2 gliomas. Graphs for CD6 (A, B), CD8 (C, D) and CD4 (E, F) positive cells. For all figures, Wistar rats are repr esented by white bars and F344 rats by black bars. There were si gnificantly greater numbers of T lymphocytes in Wistar rats versus F344 rats at most time points both intratumorally (A, *p = 0.003, **p < 0.0001, not enough tumor mass in Wist ar rats for comparison) and peritumorally (B, *p < 0.0001). Simila rly, there were significantly greater numbers of CD8 positive cells in Wistar ra ts versus Fisher rats, except at Day 4, where this trend was revers ed (C, intratumoral, *p < 0.0001, cell count based on 1 oil field due to small amount of identifiable tumor; D, peritumoral, *p = 0.015, **p < 0.0001). CD4 immunoreactivity was less consistent, but generally showed significantly greater ce ll counts in Wistar rats in both intratumoral (E, *p = 0.0002, **p = 0.0019, ***p = 0.03 [limited number of fields]) and peritumoral (F, *p < 0.0001, unable to count due to intense microglial reaction) locations.
45 CHAPTER 3 NONSPECIFIC IMMUNOTHERAPY IN SUBCUTANEOUS RG-2 GLIOMAS Introduction In both experimental glioma models and naturally occurring gliomas in humans, there is extensive invasion of microglia, macrophages and T lymphocytes, but these immune cells are unable to halt the gr owth of the tumor (86, 92, 93, 95, 192). The reasons for the ineffectiveness of the immune cells is likely multifactorial, and may include a lack of expression of distinct tu mor antigens, altered major histocompatibility complex molecules, the secretion of imm unosuppressive cytokines (e.g., transforming growth factor [TGF], interleukin [IL]-10, prostaglandi n E2), the expression of death ligands on the surface of tumor cells, and th e immune suppressing action of CD4+CD25+ regulatory T cells (Treg) (116-119, 125, 126, 192). Methods of overcoming this immune suppression include interfering with the sp ecific mechanisms described above or encouraging the immune cells to overcome the suppressive effects of the tumor by immunostimulation. When activated, both microglia and macrophages have the capacity for phagocytosis and the ability to secrete a variety of cytokines, including TNFand IL-1 (64, 65). A number of substances can l ead to activation of these cells, including lipopolysaccharide (LPS), zymosan A (Zym A) and granulocyte-macrophage colony stimulating factor (GM-CSF) (218-220). When activated, microglia and macrophages may display tumoricidal activity, and may serv e an integral role in the connection of innate and adaptive immunity (184, 221). Th e objective of these experiments was to
46 evaluate the anti-tumor prope rties of LPS, ZymA, GM-CSF and combinations of LPS and IFNin a syngeneic rat glioma mo del in a subcutaneous location. Lipopolysaccharide LPS is derived from the outer cell membrane of gram-negative bacteria, and is composed of an oligosaccharide chain with an attached glycolipid moiety. The biologically active component of this complex is a molecule known as lipid A (148). LPS binds to one of a group of pattern recep tors recognizing microbes, known as toll-like receptors (TLR), specifically TLR4 (218). TLR4 is present on cells of the innate immune system, including macrophages, dendritic cells and microglia, and lig and binding leads to several effects in these cells , including upregulation of pr oinflammatory cytokines such as TNF, IL-1, IL-6, and IL-12, production of induc ible nitric oxide synthase (iNOS), and expression of costimulatory molecules su ch as CD80 and CD86 (218). These effects are mediated through several different signal transduction pathways within the immune cells. Zymosan A ZymA is derived from the cell wall of the yeast Saccharomyces cerevisiae and contains a number of components, including 1 3-glucans, which are polymers of D-glucose and are believed to represent th e active component of this compound (220). ZymA also binds to TLRs on the surface of innate immune cells, specifically TLR2 and TLR6, and leads to the upregulation of inflammatory gene products (38, 218). Granulocyte-Macrophage Colony Stimulating Factor GM-CSF is a cytokine produced by T cells , NK cells and macrophages and binds to its receptor on macrophages, dendr itic cells and microglia (222). It has a variety of effects on these cells, including promoting phagocytosis, proliferation and antigen
47 presentation (222, 223). GM-CSF can dramati cally increase proliferation in microglial cells, and lead to upregulation of a number of markers characteris tic of DC, including CD11c, DEC-205 and the costimulatory molecule CD80 (56). It was the most effective substance at stimulating anti-tumor immunity in a screening study of a large number of cytokines, costimulatory receptors and adhe sion molecules (224), and has been evaluated in a number of pre-clinical and clinical tr ials for activity against a variety of cancers (225-228). Interferon-gamma Interferonis a cytokine secreted primarily by activated T cells and NK cells. It has many complex functions which are mediat ed through its specific receptor and the JAK-STAT signal transduction pathway, and generally promotes the Th1 arm of the adaptive immune response (229). Specific func tions induced in macrophages or dendritic cells include the promotion of phagocytic activity, upregulation of MHC molecules and costimulatory molecules involved in antigen presentation, and increa sed secretion of a variety of chemokines and cytokines, notably IL-12 (229, 230). Interferonhas also been reported to have anti-proliferative effects and may upregul ate Fas expression on neoplastic cells (231, 232). For a number of years, IFNhas been used in combination with LPS to activate macrophages for tumor cell killing, leading to upregulation of TNF, MHC II molecules, nitric oxide activity and phagocyt osis (233). In fact, IFNhas been considered the priming agent for macrophages, and is re quired for macrophage activation by LPS in vitro (234). Synergistic effects seem to occu r at several locations within and between cells, including at the level of surface recep tors, by augmentation of signal transduction pathways and transporter associated with an tigen presentation prot ein 1 (TAP-1) kinetics
48 and through modulation of autocrine and paracrin e loops (235). Held et al. (235) showed that IFNenhanced macrophage responses to LPS by augmentation of the NFB signal transduction cascade and through an autocrine loop involving TNFand IL-1. Powell et al. (235) determined that IFNcan synergize with LPS as well as other TLR ligands such as ZymA and CpG oligonucleotides in macrophage activation. The mechanism was NFB dependent, and occurred at the transcriptional level, at least for the production of monokine induced by IFN(MIG), a chemokine induci ng T cell migration. Similar results have been seen with microglia, and combinations of LPS and IFNhave been shown to work synergistically to pr omote tumor cell killing by microglia in vitro (186). Routes of Administration: Intrat umoral versus Systemic Treatment Attempts at nonspecific immunotherapy in human patients and experimental models have been made for over 100 year s, with some limited success (141-143, 165, 174, 236). The vast majority of studies ha ve utilized systemic administration of immunotherapeutic compounds, typically in travenously or intraperitoneally. Administration of the immunostimulatory co mpounds directly into the tumor mass has the potential to diminish systemic side effects by reducing the absorption of these compounds into the circulation. In addition, as the treatment is delivered directly to the desired site of action, the ther apeutic effect is maximized, and a lower total dose may be used for therapy. Intratumoral therapy fo r brain tumors has been described using chemotherapeutic compounds (237-241), radia tion treatment with radiation implants (242) or radiolabeled an tibodies (241, 243), immunotoxi ns (244), adoptive cellular therapies (245-248), and nonspecifi c immunotherapy (163, 249).
49 Materials and Methods Animals Male and female Fisher 344 (F344) rats and male nude (rnu/rnu) rats weighing 175200 g were obtained from Harlan (Indianapolis , IN). Animals were housed in a climate and humidity controlled specific pathogen fr ee environment and provided free access to a standard rodent diet (Harlan Teklad, Madison, WI) and water. All experimental animal procedures were approved by the Institutional Animal Care and Use Committee at the University of Florida. Cell Lines, Culture and Tumor Implantation The RG-2 rat glioma cell line was a gift from Dr. Keith Black (University of California, Los Angeles). Cells were gr own in Dulbeccoâ€™s Modified Eagle Medium (Gibco-Invitrogen, Carlsbad, CA) supplemente d with 10% fetal bovine serum (GibcoInvitrogen), penicillin (100 U/ml, Gibc o-Invitrogen) and streptomycin (100 g/ml, Gibco-Invitrogen) and incubated with 5% CO2 at 37C in a humidified chamber. Immediately before implantation, the cells were trypsinized with 0.25% trypsin (Mediatech, Herndon, VA), centrifuged, washe d, counted with a hemacytometer, and resuspended in sterile phosphate buffered NaCl solution (PBS). For tumor implantation, rats were an esthetized with isoflurane (Phoenix Pharmaceutical, St. Joseph, MO) in oxygen. The hair was clipped between the scapulae and approximately 500,000 tumor cells in a volume of 30 to 130 l were injected subcutaneously with a tuberculin syringe and 25 gauge needle. Animal Observation and Tumor Measurement Animals were observed daily for evaluati on of tumor growth and appearance and changes in behavior or mobility. To ev aluate tumor size, animals were briefly
50 anesthetized with isoflura ne in oxygen and measurements taken in two orthogonal directions with digital ca lipers (Mitutoyo America, Auro ra, IL). Tumor volume was calculated by the following formula: Volume = 0.52 x ab2, where a is the larger measurement and b the smaller. In some cases, multiple, distinct masses were present, in which case each was measured as described above and the volumes added together to reach a total tumor volume. If tumors exceeded 5000 mm3, animals were euthanatized. Tumor Treatments Animals were anesthetized with isoflura ne in oxygen for treatments, which were performed in conjunction with tumor measurements. Tumors were treated when they reached a volume of 200 mm3. Lipopolysaccharide (strain O5:B5, Sigma-Aldrich, St. Louis, MO), ZymA (Sigma-Aldrich), or GM-CSF (R&D Systems, Minneapolis, MN) were administered at varying dose schedules (see below). Treatments were reconstituted in NaCl solution (0.9%), standa rdized to a volume of 40 to 50 l and administered with a tuberculin syringe and 27 gauge needle directly into the tumor (intratumorally [IT]). Histology and Immunohistochemistry Rats were euthanized at end points descri bed above or in some cases, at specific time points after tumor inoculation for histopathology. Sodium pentobarbital (Beuthanasia-D, 200-400 mg/kg intraperit oneally, Schering-Plough Animal Health, Union, NJ) was administered and the anim als were perfused with intracardiac paraformaldehyde (4% in PBS). The tu mors were removed and fixed in 4% paraformaldehyde for 1 to 2 hours, then washed in PBS. Tissue samples were then placed in 30% sucrose for cryopreservation, froz en, cut in serial sections (10 to 20 m) with a cryostat and mounted on slides.
51 In rats euthanatized because of tumor volume in excess of 5000 mm3, conventional staining with cresyl violet or hematoxylin as well as immunohistochemistry for the antibodies listed above was performed. In addition, a number of rats (n = 6) were sacrificed at specific time poi nts after immunostimulatory trea tment (7 or 14 days) to try and define differences in cellular infiltra tion and cell surface mark ers between treatment groups. Routine staining was performed with cres yl violet (Sigma-Aldrich) or Mayerâ€™s hematoxylin (Vector Laboratories, Bur lingame, CA). For immunohistochemical evaluation, primary antibodies for m acrophages (OX-42, 1:500 and ED2, 1:100), dendritic cells/integrin E2 (OX-62, 1:100), NK cells/CD161 (3.2.3, 1:100, gift from W. Chambers, University of Pittsburgh), CD 6/T cells (OX-52, 1:100), CD4 (W3/25, 1:100), CD8 (OX-8, 1:100), MHC I (OX-18, 1: 100), MHC II (OX-6, 1:100), CD25 (OX-39, 1:100), CD80 (3H5, 1:100, BD Biosciences, San Jose, CA) and CD86 (24F, 1:100, BD Biosciences) were utilized. All antibodies were obtained from Serotec (Raleigh, NC) unless otherwise identified. Prior to primary antibody application, endogenous peroxidase was quenched with 1% hydrogen peroxide in methanol for 30 minutes, sections were permeabilized with NaCl so lution (PBS) + 0.5% Triton-X for 15 minutes and blocked with 10% normal goat serum (N GS, Gibco-Invitrogen) in NaCl solution (PBS) + 0.3% Triton-X for 30 to 60 minutes. Primary antibodies were diluted in NaCl solution (PBS) + 0.3% Triton-X + 3% NGS and applied for 60 minutes at room temperature or overnight at 4 C in a humidified chamber. Biotinylated secondary antibody used was goat anti-mouse (1:500, Vector Laboratories) dilute d in NaCl solution (PBS) + 0.3% Triton-X + 3% NGS and applied for 30 minutes at room temperature.
52 Avidin-horseradish peroxidase (AV-HRP) complex (1:500, Vector Laboratories) was diluted in NaCl solution (PBS) + 0.3% Trit on-X and applied for 30 minutes at room temperature. Visualization was achie ved with 3, 3â€™-diaminobenzidine (DAB, Sigma-Aldrich) activated with hydrogen per oxide. Some sections were counterstained with cresyl violet or hematoxylin. The secti ons were washed three times in PBS after the primary and secondary antibodies, AV-HRP a nd DAB were applied. Sections were examined with brightfield microscopy (Carl Zeiss, Thornwood, NY) and photographed with a digital camera and SPOT software (Diagn ostic Instruments, Sterling Heights, MI). Statistical Analysis Survival curves were constructed usi ng the Kaplan-Maier product limit method with a statistical software program (Prism 4.0, Graphpad Software Inc., San Diego, CA). Curves were constructed from the first day of treatment (i.e., First day of treatment is Day 0 on graph). The log-rank test was used for comparison between groups. A P value of 0.05 was considered significant. Results Lipopolysaccharide Preliminary experiments examined a range of doses of LPS (4, 8, 100 or 500 g), ZymA (4, 8 or 500 g) or saline administered 3 time s weekly in both male and female rats when tumors reached 10 mm in diamet er. Control animals received identical volumes of saline IT. It quick ly became obvious that, althoug h tumors grew initially in female rats, often to sizes where treatment was initiated, the tumors regressed in almost all cases, regardless of treatment, including co ntrol animals (Figure 3-1). As a result, all subsequent experiments utilized male Fisher 344 rats.
53 Initial experiments in male rats showed that multiple LPS doses (1000 or 2000 g) given 3 times per week led to tumor regressi on (Figure 3-2). Further experiments were undertaken to try and establish a dose-respon se level and effective treatment paradigm. The first experiment evaluated IT LPS at the following doses: 500 g once or 3 total doses one week apart (3X); 1000 g once or 3X; 2000 g once or 3X; and 5000 g once or 3X (n = 22). A second experiment evaluated IT LPS doses as follows: 50 g 3X; 100 g 3X; 250 g 3X; and 500 g 3X (n = 12). Control rats received IT NaCl solution 3X (n = 6). The endpoint of the experiment occurred when the tumor volume exceeded 5000 mm3 (â€œsurvivalâ€). The combined results for these 2 studies are shown in Figure 3-3 and Table 3-1. For analysis, animals were grouped as 50 to 250 g 3X, 500 to 5000 g once and 500 to 5000 g 3X. All three treatment groups show ed prolonged survival compared to control rats and were statisti cally different from each other (Table 3-1). There was no difference in survival when animals receiving 500 to 5000 g 3X were compared with animals receiving 1000 to 5000 3 times weekly until regression (data not shown). When all animals receiving multiple doses of LPS (3 total or 3 times weekly until regression) at a dose of 1000 to 5000 g were combined, 70.6% (12/17) of the tumors regressed completely and only 1 tumor eventually exceeded 5000 mm3. The remaining 4 rats were euthanatized due to unrelated causes (see below). Treatment-related morbidity in all ra ts receiving a dose of less than 5000 g was limited to local tissue swel ling. Rats receiving 5000 g often showed mild signs of illness, including periocular porphyrin accumu lation and lethargy, which resolved after
54 several days. One rat in this dose group show ed moderate lethargy after the initial dose and was found dead 24 hours after the second dose. Zymosan A Initial experiments evaluated IT ZymA at doses of 500 g, 1000 g, 2000 g, and 5000 g administered three times weekly unt il regression was observed or the tumor exceeded 5000 mm3 (n = 8). Subsequent experiments evaluated 2000 g ZymA either once or 3X (n = 10). Control rats received IT NaCl solution at similar injection schedules (n = 6). Results are shown in Figure 3-4 and Table 3-2. When all animals receiving multiple doses of ZymA (500 to 5000 g; 3 total doses or 3 times weekly until regression) were combined, 50% (6/12) of th e tumors regressed completely and only 1 tumor eventually exceeded 5000 mm3. In one rat, the tumor regressed to a very small (4.68 mm3) but persistent nodule, and the remaini ng 4 rats were euthanatized due to unrelated causes (see below). Granulocyte-Macrophage Colony Stimulating Factor GM-CSF was reconstituted in NaCl solu tion with 0.1% bovine serum albumin and evaluated IT at doses of 0.02 g 3X, 0.2 g 3X and 2.0 g 3X (n = 9). Control animals received IT NaCl solution 3X (n = 3). The results are show n in Figure 3-5 and Table 3-3. No survival benefit was noted with any of these treatment and tumors in all rats eventually exceeded 5000 mm3. Lipopolysaccharide Combined with InterferonAs LPS combined with IFNhas clear synergistic effects in vitro , a combination of these agents was evaluated for anti-tumor effects in vivo . Rats received one of the following IT treatment protocols: 1) 250 g LPS 3X, 2) 10 g IFN3X, 3) 250 g LPS
55 + 0.1 g IFN3X, 4) 250 g LPS + 1.0 g IFN3X, 5) 250 g LPS + 10.0 g IFN3X, 6) saline (n = 18). The re sults of treatment are shown in Figure 3-6. Two rats had complete regression of thei r tumors; one received 250 g LPS 3X and the second received 250 g LPS + 10.0 g IFN3X. The only statistically significant difference between groups in this experiment was seen with the combination of LPS + 0.1 g IFN, which showed prolonged survival compared with 10 g IFNalone (P = 0.0246). A number of rats were euthanized due to unrelated causes (see below). Morbidity and Mortality Unrelated to Treatment Some rats developed unilateral or bilateral pelvic limb lameness in association with swelling of the hock joint. Cytologic exam ination of synovial fluid from this joint showed increased numbers of lymphocyte s and large mononuclear cells/macrophages, consistent with an inflammatory arthritis. In addition, some rats developed tachypnea, associated with metastatic spread of the tumor to the lungs. Rats showing lameness or tachypnea were euthanatized and were censored at the time of euthanas ia in the survival analysis. The cause for the arthritis is unknown, but appear s to be unrelated to the treatment, as saline-treated animals were a ffected as well. The proportion of animals with complete tumor regression would most likely be higher had these complications not occurred, as many of the rats had shown subs tantial regression of their subcutaneous tumors at the time of censoring. Athymic Nude Rat Studies Male athymic nude rats (n = 6) were impla nted with subcutaneous RG-2 tumors as described above. When tumors reached 200 mm3, rats were treated with either 2000 g LPS IT 3X or NaCl solution IT 3X. Resu lts of treatment are shown in Figure 3-7.
56 Median survival times for both groups were identical (17 days) and there was no significant difference in survival (p = 0.5625). Rechallenge Studies Some male rats that experienced complete tumor regression after treatment with LPS or ZymA were subsequently rechallenged with a second implantation of RG-2 cells subcutaneously (n = 6). Animals were rech allenged from 229 to 367 days after initial treatment (194 to 332 days after comple te regression) with 500,000 to 1,000,000 RG-2 cells implanted adjacent to the initial tumor. Some of th ese rats formed small nodules (less than 10 mm diameter). However, thes e nodules regressed without further treatment in all animals, suggesting th e presence of immunologic memory.Three female rats that experienced complete tumor regression after ei ther LPS (n = 2) or no treatment (n = 1) were similarly protected afte r subcutaneous rechallenge. Histology and Immunohistochemistry Histological staining with hematoxylin showed tumor cells admixed with heterogeneous populations of inflammatory cells. Areas of necrosis were common, especially in tumors examined after tr eatment failure (i.e., greater than 5000 mm3). There was robust infiltration of immune cells into most of the tumors examined. A variety of cell morphologies were observed after immunohistochemical staining, some of which were unexpected. Staining for CD11b (OX-42) revealed large, oval cells within the center of the tumor, consistent with macrophages (Figures 3-8A and 3-9A). Cells with a dendritic morphology were also observed more periph erally within the tumor and peritumoral areas. The cells with macrophage morphology in the center of the tumors were also positive for CD163 (ED2), further supporti ng their likely identity. The overall
57 distribution of ED2 staining was similar, but not identical (Figures 3-8B and 3-9B). Integrin E2 (OX-62) immunostaining was not as robust, but positive cells were noted with several different morphologies. Punctate staining of large cells within the dermis and peripheral tumor was noted. In a ddition, small round cells suspected to be lymphocytes were also OX-62 positive (Figures 3-8C and 3-9C). Robust MHC II staining was noted both within and surrounding the tumors (Figures 3-8F and 3-11). Th is staining was shown mainly by macrophage-like cells in the central tumor and dendritic-like cells peri pherally. As expected, MHC I stained most cells surrounding tumors and many cells with in, including some tumor cells (Figure 3-8E). Although CD80 did not show any positi ve cells (Figure 3-8K), CD86 staining showed scattered positive cel ls with a dendritic or macrophage morphology, mainly in the edges of the tumor (Figures 3-8L and 312). Positive cells were relatively sparse compared with other antibody stains, and occu rred in groups or clumps of cells. Interestingly, CD25 immunostaining showed moderate numbers of positive cells within the tumor (Figure 3-8J). Although some cells showed the expected lymphocytic morphology, the majority either displayed a dendritic morphology or appeared to be tumor cells (Figure 3-10D). CD6 immunostaining showed cells with a small, round morphology primarily within the peripheral tumor tissue and imme diate peritumoral area, consistent with lymphocytes (Figures 3-8G and 3-10A). Immunostaining for both CD4 and CD8 showed robust staining within tumors and particularly in peritumoral areas. CD8 also stained macrophages within the tumor center, which were not CD4 positive (Figure 3-8H, I). Although cells with a sma ll, round, lymphoid morphology were noted, many positive
58 cells displayed a more stellate or dendri tic morphology consistent with DC (Figure 3-10B, C). Interestingly, immunostaining for CD161, a marker for NK cells, showed several populations of positive cells. Small to medi um-sized, round cells within the tumor and peritumoral areas were noted, consistent w ith lymphocytes. However, the large oval, macrophage-like cells within th e center of the tumor were al so positive, as were the DC-like cells in the tumor periphery a nd peritumoral area that were OX-62 positive (Figure 3-9C, D). CD161 st aining had a similar distribu tion to CD11b staining when viewed at low power (Figure 3-8D). The immunostaining for these markers was performed primarily to identify differences in cell populations between tr eatment groups. However, no consistent differences were detected. Although large numbe rs of inflammatory cells were noted in some tumors treated with LPS or ZymA, such cells were also noted in some saline treated control animals. Discussion Both LPS and ZymA led to delayed tumor growth and complete tumor regression when administered IT in this model. Th ese were dose-dependent effects, and multiple dose treatment protocols clearly outperformed single dose protocols. Experiments with rats deficient in T cells and rechallenge studies in immunocom petent rats suggest a role for T cells in these anti-tumor responses, both in the initial imm unologic response as well as in the development of long-term immunologic memory. Almost invariably (with one exception), female rats rejected RG-2 tumors after a short period of growth. This was observe d regardless of treatment, and even in saline-treated controls. There are several possible e xplanations for this observed gender
59 difference. It has long been recognized that there are gender differences in several components of the immune response. Female s are more resistant to viral (250) and parasitic (251) infections and demonstrate gr eater humoral and cell-mediated responses to exogenous antigen than males (252, 253). Th e incidence of autoimmune diseases, including multiple sclerosis, rheumatoid arth ritis, and lupus, is also higher in women (254). These differences have often been attr ibuted to the presence of sex hormones, and estrogen has been shown to promote dendritic cell viability and to increase the expression of MHC and costimulatory molecules (255). Studies in human lung cancer patients have shown that lymphocytes taken from fema les exhibit greater blastogenesis after phytohemagglutinin exposure than males (256) and male myeloma and leukemia patients receiving a female hematopoeitic bone marrow tr ansplant experience greater graft versus tumor reactions than other gender matches (257, 258). Alternatively, the rejection of RG-2 glio mas in female rats may be an entirely immune-mediated phenomenon, independent of the influence of sex hormones. The RG-2 cell line was originally cultured from a g lioma that developed in the offspring of a pregnant rat treated with N-ethyl-N-nitrosurea (2 59). However, the gender of the rat with the tumor was not disclosed. Minor histoc ompatability antigens encoded by genes found on the Y chromosome (designated H-Y) may produce an immune response in females. This has been shown in studies evaluating skin grafts from ma le mice to syngeneic females that were subsequently rejected (260, 261) and in human studies of bone marrow rejection in histocompatibility locus antigen (HLA)-matched females receiving a male marrow transplant (262)(Vogt). It is possi ble that the RG-2 cells were originally generated in a male rat and th at an immune response to such antigens lead to destruction
60 of the RG-2 cells in the female rats. Further investigation of the original identity of the RG-2 line and the role of sex hormones in this rejection process will be necessary to answer these questions. LPS has been known for decades to have anti-tumor effects when administered systemically to tumor-bearing rodents or human patients with cancer (150, 156-158, 263). Hemorrhagic necrosis of the central area of th e tumor is seen, which appears to be related to ischemia secondary to the effects of LPS on tumor vasculat ure (263). Previous studies have shown that tumor cells su rviving in the areas around the centrally necrotic areas are capable of regrowth, and that complete tumor regression after LPS administration requires an immune response involving T cells (150). Several recent studies have also demonstrated the anti-tumor properties of L PS when administered IT in a subcutaneous murine glioma model (163, 249). Several fungus-derived compounds containing -glucans have been previously evaluated for anti-tumor activity (221). A rat study evaluating 9L gliomas showed that GM-CSF, given as a constant subcutaneous infusion over 28 days (0.1-10 ng/day) in combination with intermittent injections of irradiated tumor cells, was effective in eliminating subcutaneous tumors in the c ontralateral hind limb and partly protective against intracranial tumors (223). In contrast to this, and as opposed to LPS and ZymA, GM-CSF showed no anti-tumor activity in our study. A once-weekly dose in the range used here may not be adequate in stimulating an effective immune response, and it is possible that a higher dose or a more frequent dosing schedule might l ead to an anti-tumor response. However, the total dose administered in our study was higher than in the 9L glioma study
61 previously mentioned (223). One possible e xplanation for these c onflicting results may relate to the varying immunogenicity of th e different tumor models. Whereas RG-2 gliomas are considered to be weakly i mmunogenic, the 9L gliosarcoma model has substantial inherent immunogenicity in its syngeneic host, which may facilitate its regression (193). As naturally occurri ng human tumors also display varying immunogenicity, GM-CSF may be more effi cacious against other tumor types (e.g., melanoma) than in the immunosuppressive e nvironment of most gliomas (264). These results are in line w ith recent observations that endogenous inflammatory mediators and cytokines, although effective in the activation of de ndritic cells or other antigen-presenting cells, are not sufficient to generate an e ffective helper T cell response, in contrast to microbial substances activati ng TLRs. Pasare and Medzhitov (265) showed that TLR ligation on dendritic cells was re quired for effector T cells to escape the suppressive effects of Treg cells. This activity was mediated in part by the secretion of IL-6 but could not be replaced by the admini stration of inflammatory cytokines alone. We also failed to observe an anti-tumor response with the IT administration of interferonalone (10 g 3X). It appears that such cytokines may be effective in augmenting a T cell response but without ad equate danger signals provided by TLR ligands, are ineffective at generating effectiv e T cell help and ther efore an anti-tumor response (266). The observation that multiple doses of LPS are superior to a single dose in leading to an anti-tumor effect is notable, as LPS tolerance is a well-recognized phenomenon. Animals and humans exposed to sublethal dose s of LPS become refractory to the effects of subsequent LPS administration, even at high doses (218). This is accompanied by a
62 decreased release of pro-inflamma tory cytokines, including TNF, IL-1 and IL-6 (267). However, this tolerance did not interfere w ith the anti-tumor effects seen with the multiple dosing protocols applied in our st udy. Possible explanations for this may include the temporary nature of the toleroge nic state, with rene wed receptiveness after one week (although similar effects were s een with LPS administration three times weekly), higher doses of LPS being able to overcome the tolerant st ate, or the possibility that the anti-tumor effects are mediated through mechanisms not involving traditional pro-inflammatory cytokines. Interestingly, a recent study evaluating the ability of tumor vaccines to break Treg cell-mediated tolerance found that multiple doses of LPS with dendritic cell vaccination broke tolerance and led to an effective CD8+ T cell response, while a single LPS exposure was ineffective, suggesting that persistent TLR4 signaling was critical for the anti-tumor effects (268). A number of observations suppor ted the idea that T cells we re integral in the antitumor response mediated by LPS and ZymA. Nude rats lacking T cells failed to show an anti-tumor response at an LPS dose leading to tumor regression in immunocompetent rats (Figure 3-6). In addition, F344 rats treat ed with LPS or ZymA that had complete regression of their tumor remained imm une to subsequent subcutaneous tumor rechallenge, even when implanted with very high numbers of cells up to one year later, suggesting the presence of memory T cells. Robust MHC II staining was also noted within treated tumors. Toge ther, these observations suppor t the idea of LPS or ZymA activated macrophages or dendritic cells pr ocessing and presenting tumor antigen to T cells, thus linking the innate and acquired immune responses.
63 An acquired response is likely critical to long-term protection from tumor recurrence. Traditionally, CD8+ T cells have been considered to be the effector cell responsible for killing tumor cells in tumor-immune hosts (2 69-271). However, studies in some tumor models have brought this co nclusion into question, as the tumoricidal activity of tumor-infilt rating lymphocytes is often low ( 272), cytotoxicity is limited by T cell migration away from the site of action (273) and tu mor cell death after adoptive CD8+ cell transfer may be independent fr om Fas and perforin-mediated mechanisms (232). In addition, many cancer vaccines have had poor clinical efficacy in human clinical trials; despite the generation of a robust CD8+ cell response (133, 274). These observations suggest an altern ate explanation, and it is possi ble that macrophages or other immune cells may be responsible for tumor cy totoxicity after receiv ing T cell support. In this scenario, T cells may pr ovide cytokines (e.g., interferon, IL-2) or other critical signals to facilitate tumor killing (231, 275). Macrophages have been identified as the cytotoxic effector cells in other mode ls of tumor regression after immune system activation (276), and several in vitro studies have demonstrat ed macrophage killing of tumor cells after incubation with lymphoc ytes or lymphocyte products (277, 278). Although macrophages, dendritic cells, NK cells and T cells were lik ely all involved in the anti-tumor response, identi fication of the specific effect or cells responsible for tumor cell killing in this model will require further investigation. Similarly, the contribution of regulatory T cells in this model is currently undefined. Whereas Treg cells were hypothesized to be re sponsible for the refractoriness of tumors to LPS treatment in early studies (279), more rece nt studies clearly show that
64 persistent LPS administration can overcome these effects (268). Differences in model systems may again explain some of these discrepancies. Although LPS and other microbi al products have been used in cancer therapy for many years, clinical results ha ve often been disappointing an d are limited by considerable toxicity (156, 158). Most previous studies have evaluated systemic administration (intravenous or intraperitoneal) of thes e compounds, and IT administration, as demonstrated here, may abrogate many of thes e deleterious effects while maintaining an effective anti-tumor response. Toxic e ffects were rare in our study, despite the administration of high doses of both LPS and ZymA. Consistent differences in immune cell infiltration were not identified between treatment groups. There may be a number of reasons for this observation. The primary endpoint in most of the experiments was tumor size, and histology was necessarily performed only after animals had failed treat ment. Thus, inflammatory cell populations in these animals likely do not reflect those associated with successful immunostimulatory treatment. For this reason, some animals were examined at predetermined time points, specifically 7 and 14 days after IT treatment. It is possible that this strategy missed significant inflammatory cell changes occurring either before or after these point s. Although a large number of immune cell markers were examined, it is likely that other markers potentially important to the anti-tumor effects were not. In addition, cytoki ne levels within the tumors play important roles in the anti-tumor immune response ( 268), and were not examined in our study. Functional differences in the invading i mmune cells can not be appreciated with immunostaining for selected ce llular markers. Finally, th ere is a certain amount of
65 individual variation in this system, despite using genetically similar animals. Although tumor regression was common after IT LPS and ZymA administration, this did not always happen over a consistent time frame. Based on monitoring tumor size, some of the tumors examined histologically at set time points were regressing, while some were growing larger. There appeared to be mo re intense staining for MHC II and CD161 in regressing tumors, although the numbers of tu mors examined are too small to make any objective conclusions. Clearly there was a complex interplay between the tumors and the immune system in this model. Cells with a dendri tic morphology expressed not only integrin E2, but also markers classically associated with T cells (CD4, CD8, and CD25) and NK cells (CD161). Numerous subsets of macrophages an d DC have been identified in humans and mice (280, 281) and are beginning to be r ecognized in rats (282-285). A newly recognized subset of cells expressing both DC and NK cell markers, termed interferon-producing killer de ndritic cells (IKDC) has been recently described in mice and appear to play a role in tumor su rveillance (286, 287). These IKDCs produce IFNand IL-12 upon stimulation with TLR9 liga nds, kill tumor cells, and migrate to lymph nodes where they have the capacity to present captured antigen (286) . Identification of the specific cell types involved in this model will take additional study and other experimental techniques, such as flow cyto metry. Functional analysis of immune cells after various treatments (and at a greate r number of time points) should also be considered in future studies.
66 Figure 3-1. Differences in RG-2 glioma behavi or in male versus female rats. KaplanMeier survival curve showing summ ary of 3 preliminary experiments evaluating IT LPS (4, 8, 100, or 500 g 3X weekly), ZymA (4, 8 or 500 g 3X weekly) or saline in male and female F344 rats with subcutaneous RG-2 tumors. Female rats showed signifi cantly longer survival (P < 0.0001), even when treated with saline. Tick ma rks represent censored observations. Figure 3-2. Preliminary high dos e LPS experiment in male rats with subcutaneous RG-2 gliomas. Male F344 rats given IT LPS at higher doses show prolonged survival and complete tumor regression.
67 Figure 3-3. Lipopolysaccharide dose response expe riment in male rats with subcutaneous RG-2 gliomas. Treatment with IT L PS delays tumor growth and leads to long-term survival in male F344 rats. Multiple doses were superior to a single dose within the sa me dose range. Statistical comparisons between groups are shown in Table 3-1. Figure 3-4. Zymosan A treatmen t of subcutaneous RG-2 glio mas. Treatment with IT ZymA delays tumor growth and leads to long-term survival in male F344 rats. Multiple doses were superior to a single dose. Statistical comparisons between groups are shown in Table 3-2.
68 Figure 3-5. Granulocyte-macrophage co lony stimulating factor treatment of subcutaneous RG-2 gliomas. Intratumoral GM-CSF does not show antitumor effects in male F344 rats wi th subcutaneous RG-2 gliomas. Statistical comparisons between groups are shown in Table 3-3. Figure 3-6. Combination LPS and IFNtreatment of subcutaneous RG-2 gliomas. Intratumoral IFNalone appeared to shorten surivival, although this was only significant when compared to the LPS + 0.1 g IFNgroup (P = 0.0246). There was no difference in survival between any other groups.
69 Figure 3-7. Treatment of s ubcutaneous RG-2 gliomas in male athymic nude rats (rnu/rnu). Intratumoral LPS does not s how anti-tumor effects in nude rats with subcutaneous RG-2 gliomas.
70 Figure 3-8. Immunohistochemical evaluation of a subcutaneous RG-2 glioma (low power). Sections are from similar areas of a tumor in a rat treated with IT LPS (2000 g) 7 days previously. Tumo rs are stained with the peroxidase method and counterstained with he matoxylin. The dotted line in A delineates the tumor edge, and is si milarly oriented in all frames. Immunostained molecule with clone designations in paretheses are as follows: A) CD11b (OX-42) B) CD163(ED2) C) Integrin E2 (OX-62) D) CD161 (3.2.3) E) MHC I (OX-18) F) MHC II (OX-6) G) CD6 (OX52) H) CD4 (W3/25) I) CD8 (OX-8) J) CD25 (OX-39) K) CD80 (3H5) L) CD86 (24F).
71 Figure 3-9. Immunohistochemical staining in a subcutane ous RG-2 glioma (higher power). Sections are from similar areas of a tumor in a rat treated with IT saline 7 days previously. Tumors ar e stained with the peroxidase method and counterstained with hematoxylin. Immunostained molecule with clone designations in paretheses are as foll ows: A) CD11b (OX-42). Most cells have a large, round to oval mor phology characteristic of macrophages (arrow). B) CD163 (ED2). Cells wi th both a macrophage (arrowhead) and dendritic (arrow) morphology are seen. C) Integrin E2 (OX-62). Larger cells with a punctuate staining patte rn (arrow) and small round cells suggestive of lymphocytes (arrowhead ) are indicated. D) CD161 (3.2.3). Similar to C, large cells with punctuate staining (a rrow) and smaller, round, lymphocytic cells (arrowheads) are stained positive.
72 Figure 3-10. Immunohistochemi cal staining in a subcutaneous RG-2 glioma (higher power). Sections are from similar areas of a tumor in a rat treated with IT saline 7 days previously. Tumors ar e stained with the peroxidase method and counterstained with hematoxyli n. Immunostained molecule with clone designations in paretheses are as follows: A) CD6 (OX-52). Immunopositive cells are almost exclusively small round cells characteristic of lymphocytes. B) CD4 (W3/25). Infiltration of small round lymphocytes (arrowheads) is not ed, but cells with stellate and dendritic morphologies are also seen (a rrow). C) CD8 (OX-8). Similar to B, lymphocytic cells (arrowheads) and dendritic-like cells (arrows) are both noted. Large round cells characte ristic of macrophages also stained positive (not indicated). D) CD25 (OX-39). Multiple cell types were immunopositive, including lymphocytes , cells with dendritic morphology and what appeared to be tumor cells (arrows).
73 Figure 3-11. Immunostaining for MHC II in a subcutaneous RG-2 glioma. Sections are from a rat treated with IT saline 7 da ys previously. Tumors are stained with the OX-6 antibody using the per oxidase method and counterstained with hematoxylin. Robust staini ng of cells with both macrophage (arrowheads) and dendritic (a rrows) morphologies are noted. Figure 3-12. Immunostaining for CD86 in a s ubcutaneous RG-2 glioma. Sections are from a rat treated with IT saline 7 da ys previously. Tumors are stained with the 24F antibody using the peroxidase method and counterstained with hematoxylin. Intermittent stai ning of cells, often in small groups (arrows) was noted within the tumors . Immunopositive cells showed both macrophage and dendritic morphologies.
74 Table 3-1. Intratumoral lipopol ysaccharide for subcutaneous RG-2 gliomas: Treatment outcomes Comparison vs. other groupsa Treatment Group Median Survival (Days) A B C D Animals with Complete Regression (%) A Saline 15 0.0029<0.0001 <0.0001 0 B 50-250 g 3X 36 0.0029 0.0164 0.0143 0 C 500-5000 g once 53 <0.0001 0.0164 0.0476 0 D 500-5000 g 3X Undefined <0.0001 0.01430.0476 50 aP values calculated using the log-rank test for comparison of survival curves generated using the Kaplan-Meier method (Figure 3-1) Table 3-2. Intratumoral zymosan A for s ubcutaneous RG-2 gliomas: Treatment outcomes Comparison vs. other groupsa Treatment Group Median Survival (Days) A B C D Animals with Complete Regression (%) A Saline 17 0.01670.0120 <0.0001 0 B 2000 g once 39.5 0.0167 0.1092 0.0014 0 C 2000 g 3X Undefined 0.0120 0.1092 0.2207 50 D 500-5000 g UR Undefined <0.0001 0.00140.2207 50 aP values calculated using the log-rank test for comparison of survival curves generated using the Kaplan-Meier method (Figure 3-4) Table 3-3. Intratumoral granulocyte-macrophage col ony stimulating factor for subcutaneous RG-2 gliomas: Treatment outcomes Comparison vs. other groupsa Treatment Group Median Survival (Days) A B C D Animals with Complete Regression (%) A Saline 18 0.02950.6141 0.4876 0 B 0.02 g 3X 16 0.0295 0.0295 0.0295 0 C 0.2 g 3X 18 0.6141 0.0295 0.4876 0 D 2.0 g 3X 17 0.4876 0.02950.4876 0 aP values calculated using the log-rank test for comparison of survival curves generated using the Kaplan-Meier method (Figure 3-5)
75 CHAPTER 4 COMBINATION RADIATION AND IMMUNOTHERAPY IN SUBCUTANEOUS RG2 GLIOMAS Introduction Although immunotherapy may be successful in generating an anti-tumor response, it also has the potential to create adverse e ffects related to inflammation in the region of treatment and possibly immune-mediated attack of normal tissue structures. Reducing the potential for these side effects by utilizing the lowest effective dose of immunostimulant should be a tr eatment goal. This might be facilitated by combining immunostimulatory therapy with a second ther apy aimed at killing or removing tumor cells. In addition, the current standard of care for malignant gliomas includes surgical debulking, fractionated external beam radi ation therapy, and often chemotherapy, and any future therapy will almost certainly need to be evaluated in combination with these standards (288, 289). Cytotoxic therapies have often been view ed as being potentially antagonistic to immunotherapy, due to their potential immunosu ppressive effects and the fact that they kill cells mainly through apoptotic m echanisms, which has been viewed as immunologically benign and potentially toleroge nic. Radiation therapy kills tumor cells primarily through the generation of oxygen fr ee radicals, which damage DNA and trigger apoptotic cell death. Chemotherapy agents have a variety of different ways of interfering with cellular processes and killing tumor cells, but most also lead to apoptotic cell death (290, 291).
76 Apoptotic cell death is a frequent occu rrence during development and normal body functioning, and has usually been regarded as being immunologically bland or tolerizing. However, it is likely that not al l apoptotic death is alike, and in some cases, apoptosis can trigger a powerful immune reaction (292). Release of large amount s of antigen due to massive cell death may occur after chemot herapy or radiation treatment, and may overwhelm the immune systemâ€™s tolerogenic mechanisms (291). Thus, in appropriate amounts and with appropriate alerting or danger signals, apoptotic cell debris may sensitize, not tolerize the immune system to the tumor (291, 293). There are other potential mechanistic be nefits associated with combining immunotherapy with a cytotoxic therapy such as radiation or chemotherapy. The first is debulking of the tumor mass, which reduces the tumor burden and the likelihood of antigen loss variant cells among the surviving population (29 1, 294). Cytotoxic therapies may increase the infiltration and activation of immune cells (291, 294, 295). In the brain, radiation therapy can disrupt the blood-brain barrier (296298) and upregulate adhesion molecules for immune cells in the local va sculature (299, 300). Radiation therapy may also contribute to the generation of a proinflammatory environment (300), and may counteract the immunosuppre ssive effects of TGF(301). Perhaps most importantly, cytotoxic therapies can lead to the release of increased am ounts and a greater variety of tumor antigen for processing by cells of th e immune system (291, 293, 294). In some cases, the cytotoxic therapy may produce anti gens not present on untreated tumor cells, allowing for exquisite sensitivity when combined with immunotherapy (302). Results in other experimental tumor m odels have supported this rationale. Combinations of radiation therapy and Cp G oligodeoxynucleotides were effective in
77 delaying tumor growth in rodents with either muscle (303) or brain tumors (304). CpG oligodeoxynucleotide administra tion with chemotherapy has also shown efficacy in experimental models (305-307). Radiati on or chemotherapy have been combined successfully with other immunotherapeutic approaches, including other nonspecific immunostimulants (308), TNF-a gene therapy (309), FasL administration (310), adoptive T cell therapy (311), cellul ar vaccination protocols ( 312, 313) or dendritic cell administration (314-316). Other immunostimul atory and cytotoxic combinations have also been used successfully (317). Materials and Methods Animals, cell lines, tumor implantation and monitoring procedures were similar to those described in Chapter 3. Treatment was again initiated when the tumors reached a volume of 200 mm3 and animals were euthanized if the tumor exceeded 5000 mm3. Tumor Treatments Animals were anesthetized with isoflura ne in oxygen for treatments, which were performed in conjunction with tumor measurements. Tumors were treated when they reached a volume of 200 mm3. Lipopolysaccharide (strain O5:B5, Sigma-Aldrich, St. Louis, MO), was reconstituted in NaCl solu tion (0.9%), standardized to a volume of 40 l and administered with a tuberculin syringe and 27 gauge needle dire ctly into the tumor (intratumorally [IT]). For radiation treatment, a 6 MeV lin ear accelerator (CLINAC 600C, Varian Medical Systems, Palo Alto, CA) was used to deliver a horizontal photon beam to encompass the entire tumor. The treatme nt field was 4 cm (horizontal) by 8 cm (vertical). The tumor was raised dorsally wi th the aid of a custom-made clamp device and the majority of the body shielded with a l ead block in order to avoid irradiation of
78 critical structures (e.g., spinal cord). Tissue-equivalent ma terial was placed over the mass to provide the build-up required to deliver the correct radiation dose. A variety of doses were given as a single treatment (see below). Radiation Dose-Response Study A dose response experiment was performed to evaluate the response of the subcutaneous tumors to irradiation. Rats received a single treatment dose of 5, 10, 15, 25 or 50 Gray (Gy) when tumors reached 200 mm3 (n = 23). Control rats were not irradiated (n = 2). Combination Treatment Study The results of the radiation and imm unostimulatory dose response experiments were used to design an experiment evaluati ng a combination of these two therapies. When subcutaneous tumors reached 200 mm3, animals were assigned to one of the following treatment groups: 5 Gy, 10 Gy, 100 g LPS 3X, 5 Gy + 100 g LPS 3X or 10 Gy + 100 g LPS 3X (n = 15). Control rats received IT NaCl so lution (n = 2). For rats receiving combination thera py, the initial dose of LPS wa s administered immediately following irradiation. Histology and Immunohistochemistry Histological and immunohistochemical examination of tumors from rats euthanatized because of tumor volumes in excess of 5000 mm3 was performed in some cases. In addition, a number of rats (n = 6) were sacrifi ced at specific time points after radiation and immunostimulatory treatment (7 or 14 days) to try and define differences in cellular infiltration and cell surface markers between treatment groups. The procedures used to process and stain tumor tissue and the antibodies used are described in Chapter 3.
79 Statistical Analysis Survival curves were constructed usi ng the Kaplan-Maier product limit method with a statistical software program (Prism 4.0, Graphpad Software Inc., San Diego, CA). The log-rank test was used for compar ison between groups. A P value of 0.05 was considered significant. Results Radiation Dose-Response Study The median survival times and comparis ons between groups are shown in Figure 4-1 and Table 4-1. There was a clear dose-de pendent response of the tumors to the radiation therapy. Whereas the animals receivi ng 5 Gy showed no increase in survival compared to control rats, all of the other treatment doses showed delayed tumor growth and prolonged survival. Complete tumor regr ession was only seen in the 50 Gy treatment group, in which 75% of the rats were cured of their tumors. Adverse effects were limited to moist dermatitis in rats receiving 50 Gy, which resolved without treatment. Combination Treatment Study The median survival times and comparis ons between groups are shown in Figure 4-2 and Table 4-2. Similar to the radiation dose response experiment, rats receiving 5 Gy of radiation showed no improvement in surviv al, while those treated with 10 Gy showed a modest survival advantage. Treatment with LPS alone also showed a small but significant survival advantage. In contra st, both groups receivi ng combination therapy showed prolonged survival when compared with control rats or t hose receiving single treatments only. In addition, 33% of rats treated with 10 Gy + LPS showed complete tumor regression. No adverse effects rela ted to treatment were noted in our study.
80 Rechallenge Studies Three rats treated with radiation at 50 Gy that had complete tumor regression were subsequently rechallenged with a second subcutaneous injection of approximately 500,000 tumor cells. Although all 3 rats devel oped moderate-sized nodules, with peak volumes of 483 mm3, 420 mm3 and 396 mm3, interestingly, in all 3 cases the tumors subsequently regressed. These rats were then subsequently subjected to an intracranial challenge of RG-2 cells (Chapter 5). Morbidity and Mortality Unrelated to Treatment A proportion of rats developed lameness sec ondary to swelling of the hock joint or tachypnea related to metastasis of the tumor to the lungs. If euthanasia was necessary due to these complications, animals were censore d in the survival anal ysis as described in Chapter 3. Histology and Immunohistochemistry Histological and immunohistoche mical evaluation of tissues was very similar to the results reported in Chapter 3. There was r obust infiltration of a variety of different immune cells into the tumors. Cells w ith a macrophage morphology that were CD11b and ED2 positive were prominent within the center of the tumors, while cells with lymphoid morphologies and staining positive for CD6, CD4, and CD8, were generally found in the tumor periphery and peritumoral su bcutaneous tissues. Cells with dendritic morphology and staining positive for CD4, CD8, CD25, integrin E2 and CD161 were noted as described in Chapter 3. Simila rly, no clear differences were noted in immunostaining between treatment groups.
81 Discussion Combining NSI with a cytotoxic treatment has the potential to increase efficacy and allow dose reductions of both modalities, which may also limit toxicity. We show here that the combination of external beam radiation and IT LPS produces a much greater anti-tumor effect than either modality alone (Figure 4-2). Similar observations have been made by others after combina tions of radiation and TLR9 ligands (303, 318). There are many possible explanations for the contribution of cytotoxic therapy to this synergistic effect, including debulking of the tumor mass, increased infiltrati on and activation of immune cells, and the release of increased amounts and a greater variety of tumor antigen for processing by these cells (291). Treatment with a single radiation dose ha d a dose-dependent e ffect in delaying tumor growth in this model. Traditional radi ation therapy protocols, however, divide the total radiation dose into fractions, which allows cells to repair sublethal damage in the time between treatments. As most healthy tissu e is more efficient at this repair process than tumor tissue, fractionation maximizes tumor death while sparing normal tissue within the radiation field, a nd is the basis for conventional radiotherapy. A single dose of radiation was chosen here primarily for c onvenience and availability of the radiation facility. Although adjustment factors are av ailable to convert single high doses into fractionated equivalents, a vari ety of other factors would caut ion against extrapolating the synergism between NSI and radiation seen here to a fractionated scenario. These would include the effect of radi ation dose in leading to pr oinflammatory or possibly anti-inflammatory effects (299, 300) and th e unknown effect of multiple doses on the immune cell population within and surrounding the tumor. For example, it is possible that multiple doses may sequentially kill infiltrating lymphocytes and abrogate a
82 potentially beneficial immune response. As a result, experimental evaluation of NSI in conjunction with fractionated radiotherapy s hould be considered in the future. Stereotactic Radiosurgery (SRS) is a technique using a fo cused, high dose of radiation, delivered as a single fraction to pati ents with a variety of conditions, primarily brain tumors (319, 320). It has been extensiv ely evaluated in the tr eatment of malignant gliomas, with mixed results (242, 321-323). A randomized, trial comparing SRS as a boost to conventional EBRT failed to show a beneficial effect in patients with GBM (324). However, the topic remains controversia l, and some clinicians believe that SRS may still benefit a certain subset of patients (325, 326). The radiation doses used in our study are very similar to those used in SRS for clinical patients, and it is possible that a combination of NSI and SRS may prove bene ficial for these patients in the future. The resistance of rats cured of subcut aneous tumors with high dose (50 Gy) radiation therapy to subsequent tumor rech allenge was an interesting and unexpected finding. It suggests that thes e animals developed an effec tive immune response to the original tumors that involved memory T ce lls. One possible explanation is that the tumors were inherently immunogenic, and th at a priming â€œdoseâ€ of tumor antigen is sufficient to confer future immunity. Howe ver, the RG-2 line has been shown to be weakly immunogenic in the pas t; that is, animals having an original tumor surgically removed and inoculated with non-viable (irr adiated) RG-2 cells are not protected from subsequent tumor rechallenge ( 113, 193). Fisher 344 rats with subcutaneous tumors that are cured with IT mitoxantrone (a chemothe rapeutic agent) using this identical model system, are not protected from rechallenge with tumor cells (data not shown). In addition, of 3 athymic nude rats treated with 50 Gy radiation, only 1 had complete tumor
83 regression and this rat was not protected fr om a subcutaneous rechallenge (data not shown). This suggests that immune system activity may play a role in the treatment responses attributed to single high-dose radiation therapy. Figure 4-1. Dose-response experiment of subcutaneous RG-2 gliomas treated with external beam radiation therapy (E BRT). Single fraction EBRT shows a dose-dependent delay in tumor growth in male F344 rats. Comparisons between groups are shown in Table 4-1. Table 4-1. Single-dose external beam radiation therapy for subcutaneous RG-2 gliomas: Treatment outcomes Comparison vs. other groupsa Animals with Complete Regression (%) Treatment Group Median Survival (Days) A B C D E F 0 A Saline 26 0.25600.01770.00820.00820.0177 0 B 5 Gy 40 0.2560 0.35490.04990.00420.0101 0 C 10 Gy 53 0.0177 0.3549 0.71740.09810.0510 0 D 15 Gy 50 0.0082 0.04990.7174 0.01980.0170 0 E 25 Gy 76 0.0082 0.00420.09810.0198 0.0620 0 F 50 Gy Undefined 0.0177 0.01010.05100.01700.0620 75 aP values calculated using the log-rank test for comparison of survival curves generated using the Kaplan-Meier method (Figure 4-1)
84 Figure 4-2. Combination therapy with EBRT and LPS in rats with subcutaneous RG-2 gliomas. Combination therapy shows synergistic effects in delaying tumor growth and leads to long-t erm survival in male F344 rats with subcutaneous RG-2 gliomas. For statistical co mparison between treatment groups, see Table 4-2. Table 4-2. Combination singledose external beam radiatio n therapy and intratumoral lipopolysaccharide for subcutaneous RG -2 gliomas : Treatment outcomes aP values calculated using the log-rank test for comparison of survival curves generated using the Kaplan-Meier method (Figure 4-2) bLPS dose for all groups was 100 g 3X Comparison vs. other groupsa Animals with Complete Regression (%) Treatment Group Median Survival (Days) A B C D E F 0 A Saline 19.5 0.03890.11610.03890.0389 0.0389 0 B LPSb 25 0.0389 0.30180.85840.0295 0.0295 0 C 5 Gy 21 0.11610.3018 0.34300.0246 0.0246 0 D 10 Gy 24 0.03890.85840.3430 0.0363 0.0246 0 E 5 Gy + LPS 53 0.03890.02950.02460.0363 0.5151 0 F 10 Gy + LPS 63 0.03890.02950.02460.02460.5151 33
85 CHAPTER 5 NONSPECIFIC IMMUNOTHERAPY WITH OR WITHOUT RADIATION IN INTRACRANIAL RG-2 GLIOMAS Introduction A subcutaneous location is obviously not idea l to model the clinical situation seen in human glioma patients. Nevertheless, it provided a convenient wa y to screen initial immunostimulatory treatment substances and dos ing protocols, facilitating treatment as well as serial monitoring of tumor size. It is possible that protocols effective in subcutaneous tumors will not be effective when the same cell type is implanted intracranially, which has been suggested by other investigators (249). The local environment of the brain may not be as c onducive to the generation of an effective immune response as the subcutaneous comp artment. Although microglia are able to present antigen, they are less efficient at this task than dendritic cells (71). As dendritic cells are not normally present within the brain, it is possible that this treatment paradigm may fail in an IC setting. However, LPSactivated microglia have been shown to kill tumor cells in vitro (184, 186), and IT injection of CpGs (a TLR9 ligand) was effective in prolonging survival in an intr acranial CNS-1 rat model of glioma (174). In addition, there are many anecdotal observations of tumor regression in patients developing infections, including patients with gl iomas (327). Therefore, nonspecific immunostimulation may still be effic acious against intracranial lesions. As mentioned in Chapter 3, IT treatmen t has many potential advantages over systemic therapy, including maximizing the eff ects at the intended site of action, and
86 reduction of the total administered therapeu tic dose, which may reduce potential side effects. For brain tumors, IT administration also circumvents the difficulties associated with getting therapeutic s ubstances past the blood-brain barrier, which can be a formidable task. This barrier can impede the passage of pharmaceuticals, biological agents and cells from the systemic circulation into the brain. It consists of multiple strucutes, including tight junc tions between endothelial cells, pericytes, and astrocytic foot processes. Transport proteins are also involved, such as P-glycoprotein, which pumps selected molecules that have entered th e brain back into the systemic circulation. Based on the results of the studies performe d in the subcutaneous RG-2 model, IT LPS and combinations of IT LPS with IFNand IT LPS with radiation were evaluated in intracranial RG-2 gliomas. In addition, subcutaneous admini stration of irra ditated tumor cells with LPS as a treatment for intracranial tumors was also examined. Materials and Methods Animals Fisher 344 (F344) rats weighing 175-200 g were obtained from Harlan. Housing and feeding were as described in previous chapters. In addition, a number of male and female rats that had experienced complete regression of subcutaneous tumors after treatment with nonspecific immunotherapy or radiation were subsequently rechallenged with an intracranial injection of tumor cells . All experimental animal procedures were approved by the Institutional Animal Care and Use Committee at the University of Florida. Tumor Implantation Two experimental implantation paradigms were developed, depending on the need for single or multiple intracrania l injections of immunostimulatory substances. Rats were
87 anesthetized with isoflurane in oxygen. The hair over the skull was clipped, the skin surgically prepared, and the animals were plac ed into a stereotactic head frame apparatus (David Kopf Instruments, Tujunga, CA). A midline incision was made in the skin over the skull and the skin and subcutaneous tissu es were reflected laterally. A burr hole was made in the skull 3 mm lateral to bregma with the aid of an electric drill. For animals being treated with a single immunostim ulatory treatment, approximately 10,000 tumor cells were injected with a Hamilton syringe in to the right striatum (5 mm ventral to the dura). The injection was made over a period of 2 minutes, after which the syringe was slowly withdrawn. The burr hole was filled wi th bone wax and the skin incision closed with autoclips. Buprenorphine (0.02-0.04 mg /kg subcutaneously) was administered for post-operative analgesia. For animals to be treated with multiple immunostimulant doses, the animal was prepared for surgery, a skin incision made a nd a burr hole drilled as described above. A teflon guide cannula (Plastics One, Roanoke, VA) was then lowered into the brain at the same coordinates (3 mm lateral to bregma, 5 mm deep to the dura). Three plastic screws were then placed into the skull and secured to the guide cannula with polymethylmethacrylate bone cement (Lang Dental Manufacturing Company, Wheeling, IL). A dummy cannula (Plastics One) was pl aced into the guide cannula to maintain patency. After cement curing, the skin was closed around the edges of the polymethylmethacrylate with autoclip s. Buprenorphine (0.02-0.04 mg/kg subcutaneously) was administered for post-operative analgesia. The cannula was then left in place for 6 to 7 days until tumor impl antation. For this procedure, rats were anesthetized and the tumor cells (approximately 10,000) were implanted using an internal
88 cannula (Plastics One), whic h fit through the guide cannula into the striatum. A Hamilton syringe attached to the internal cannula with polye thylene tubing was used to deliver the cells. Tumor Treatments Based on the results obtained in the subcut aneous RG-2 model, intracranial tumors were treated with either LPS alone or a comb ination of LPS and external beam radiation therapy. For single injections of LPS into established tumors, a skin incision was made and the subcutaneous tissues reflected laterally. The burr hole used for tumor implantation was again used for treatment after removal of the bone wax. A Hamilton syringe with attached needle was used to deli ver the LPS. The depth of needle placement was either 5 mm below the dura or was ba sed on pre-operative magnetic resonance imaging scans, when available. Injection of LPS or saline (controls) was made over two minutes, after which the needle was slowly withdrawn. Bone wax was replaced in the burr hole, the wound was closed and buprenor phine was administered as described previously. For multiple LPS injection paradigms, treatment was delivered with an internal cannula and attached Hamilton sy ringe through the existi ng guide cannula over two minutes. Radiation Treatment and Combination Therapy For irradiation of intracranial tumors, rats were anesthetized with isoflurane in oxygen and placed in sternal recumbency. A vertical photon beam from a 6 MeV linear accelerator (Chapter 4) was used for treatmen t. A 14-mm diameter, circular collimator was used to shape the beam and it was targ eted based on anatomical landmarks of the skull and the site of previous surgery, or in the case of animals with a guide cannula in
89 place, was targeted based on this. The diameter of the beam was subs tantially larger than the diameter of the tumor, in order to avoid missing the edges of the target. For combination therapy with radiation a nd LPS, intracranial tumors were first treated with radiation as desc ribed above. An IT injection of LPS was then immediately administered while the rat was still anestheti zed, and the animals were then subsequently recovered. Repeat IT LPS treatments were ma de at weekly intervals, but the radiation therapy was always limited to a single treatment. Vaccination with Irradiated Tumor Cells To evaluate treatment of an establis hed intracranial RG-2 glioma with a subcutaneous vaccination protocol, rats were in jected subcutaneously with either healthy live or irradiated RG-2 cells in combination with LPS. The animals were implanted with tumor cells in the right striatum using a Ham ilton syringe as described above. The rats were imaged with MRI on Day 8 and then tr eated on Day 9 after tumor implantation. Live RG-2 cells were grown as previously de scribed. For irradiation, cells were grown as previously described in 175 cm2 flasks. When the cells neared confluence, they were irradiated at 100 Gy with a 6 MeV linear acceler ator. Both live cells and irradiated cells were trypsinized, collected, washed and count ed with a hemacytometer before injection, which was performed between the shoulder blad es in anesthetized rats. Injection of tumor cells (500,000 cells) was performed firs t, followed immediately by LPS (2000 g) injection into the same region. These subcut aneous treatments were then given every 7 days for a total of 3 treatments or until euthanasia. Rechallenge Studies Some animals that had completely reject ed subcutaneous tumors after treatment were rechallenged with an intracranial inject ion of RG-2 cells. These injections were
90 made with cells grown in a standard ma nner using a Hamilton syringe as described above. The time from tumor regression to r echallenge was variable. Some of these animals were female rats that had spontane ous tumor regression (i.e., no treatment). In addition, some of these animals had had a prio r subcutaneous recha llenge that had also regressed or were concurrent ly rechallenged with both subc utaneous and intracranial injections. Magnetic Resonance Imaging For MRI, rats were anesthetized and instrumented as described in Chapter 2. All animals were imaged in the 4.7 T Oxford Magnet with Paravision software. Imaging parameters were identical or very similar to those described in Chapter 2, and T1-weighted, T2-weighted and postgadodiam ide T1-weighted images were acquired. Histology and Immunohistochemistry Rats were euthanized when they showed signs of increased intracranial pressure, including lethargy or obtundation, reluctance or inability to ambulate, porphyrin staining of eyes or muzzle and lack of grooming. Th ese signs typically precede death from tumor growth by approximately 24 hours. At necrops y, most rats were noted to have caudal displacement of the cerebellum into the foramen magnum, consistent with increased intracranial pressure. Survival Calculation and Statistical Analysis The date of death due to eu thanasia or spontaneous causes was used for survival analysis. Survival curves were constructe d using the Kaplan-Maier product limit method with a statistical software program (Prism 4.0) . Curves were constructed from the day of tumor implantation (i.e., Day of implantation is Day 0 on graph). The log-rank test was used for comparison between groups. A P va lue of 0.05 was considered significant.
91 Results Treatment of Intracranial Tumors with Intratumoral Nonsp ecific Immunotherapy Alone Lipopolysaccharide and InterferonMale F344 rats (n = 8) were implanted with intracranial tumors using a Hamilton syringe. On Day 10 after implantation, the ra ts were imaged with MRI, which showed small to moderate sized contrast enhancing ma sses in the right striatum of most of the rats (Figure 5-1A, B). The inte nt in this experiment was to treat the tumors when they had reached a size identifiable by MRI..One rat developed a systemic illness and was not included in the treatment groups. The followi ng day (Day 11), most of the rats were treated with either LPS (4 g) and IFN(100 ng; n = 3) combined in the same Hamilton syringe (total volume 8 l) or 8 l saline as a control (n = 3). A second MR imaging session was conducted a week following the firs t (Day 17) and showed enlarged tumors in all cases (Figure 5-1C, D). On the followi ng day (Day 18), rats st ill alive received a second dose of LPS/ IFNor saline and the remaining untreated rat was treated with the same dose of LPS and IFN. Survival curves for these animals are shown in Figure 5-2. There was no difference in survival betw een the two groups (P = 0.5498). In fact, several rats treated with LPS and IFNdied on the day of or the day following treatment. All rats had signs of increased intracranial pressure, and it appeared that the inflammatory response related to treatment led to swelling and subsequent death. Lipopolysaccharide After the results of the subcutaneous e xperiments revealed that IT LPS may be effective when used alone, a small experiment was designed to preliminarily evaluate IT
92 administration of LPS to treat intracranial tumors . Male F344 rats (n = 6) had intracranial guide cannulas placed as described above. One rat died afte r the procedure was completed. RG-2 cells were injected into the remaining animals 6 days later (Day 0). On Day 9, the rats were treated with either 5 g LPS or 10 g LPS IT (n = 2 animals per group). The remaining rat pulled out th e guide cannula on Day 9 and was followed without treatment as a control. Survival curves for these animals are shown in Figure 5-3. There was no difference in survival be tween the groups (P = 0.5780) and rats treated with immunostimulatory th erapy died shortly following the second IT treatment. These rats again had signs of increased intracranial pressure. Radiation Dose-Response Experiment for Intracranial RG-2 Tumors Based on the results obtained with combin ation therapy in subcutaneous tumors, it was hypothesized that combination of radiation therapy with LPS may be able to increase the therapeutic index in intracranial tumors, th at is, lead to improve d anti-tumor efficacy with reduced side effects. To provide a baseline indication of the responsiveness of intracranial tumors to radi ation therapy, a dose-response experiment was performed. Male F344 rats (n = 14) had RG-2 cells im planted into the right striatum with a Hamilton syringe. Magnetic resonance imag ing was performed on Days 16, 17 and 23 to monitor tumor size. On Day 9 after tumor im plantation, rats were tr eated with a single dose of radiation (5, 10, 15, 25 Gy; n = 3 rats pe r group) or followed without treatment (n = 2) as controls. Survival curves for these animals are shown in Figure 5-4. Only rats treated with 25 Gy showed prolonged survival compared to untreated controls (P = 0.0389). All animals died or were euthanized with signs of increased intracranial pressure.
93 Combination Lipopolysaccharide and Radi ation Therapy for Intracranial RG-2 Tumors Treatment of tumors on Day 9 Based on the results of the intracranial dose-response experiment, an experiment evaluating combination therapy with LPS and radiation was designed. Male F344 rats (n = 20) had intracranial guide cannulas placed as described above on two consecutive days. RG-2 cells were injected into the remaining an imals 6 to 7 days later (Day 0). On Day 9, the rats were randomly assigned to one of the following treatment groups: 1) 10 Gy radiation (n = 3) 2) 15 Gy radi ation (n = 3) 3) 10 g LPS IT (n = 4) 4) 10 Gy radiation + 10 g LPS IT (n = 4) 5) 15 Gy radiation + 10 g LPS IT (n = 4). Only a single dose of radiation and a single IT LPS treatment wa s given. Two additional rats served as untreated controls. Survival curves for these animals are shown in Figure 5-5A and comparisons between groups are shown in Table 5-1. In this experiment, animals treated with 15 Gy of radiation had prolonged survival compared with controls (P = 0.0389) alt hough the 10 Gy group did not. The animals receiving LPS alone showed decreased surviv al compared with controls (P = 0.0446) and the single modality radiation groups (P = 0.0189). In addition, the rats receiving combination therapy did not show prolonged surv ival when compared to any of the other treatments. These results can be attributed to toxicity related to LPS treatment. Significant morbidity and mortality was noted on the day following IT LPS therapy. Similar to the previous studies described above, this was associated with cerebral edema, swelling of the brain and herniation of th e intracranial contents through the foramen magnum. If animals dying as a result of th e therapy are censored from the survival analysis, there is a trend towards longer su rvival in those treated with combination
94 therapies (Figure 5-5B). However, the numbers of animals are very small after censoring, and differences do not reach statistic al significance (Salin e vs. 10 Gy + LPS: P = 0.0896; Saline vs. 15 Gy: P = 0.2253). Treatment of tumors on Day 3 The mortality associated with brain hernia tion is directly related to intracranial pressure, which in turn depends on the volum e of tissue, blood a nd cerebrospinal fluid within the confines of the in tracranial vault. RG-2 tumors reach a substantial size by 9 days post-implantation, and contri bute to increases in intracranial pressure. To evaluate combination therapy while attempting to minimize some of the toxicity associated with therapy, treatment of a small number of ra ts was attempted on Day 3 post-implantation. It was also anticipated that this would allo w for the administration of multiple LPS doses, which had been more effective in the subcutaneous trials. Male F344 rats (n = 6) had intracranial guide cannulas placed as described above and RG-2 cells were implanted 7 days later (Day 0). On Day 3, animals received one of three treatments: 1) 15 Gy radiation + saline IT 2) 15 Gy radiation + 5 g LPS IT 3) 15 Gy radiation + 10 g LPS IT. Only a single radiati on treatment was administered, but rats received IT LPS or saline weekly fo r 3 doses (3X). The initial LPS or saline treatment was administered with an internal cannula through the implanted guide cannula immediately following the radiation treatment. The results of this experiment are shown in Figure 5-6. This treatment paradigm did allow the ad ministration of 3 doses of LPS, although the rats showed significant morbidity followi ng the treatments. This was characterized by porphyrin staining of the eyes and muzzle, lethargy and in some cases, walking in circles towards the side of the tumor. Magnetic resonance imaging of the rats was
95 performed on Days 9 and 16 and showed hyperi ntensity involving the majority of the treated hemisphere on T2-weighted images, cons istent with cerebral edema (Figure 5-7). Increased mortality was noted following the thir d IT treatment, and there was no survival differences between the trea tment groups (P = 0.8607). Treatment of Intracranial RG-2 Gliomas with Subcutaneous Administration of Irradiated Tumor Cells and Lipopolysaccharide To evaluate the potential for subcutaneous vaccination with irradiated tumor cells and LPS to treat intracranial tumors, this appr oach was evaluated in male F344 rats (n = 12) with established intracranial RG-2 glioma s. On Day 9 after implantation, rats were treated with one of 6 protoc ols: 1) Irradiated RG-2 cells (500,000 cells) + LPS (2000 g) 2) Viable RG-2 cells + LPS 3) Irradiated RG-2 cells alone 4) Viable RG-2 cells alone 5) LPS alone 6) Saline control. Treatments were planned for week ly administration of 3 total doses, although no rats survived until th e third dose. Monitoring with MRI was performed on the day before initia l treatment (Day 8) and on Day 15. The survival curves for this experiment ar e shown in Figure 5-8. No difference in survival was noted between treatment groups (P = 0.6112). Magnetic resonance imaging showed steady tumor growth in all rats. Rats injected with viable RG-2 cells subcutaneously generally developed sma ll nodules, with the exception of one rat receiving viable cells with LPS. As in pr evious studies, the animals were euthanized because of signs of increased intracranial pressure, including lethargy, porphyrin staining of eyes and muzzle and turning in circle s towards the side of the tumor. Rechallenge Studies A number of rats (n = 22) that had co mplete regression of their subcutaneous tumors were subsequently rechallenged in tracranially with RG-2 cells (10,000 cells as
96 above). There were 3 female and 19 male rats. Some animals had been previously rechallenged subcutaneously, and had rejected the rechallenged cells. Some rats were rechallenged with concurrent subcutaneous a nd intracranial injectio ns. The details of these animals and survival times are shown in Table 5-2 and Figure 5-9. Compared with intracranial rechallenge only, anim als had prolonged surivival if they had been previously rechallenged subcutaneously (P = 0.0010) or if they received concurrent intracranial and subcutaneous rechallenge (P = 0.0054). Th ere was no difference between concurrently rechallenged rats and those with prior subcutaneous rechallenge (P = 0.2359). Histology and Immunohistochemistry Similar to the subcutaneous tumors, there was a large influx of inflammatory cells into and surrounding intracrania l RG-2 gliomas. Immunostaining for a number of different immune cell molecules revealed mixed populations of cells, some of which were again unexpected. CD11b immunostain ing, as expected, labeled microglia surrounding and invading tumor edges, and la rge oval macrophage-like cells within the tumor (Figures 5-9A and 5-10E). Both of these cell types showed robust MHC II staining (Figures 5-9D and 5-10C, D) and positive but less dramatic MHC I staining (Figures 5-9C and 5-10A, B). CD86 staining primarily labeled cells with an activated microglial morphology in peripheral tumor tissu e. These cells were relatively less common and generally occurred in groups (Figure 5-10F). Staining for CD6 was fairly sparse, a nd positive cells with lymphocyte morphology were rare. Both CD4 and CD 8 showed la rge numbers of positive cells with macrophage and activated microglial morphol ogies (Figures 5-9E, F and 5-11A-D). CD8 staining was generally more robust than CD4. Immunos taining for CD161, classically an NK cell marker, showed positive cells with both lymphocytic and microglial morphologies
97 (Figure 5-11E). Interestingl y, immunostaining for integrin E2, a dendritic cell marker, staining almost exclusively cells resembling lymphocytes. Similar to the situation in subcutaneous RG-2 gliomas, there was a fair amount of individual variability in imm une cell infiltration. LPS treated rats appeared to have markedly higher numbers of invading immune cells compared with some control animals, but not compared with others. Thus, a c onsistent correlation be tween treatment and immune cell response could not be made in our study. Discussion As opposed to the results seen with IT tr eatment of subcutaneous tumors (Chapter 3), IT treatment with LPS was not effective against intracranial RG-2 tumors in F344 rats. Part of the failure of therapy was related to increased toxicity related to administration of this immunostimulant into the central nervous system. The toxic effects were related to increas ed cerebral edema (Figure 5-7), swelling of the brain and in some cases, herniation of the intracranial co ntents through the foramen magnum, leading to death of the animal. Local swelling was also noted after treatment of subcutaneous tumors, but was much bette r tolerated in this re gion of the body. Although a dose-sparing effect of radiation treatment on IT LPS was noted in the subcutaneous tumors and was anticipated in the intracranial model, combination therapy did not alleviate the toxicity se en, at least with the dosing paradigm used here. It is possible that further adjustment of the dosing schedule, either with amounts of LPS or radiation administered, or with the relative timing of the treatments, may still show a beneficial treatment e ffect. However, alternative ap proaches to reduce toxicity and/or increase efficacy are like ly to be required if this ther apy is to be successful in the
98 intracranial compartment. Before discussing these approaches, it is useful to note a few caveats regarding this tu mor model system. Many of the glial tumor cell lin es available for study in th e rat are syng eneic to the F344 rat, including the F98, 9L and RG-2 ce ll lines. Of these, the RG-2 line is considered to be the least immunogenic, and therefore the mo st appropriate for immunotherapy studies (113, 193). One popular tumor cell line, C6, does not have an available syngeneic host, and thus is inappropria te for such studies. Initial evaluation of a newer cell line syngeneic to Lewis rats, CN S-1, showed that these tumors underwent spontaneous regression when implanted s ubcutaneously or wh en administered intracranially through the guide cannula system described above (data not shown). Fisher 344 rats are well-known to be less hardy than other rat strains, and this may have influenced the results seen in these studies . Supporting this contention, when similar doses of LPS are administered into Wistar rats (an outbred strain ) with intracranial tumors of comparable size, no deaths were noted, and the rats appeared completely healthy (data not shown). T hus, the RG-2 model system may be somewhat limited by the F344 rats sensitivity to adversity, which may not be the case in actual human patients. A number of strategies may be consider ed to improve the th erapeutic index of nonspecific immunotherapy, that is, to improve the efficacy while reducing toxicity. First, concurrent debulking of the tumor with surgery would remove a large part of the mass effect driving the increase in intracranial pressure and subsequent herniation. With a reduction in intracranial contents, the rema ining normal brain has more room to expand without the catastrophic effects of brain herniation. Such debulking is already part of the current standard of care fo r most patients with malignant gliomas (288, 328), and
99 immunostimulatory substances c ould be administered post-oper atively into the resection cavity, an approach currently used for adjuva nt chemotherapy in some patients (329). The concurrent administration of anti-ede ma medications is a second and possibly complimentary strategy to reduce intracr anial volume and prevent increases in intracranial pressure. Such drugs, notably glucocorticoids such as dexamethasone or prednisone, are commonly administered to glio ma patients for this purpose. A potential drawback of such a strategy is the general suppression of the immune response seen with the administration of glucocorticoids. Howe ver, anti-edema effects may be noted at doses below those considered to be immunos uppressive. Won et al. recently reported that dexamethasone can reduce the toxicity of LPS when given in a subcutaneous tumor model without reducing the efficacy of ther apy (163). In addition, other medications such as the 21-aminosteroids may have the potential to reduce cer ebral edema without suppressing the immune response (330, 331). A third strategy is to inve stigate the use of other i mmunostimulatory substances that may be less toxic than LPS when used in tracranially. Lipid A is considered to be responsible for the majority of the immunosti mulatory effects of LPS (148), but also appears to have reduced toxicity. When ev aluated in a subcutaneous murine model of glioma, Won et al. found that lipid A had sim ilar efficacy to LPS with fewer side effects (163). The administration of oligodexoynucle otides with CpG motifs (CpG-ODN) has been evaluated in a number of different tumor models, including gliomas (174, 332). When Carpentier et al. administered CpG-OD N directly into intracranial tumors in a CNS-1 rat model of glioma (174), 88% of rats showed long-term survival (greater than 90 days). This group also demonstrated s ynergy of CpG-ODN treat ment with radiation
100 therapy in the 9L model (304). This latter study also utilized F344 rats, suggesting that this paradigm is considerably less toxic than LPS. However, as alluded to above, the 9L glioma is naturally immunogenic, even in it s syngeneic host, and the efficacy noted may not be easily translated to human patients in the clinic. Finally, a combination of systemic a nd local immunotherapy may also be considered as a possible way to improve effi cacy. Part of the therapeutic failure of nonspecific immunotherapy in this model might be attributed to the rapid growth of tumors after intracranial implantation of RG-2 cells. Theoretically, the generation of an anti-tumor immune response, at least a T cel l response, requires an tigen capture at the tumor site, migration to secondary lymphoi d organs, activation and proliferation of effector T cells and then migration back to the site of the tumor. The time required for this response may exceed the short time required for the RG-2 tumors to grow to a lethal size. Further studies might evaluate tumo rs implanted with lower numbers of cells, giving the immune system more time to generate an effectiv e response. Alternatively, the generation of systemic immunity at an earlier date through subcutaneous vaccination with killed tumor cells and immunostimul ants may accelerate the process. The experiment evaluating subcutaneous tr eatment with LPS and irradiated tumor cells failed to show a survival benefit on ex isting intracranial tumors. If T cells are effectively generated in the periphery, however , they are likely to re quire restimulation at the site of action (i.e., intracranially) in or der to actively proliferate and generate a cytotoxic response. Treatment with nonspeci fic immunostimulants at both subcutaneous and intracranial locations might provide a me thod of achieving this end and generating an effective anti-tumor response.
101 Figure 5-1. An intracranial RG-2 g lioma treated with IT LPS and IFN(MR images). These images were obtained on Days 10 (A, B) and 17 (C, D) after tumor implantation. Panels A and C are T2-weighted images, while B and D are post-contrast T1-weighted images. Note the rapid growth of the tumor (arrowheads), leading to considerab le mass effect, de spite treatment.
102 Figure 5-2. Intratumoral LPS and IFNcombination treatment of intracranial RG-2 gliomas. There was no difference in survival between LPS/ IFN(4 g/100 ng) and saline treated rats (P = 0.5498). Day 0 is the day of tumor implantation and the arrows represent treatment dates. Figure 5-3. Intratumoral LPS treatment of intracranial RG-2 gliomas. There was no difference in survival between rats trea ted with LPS (5 or 10 g) and saline controls (P = 0.5780). Day 0 is the day of tumor implantation and the arrows represent treatment dates.
103 Figure 5-4. Dose-response experiment of intracranial RG-2 gliomas treated with radiation therapy. Male F344 rats with intracranial RG-2 gliomas were treated with single-dose radiation therapy ( 5, 10, 15 or 25 Gy) and compared with untreated controls. Only rats treate d with 25 Gy showed prolonged survival compared to controls (P = 0.0389). Da y 0 is the day of tumor implantation and the arrow represents the treatment date.
104 Figure 5-5. Intratumoral radia tion therapy and LPS combination treatment of intracranial RG-2 gliomas. Male F344 rats were tr eated with radiati on therapy (10 or 15 Gy), LPS (10 g IT) or combinations of radiation and LPS (10 Gy radiation + 10 g LPS or 15 Gy radiation + 10 g LPS IT) and compared to untreated controls. Day 0 is the day of tumor im plantation and the arrow represents the treatment date. A) Uncensored data, coun ting all deaths in su rvival analysis. Comparison between treatment groups is shown in Table 5-1. B) Rats dying as a result of treatment were censored in the survival analysis. Censor points are denoted by symbols corre sponding to treatment group.
105 Table 5-1. Combination singledose external beam radiatio n therapy and intratumoral lipopolysaccharide for intr acranial RG-2 gliomas: Treatment outcome aP values calculated using the log-rank test for comparison of survival curves generated using the Kaplan-Meier method (Figure 5-5) bLPS dose for all groups was 10 g Comparison vs. other groupsa Animals with Complete Regression (%) Treatment Group Median Survival (Days) A B C D E F 0 A Saline 15.5 0.04460.20720.03890.5826 0.8584 0 B LPSb 10.5 0.0446 0.01890.01890.2931 0.8482 0 C 10 Gy 18 0.20720.0189 0.37010.7223 0.4590 0 D 15 Gy 20 0.03890.01890.3701 0.9639 0.2082 0 E 10 Gy + LPS 15 0.58260.29310.72230.9639 0.3385 0 F 15 Gy + LPS 10 0.85840.84820.45900.20820.3385 0
106 Figure 5-6. Combination radia tion therapy and IT LPS treat ment of intracranial RG-2 gliomas. Survival curve showing male F344 rats with in tracranial RG-2 gliomas treated with radiation therapy ( 15 Gy) and IT saline, radiation therapy (15 Gy) and IT LPS (5 g IT) or radia tion therapy (15 Gy) and IT LPS (10 g IT). There is no difference in survival between treatment groups (P = 0.8607). Day 0 is the day of tumor implantati on and the arrows represent treatment dates.
107 Figure 5-7. Intracranial RG-2 gliomas treated with LPS and radiation versus radiation alone (MR images). These images were obtained 16 days after tumor implantation in rats treated with a sing le dose of radiation (15 Gy) and 5 g LPS 3X (C, D) or radiation and sali ne (A, B). Panels A and C are T2weighted images, while B and D are post-contrast T1-weighted images. Note the marked hyperintensity representing cer ebral edema in the rat treated with LPS (C). The guide cannula is visible as a linear signal void in all images.
108 Figure 5-8. Treatment of in tracranial RG-2 gliomas with subcutaneous LPS and irradiated tumor cells. Survival curve showing male F344 rats with intracranial RG-2 gliomas treated with subcutaneous LPS (2000 g), irradiated RG-2 cells (500,000 cells), vi able RG-2 cells (500,000 cells), LPS + irradiated RG-2 cells, LPS + viable RG-2 cells or saline. There is no difference in survival between treatm ent groups (P = 0.6112). Day 0 is the day of tumor implantation and the a rrows represent treatment dates.
109 Table 5-2. Intracranial rechal lenge experiment of rats with rejection of subcutaneous RG-2 tumors. Gender Subcutaneous IT treatment Time from 1st treatment to ICa rechallenge Previous subcutaneous rechallenge IC or IC + SQb Survival from rechallenge None, no tumor 195 No IC 15 2000 g LPS 3X/week 139 No IC Rejected IC tumor Saline 139 No IC Rejected IC tumor None, no tumor 195 No IC 13 1000 g LPS 3X/week 139 No IC 13 2000 g LPS 3X/week 139 No IC 15 1000 g ZymA 3X/week 326 No IC 13 2000 g ZymA 3X/week 481 Yes IC 18 5000 g ZymA 3X/week 326 No IC 12 250 g LPS 3X 172 No IC + SQ 34 500 g LPS 3X 257 Yes IC 15 1000 g 3X 227 No IC 15 1000 g 3X 255 Yes IC Rejected IC tumor 5000 g 3X 232 No IC 12 5000 g 3X 503 Yes IC + SQ 22 2000 g ZymA 3X 152 No IC + SQ 14 2000 g ZymA 3X 154 No IC + SQ 34 250 g LPS + 10 g IFN3X 172 No IC + SQ Rejected IC tumor 50 Gy 351 Yes IC 18
110 Table 5-2. Continued. Gender Subcutaneous IT treatment Time from 1st treatment to ICa rechallenge Previous subcutaneous rechallenge IC or IC + SQb Survival from rechallenge 50 Gy 351 Yes IC 18 50 Gy 351 Yes IC 24 10 Gy + 100 g LPS 296 No IC 14 aIntracranial bSubcutaneous
111 Figure 5-9. Intracranial rechallenge experiment in rats rejecting previous subcutaneous RG-2 gliomas. Survival curve showi ng male F344 rats that developed and then subsequently rejected subcutaneous RG-2 gliomas after treatment. Rats are divided into 3 groups: 1) Intracrani al rechallenge only (IC only), 2) Prior subcutaneous rechallenge (Prior SQ), 3) Concurrent intracranial and subcutaneous rechallenge (IC + SQ). Group 1 had significantly shorter survival than groups 2 (P = 0.0010) and 3 (P = 0.0054). There was no difference between groups 2 and 3 (P = 0.2359).
112 Figure 5-10. Immunohistochemical evaluation of an intracranial RG-2 glioma (low power). Sections are from a rat treat ed with 15 Gy + IT LPS (10 g) 3X. Tumors are stained with the peroxi dase method and counterstained with hematoxylin. The dotted line delineates the tumor edge, with the tumor side designated with the letter T. Im munostained molecules with clone designations in paretheses are as follow s: A) CD11b (OX-42) B) Integrin E2 (OX-62) C) MHC I (OX-18) D) MHC II (OX-6) E) CD4 (W3/25) F) CD8 (OX-8).
113 Figure 5-11. Immunohistochemical evaluation of an intracranial RG-2 glioma (higher power). Sections are from a rat treat ed with 15 Gy + IT LPS (10 g) 3X. Tumors are stained with the peroxi dase method and counterstained with hematoxylin. A) MHC I (OX-18) pos itive cells with rounded macrophage morphology. B) MHC I positive cells with activated microglial morphology (arrows). C) MHC II (OX-6) posit ive cells with rounded macrophage morphology. D) MHC II positive cells w ith activated microglial morphology (arrows). E) CD11b (OX-42) positive cells with macrophage and microglia (arrow) morphologies. F) CD86 (24F) pos itive cells with activated microglial morphology (arrow).
114 Figure 5-12. Immunohistochemical evaluation of an intracranial RG-2 glioma (higher power). Sections are from a rat treat ed with 15 Gy + IT LPS (10 g) 3X. Tumors are stained with the peroxi dase method and counterstained with hematoxylin. A) CD4 (W3/25) posi tive cells with rounded macrophage morphology. B) CD4 positive cells w ith activated microglial morphology (arrows). C) CD8 (OX-8) positive cells with rounded macrophage morphology. D) CD8 positive cells w ith activated microglial morphology (arrows). E) CD161 (3.2.3) positive cel ls with lymphocytic (arrowhead) and microglial (arrow) morphologies. F) Integrin E2 (OX-62) positive cells with lymphocytic morphologies (arrowheads).
115 CHAPTER 6 CONCLUSIONS Both of the TLR ligands evaluated in our study, LPS and ZymA, showed antitumor effects when injected IT into subcutaneous RG-2 gliomas. These effects were dose-dependent, and multiple doses had greater anti-tumor effects than single doses. In contrast, GM-CSF did not show any anti-t umor effects in our study. Although it is possible that this cytokine may be effective with a different dosing schedule, it is likely that this compound alone is unable to provide the necessary signals to immune cells to mount an attack in this w eakly immunogenic tumor model. Previous reports of efficacy in other experimental tumor models (223) and human clinical trials (226-228) may be related to the inherent immunogenicity of these tumors, which is not a feature of malignant gliomas. A synergistic effect on subcutaneous tumors was also noted when LPS was combined with single-dose ra diation therapy. This is an important observation, as radiation therapy is part of th e standard therapy for malignant gliomas, and future trials in human patients will have to occur in conjunc tion with this therapeutic modality. The synergistic effect may allow dose-sparing of both individual treat ments, potentially reducing neurologic side effects. The results of NSI in athymic nude rats a nd the resistance of successfully treated animals to tumor rechallenge suggest the involve ment of T cells in the anti-tumor effects of LPS and ZymA. Although histological evaluation of both subcutaneous and intracranial tumors showed a robust infiltration of immune cells, a correlation could not
116 be established between this infiltration and treatment success. Tumor rejection after NSI with TLR ligands appears to involve a numbe r of different immune cell populations and there is probably cooperation between these ce ll types at multiple levels. Examination of tumors at time points earlier and later than t hose shown here or at a point when they are clearly regressing may provide more information in terms of cells potentially important in contributing to anti-tumor effects. The use of flow cytometry allows cells to be labeled with multiple markers, which might aid in sorting out the complex phenotypes of these invading cells. Other approach es that might be beneficial in identifying cells contributing to tumor rejection include functional analys is of invading immune cells, isolation and adoptive transfer of specific cell populations to untreated tumor-bearing hosts and the use of animals genetically defi cient in certain cell types or immune molecules. Treatment of intracranial tumors with IT LPS with or without radiation therapy was not successful in extending survival. This was due in large part to the toxic effects of this protocol, characterized by cerebral edema a nd brain swelling. Per itumoral swelling was also noted after treatmen t of subcutaneous tumors, but is obv iously better tolerated in this location. Although these toxic effects are cl early a drawback to this therapeutic paradigm, there are a number of reasons why such a course might still be considered. Firstly, F344 rats are well-recognized as being a relatively fragile strain. When outbred Wistar rats receive a similar intracr anial LPS dose, the adverse effects are much less severe, if noted at all. Thus, toxic eff ects of this approach ma y not be as great in other strains or species. Secondly, much of the morbidity and mortality associated with this treatment was related to brain swelling a nd shifts of brain tissue, or herniation. As the contents of the intracra nial vault are enclosed by a rigid skull, swelling beyond a
117 certain tolerated point leads to catastrophic effects. The current standard of care for human patients includes surgical debulking of the tumor, leaving a resection cavity. This cavity is potentially useful as a safeguard, as it provides space for the expansion of swollen brain tissue. Although LPS was used in our study, lipid A has been shown to mediate most of the biological and immunos timulatory effects of this compound. A number of synthetic lipid A an alogues are currently available, which appear to retain the immunostimulatory effects of LPS with consid erably less toxicity. Several of these compounds have shown efficacy in the therapy of a number of different cancers. Finally, it may be possible to administer LPS, lipid A, ZymA or similar TLR agonists in conjunction with an agent that reduces cer ebral edema and brain swelling, and improves the therapeutic index of these immunostimulan ts. Medications such as glucocorticoids have these effects and are frequently admini stered to patients with brain tumors. Although these drugs can have immunosuppressive effects, preliminary studies by others have shown that they may reduce the toxic ity of TLR ligands without impairing the immunostimulatory effects (163). Although activated T cells can enter the CNS, this traffi c is still very tightly regulated, and these cells enter the brain in much smaller numbers when compared with other tissues (333). In the rats rechallenged with intrac ranial RG-2 injections, survival times were longer if rats had either prior or concurrent s ubcutaneous RG-2 rechallenge (Figure 5-8). These observations and othe rs support an approach to brain tumor immunotherapy employing combinations of system ic and local therapy. In this scenario, a systemic immune response to tumor antigen(s) is induced, resulting in the production of activated T cells. Some of these cells tra ffic to the intended site of action, where
118 microglia, activated by local immunostimulan ts, restimulate T cells, leading to clonal expansion of these cells in the brain. The T cells might reciprocally stimulate microglia, leading to a coordinated anti-tumor response. Nonspecific immunotherapy remains a promis ing therapy for malignant gliomas. Compared with other immunotherapeutic protocol s, it is technically easy, facilitating the transfer to clinical patien ts without requirements of gr owing tumor cells in culture, transfection techniques, ex vivo immune cell expansion or other demanding laboratory procedures. As demonstrated in the intracra nial allogeneic model, the immune system has the potential to completely eliminate large intracranial tumors which are causing signs of neurologic dysfunction (Chapter 2). However, the rationale of targeting multiple tumor antigens with NSI may be both advantageo us and detrimental. The advantage lies in avoiding the escape of clonal tumor popula tions that may develop as antigen loss variants in a genetically unsta ble group of cells. The disa dvantage is the potential for shared antigens between tumor and norma l brain tissues, which might lead to autoimmunity (334). Thus, future studies in this area must be approached with both cautious optimism and wary vigilance.
119 LIST OF REFERENCES 1. Akman, F., Cooper, R. A., Sen, M., Tanriver, Y., and Kentli, S. Validation of the Medical Research Council and a newly de veloped prognostic index in patients with malignant glioma: how useful are prognostic indices in routine clinical practice? J Neurooncol, 59: 39-47, 2002. 2. Basso, U., Ermani, M., Vastola, F., and Brandes, A. A. Non-cytotoxic therapies for malignant gliomas. J Neurooncol, 58: 57-69, 2002. 3. Nieder, C., Grosu, A. L., and Molls, M. A comparison of treatment results for recurrent malignant gliomas. Cancer Treat Rev, 26: 397-409, 2000. 4. Quah, B. J. and O'Neill, H. C. Maturation of function in dendritic cells for tolerance and immunity. J Cell Mol Med, 9: 643-654, 2005. 5. Banchereau, J. and Steinma n, R. M. Dendritic cells a nd the control of immunity. Nature, 392: 245-252, 1998. 6. Rodriguez, A., Regnault, A., Kleijm eer, M., Ricciardi-Castagnoli, P., and Amigorena, S. Selective transport of inte rnalized antigens to the cytosol for MHC class I presentation in dendritic cells. Nat Cell Biol, 1: 362-368, 1999. 7. Kurts, C. Cross-presentation: inducing CD8 T cell immunity and tolerance. J Mol Med, 78: 326-332, 2000. 8. Heath, W. R. and Carbone, F. R. Crosspresentation, dendritic cells, tolerance and immunity. Annu Rev Immunol, 19: 47-64, 2001. 9. Albert, M. L., Sauter, B., and Bhardwaj , N. Dendritic cells acquire antigen from apoptotic cells and induce cla ss I-restricted CTLs. Nature, 392: 86-89, 1998. 10. Norbury, C. C., Chambers, B. J., Presco tt, A. R., Ljunggren, H. G., and Watts, C. Constitutive macropinocytosis allows TAP-dependent major histocompatibility complex class I presentation of exog enous soluble antigen by bone marrowderived dendritic cells. Eur J Immunol, 27: 280-288, 1997. 11. Shen, Z., Reznikoff, G., Dranoff, G., a nd Rock, K. L. Cloned dendritic cells can present exogenous antigens on both MH C class I and class II molecules. J Immunol, 158: 2723-2730, 1997.
120 12. Mitchell, D. A., Nair, S. K., and Gil boa, E. Dendritic cell/macrophage precursors capture exogenous antigen for MHC class I presentation by dendritic cells. Eur J Immunol, 28: 1923-1933, 1998. 13. Svensson, M. and Wick, M. J. Classical MHC class I peptide presentation of a bacterial fusion protein by bone marrow-de rived dendritic cells. Eur J Immunol, 29: 180-188, 1999. 14. Kovacsovics-Bankowski, M., Clark, K., Ben acerraf, B., and Rock, K. L. Efficient major histocompatibility complex class I presentation of exogenous antigen upon phagocytosis by macrophages. Proc Natl Acad Sci U S A, 90: 4942-4946, 1993. 15. Norbury, C. C., Hewlett, L. J., Prescott, A. R., Shastri, N., and Watts, C. Class I MHC presentation of exogenous soluble antigen via macropinocytosis in bone marrow macrophages. Immunity, 3: 783-791, 1995. 16. Arina, A., Tirapu, I., Alfaro, C., Rodri guez-Calvillo, M., Mazzolini, G., Inoges, S., Lopez, A., Feijoo, E., Bendandi, M., and Melero, I. Clinical implications of antigen transfer mechanisms from maligna nt to dendritic cells. exploiting crosspriming. Exp Hematol, 30: 1355-1364, 2002. 17. Huang, A. Y., Golumbek, P., Ahmadzad eh, M., Jaffee, E., Pardoll, D., and Levitsky, H. Role of bone marrow-derive d cells in presenting MHC class Irestricted tumor antigens. Science, 264: 961-965, 1994. 18. Robinson, B. W., Lake, R. A., Nelson, D. J., Scott, B. A., and Marzo, A. L. Crosspresentation of tumour antigens: eval uation of threshold, duration, distribution and regulation. Immunol Cell Biol, 77: 552-558, 1999. 19. Selenko, N., Majdic, O., Jager, U., Sillabe r, C., Stockl, J., and Knapp, W. Crosspriming of cytotoxic T cells promoted by apoptosis-inducing tumor cell reactive antibodies? J Clin Immunol, 22: 124-130, 2002. 20. Djeu, J. Y., Jiang, K., and Wei, S. A view to a kill: signals tr iggering cytotoxicity. Clin Cancer Res, 8: 636-640, 2002. 21. Constant, S. L. and Bottomly, K. Induction of Th1 and Th2 CD4+ T cell responses: the alternative a pproaches. Annu Rev Immunol, 15: 297-322, 1997. 22. Abbas, A. K., Murphy, K. M., and Sher , A. Functional diversity of helper T lymphocytes. Nature, 383: 787-793, 1996. 23. Mosmann, T. R. and Sad, S. The expandi ng universe of T-cell subsets: Th1, Th2 and more. Immunol Today, 17: 138-146, 1996.
121 24. Bauer, S., Groh, V., Wu, J., Steinle, A., Ph illips, J. H., Lanier, L. L., and Spies, T. Activation of NK cells and T cells by NKG2D, a receptor for stress-inducible MICA. Science, 285: 727-729, 1999. 25. Screpanti, V., Wallin, R. P., Grandien, A., and Ljunggren, H. G. Impact of FASLinduced apoptosis in the elimination of tumor cells by NK cells. Mol Immunol, 42: 495-499, 2005. 26. Munz, C., Steinman, R. M., and Fujii, S. Dendritic cell maturation by innate lymphocytes: coordinated stimulation of innate and adaptive immunity. J Exp Med, 202: 203-207, 2005. 27. Lamb, L. S., Jr. and Lopez, R. D. gammadelta T cells: a new frontier for immunotherapy? Biol Blood Marrow Transplant, 11: 161-168, 2005. 28. Freedman, M. S., D'Souza, S., and Antel, J. P. gamma delta T-cell-human glial cell interactions. I. In vitro induction of gammadelta T-cell expansion by human glial cells. J Neuroimmunol, 74: 135-142, 1997. 29. Hori, S., Nomura, T., and Sakaguchi, S. Control of regulatory T cell development by the transcription factor Foxp3. Science, 299: 1057-1061, 2003. 30. Sakaguchi, S., Sakaguchi, N., Asano, M., Itoh, M., and Toda, M. Immunologic self-tolerance maintained by activated T cells expressing IL-2 receptor alphachains (CD25). Breakdown of a singl e mechanism of self-tolerance causes various autoimmune diseases. J Immunol, 155: 1151-1164, 1995. 31. Zou, W. Regulatory T cells, tumour immunity and immunotherapy. Nat Rev Immunol, 6: 295-307, 2006. 32. Shevach, E. M. CD4+ CD25+ suppressor T cells: more questi ons than answers. Nat Rev Immunol, 2: 389-400, 2002. 33. Liyanage, U. K., Moore, T. T., Joo, H. G., Tanaka, Y., Herrmann, V., Doherty, G., Drebin, J. A., Strasberg, S. M., Eberlein, T. J., Goedegebuure, P. S., and Linehan, D. C. Prevalence of regulatory T cells is increased in peripheral blood and tumor microenvironment of patients w ith pancreas or breast adenocarcinoma. J Immunol, 169: 2756-2761, 2002. 34. Zou, W. Immunosuppressive networks in the tumour envi ronment and their therapeutic relevance. Nat Rev Cancer, 5: 263-274, 2005. 35. Andaloussi, A. E. and Lesniak, M. S. An increase in CD4+CD25+FOXP3+ regulatory T cells in tumor-infiltratin g lymphocytes of human glioblastoma multiforme. Neuro-oncol, 8: 234-243, 2006.
122 36. El Andaloussi, A., Han, Y., and Lesniak, M. S. Prolongation of survival following depletion of CD4+CD25+ regulatory T ce lls in mice with experimental brain tumors. J Neurosurg, 105: 430-437, 2006. 37. Curiel, T. J., Coukos, G., Zou, L., Alvarez, X., Cheng, P., Mottram, P., EvdemonHogan, M., Conejo-Garcia, J. R., Zha ng, L., Burow, M., Zhu, Y., Wei, S., Kryczek, I., Daniel, B., Gordon, A., Myer s, L., Lackner, A., Disis, M. L., Knutson, K. L., Chen, L., and Zou, W. Sp ecific recruitment of regulatory T cells in ovarian carcinoma fosters immune priv ilege and predicts reduced survival. Nat Med, 10: 942-949, 2004. 38. Akira, S. and Takeda, K. Toll-lik e receptor signalling. Nat Rev Immunol, 4: 499511, 2004. 39. Ulevitch, R. J. Therapeutics targe ting the innate immune system. Nat Rev Immunol, 4: 512-520, 2004. 40. Nguyen, M. D., Julien, J. P., and Rivest, S. Innate immunity: the missing link in neuroprotection and neurodegene ration? Nat Rev Neurosci, 3: 216-227, 2002. 41. Takeda, K. and Akira, S. Roles of Toll-like receptors in innate immune responses. Genes Cells, 6: 733-742, 2001. 42. Hacker, H., Vabulas, R. M., Takeuchi, O ., Hoshino, K., Akira, S., and Wagner, H. Immune cell activation by bacterial CpG-DNA through myeloid differentiation marker 88 and tumor necrosis factor r eceptor-associated factor (TRAF)6. J Exp Med, 192: 595-600, 2000. 43. Kawai, T., Adachi, O., Ogawa, T., Takeda, K., and Akira, S. Unresponsiveness of MyD88-deficient mice to endotoxin. Immunity, 11: 115-122, 1999. 44. Iwasaki, A. and Medzhitov, R. Toll-like receptor cont rol of the adaptive immune responses. Nat Immunol, 5: 987-995, 2004. 45. Greene, H. S. The transplantation of tumo rs to the brains of heterologous species. Cancer Res, 11: 529-534, 1951. 46. Murphy, J. B., Sturm, E. Conditions determ ining the transplantability of tissues in the brain. J Exp Med, 38: 183-197, 1923. 47. Shirai, Y. Transplantation of rat sarcom a in adult heterogeneous animals. Japan Med World, 1: 14-15, 1921. 48. McMenamin, P. G. Distribution and phe notype of dendritic cells and resident tissue macrophages in the dura mater, lept omeninges, and choroid plexus of the
123 rat brain as demonstrated in whol emount preparations. J Comp Neurol, 405: 553562, 1999. 49. Medawar, P. B. Immunity to homologous grafted skin. III. The fate of skin homografts transplanted to the brain, to subcutaneous tissue and to the anterior chamber of the eye. Br J Exp Pathol, 29: 58-69, 1948. 50. Cserr, H. F. and Knopf, P. M. Cervical lymphatics, the blood-brain barrier and the immunoreactivity of the brain: a new view. Immunol Today, 13: 507-512, 1992. 51. de Vos, A. F., van Meurs, M., Brok, H. P., Boven, L. A., Hintzen, R. Q., van der Valk, P., Ravid, R., Rensing, S., Boon, L., t Hart, B. A., and Laman, J. D. Transfer of central nervous system auto antigens and presentation in secondary lymphoid organs. J Immunol, 169: 5415-5423, 2002. 52. Okamoto, Y., Yamashita, J., Hasegawa, M., Fujisawa, H., Yamashima, T., Hashimoto, T., Nonomura, A., Matsumoto, Y., and Kida, S. Cervical lymph nodes play the role of regional lymph nodes in brain tumour immunity in rats. Neuropathol Appl Neurobiol, 25: 113-122, 1999. 53. Hickey, W. F., Hsu, B. L., and Kimura, H. T-lymphocyte entry into the central nervous system. J Neurosci Res, 28: 254-260, 1991. 54. Graeber, M. B. and Streit, W. J. Micr oglia: immune network in the CNS. Brain Pathol, 1: 2-5, 1990. 55. Theele, D. P. and Streit, W. J. A chronicle of microglial ontogeny. Glia, 7: 5-8, 1993. 56. Santambrogio, L., Belyanskaya, S. L., Fisc her, F. R., Cipriani, B., Brosnan, C. F., Ricciardi-Castagnoli, P., Stern, L. J., Strominger, J. L., and Riese, R. Developmental plasticity of CNS microglia. Proc Natl Acad Sci U S A, 98: 62956300, 2001. 57. Graeber, M. B., Streit, W. J., and Kr eutzberg, G. W. Identity of ED2-positive perivascular cells in ra t brain. J Neurosci Res, 22: 103-106, 1989. 58. Sedgwick, J. D., Schwender, S., Imrich, H., Dorries, R., Butcher, G. W., and ter Meulen, V. Isolation and dire ct characterization of resi dent microglial cells from the normal and inflamed central nervous system. Proc Natl Acad Sci U S A, 88: 7438-7442, 1991. 59. Imamura, K., Ito, M., Suzumura, A., Asai , J., and Takahashi, A. Generation and characterization of monocl onal antibodies against rat microglia and ontogenic distribution of positive cells. Lab Invest, 63: 853-861, 1990.
124 60. Graeber, M. B., Streit, W. J., Buringer, D., Sparks, D. L., and Kreutzberg, G. W. Ultrastructural location of major histocompatibility complex (MHC) class II positive perivascular cells in histologically normal human brain. J Neuropathol Exp Neurol, 51: 303-311, 1992. 61. Hickey, W. F. and Kimura, H. Perivasc ular microglial cells of the CNS are bone marrow-derived and present antigen in vivo. Science, 239: 290-292, 1988. 62. Hickey, W. F., Vass, K., and Lassmann, H. Bone marrow-derived elements in the central nervous system: an immunohistochemical and ultr astructural survey of rat chimeras. J Neuropathol Exp Neurol, 51: 246-256, 1992. 63. Graeber, M. B., Tetzlaff, W., Streit, W. J., and Kreutzberg, G. W. Microglial cells but not astrocytes undergo mitosis follow ing rat facial nerve axotomy. Neurosci Lett, 85: 317-321, 1988. 64. Giulian, D., Baker, T. J., Shih, L. C., and Lachman, L. B. Interleukin 1 of the central nervous system is produced by ameboid microglia. J Exp Med, 164: 594604, 1986. 65. Sawada, M., Kondo, N., Suzumura, A., and Marunouchi, T. Production of tumor necrosis factor-alpha by microglia and astrocytes in culture. Brain Res, 491: 394397, 1989. 66. Aloisi, F., Penna, G., Cerase, J., Menend ez Iglesias, B., and Adorini, L. IL-12 production by central nervous system micr oglia is inhibited by astrocytes. J Immunol, 159: 1604-1612, 1997. 67. Becher, B., Dodelet, V., Fedorowicz, V., and Antel, J. P. Soluble tumor necrosis factor receptor inhibits interleukin 12 production by stimulated human adult microglial cells in vitro. J Clin Invest, 98: 1539-1543, 1996. 68. Aloisi, F., Ria, F., Penna, G., and Adorin i, L. Microglia are more efficient than astrocytes in antigen processing and in Th1 but not Th2 cell activation. J Immunol, 160: 4671-4680, 1998. 69. De Simone, R., Giampaolo, A., Giometto, B., Gallo, P., Levi, G ., Peschle, C., and Aloisi, F. The costimulatory molecule B7 is expressed on human microglia in culture and in multiple sclerosis acute lesions. J Neuropathol Exp Neurol, 54: 175-187, 1995. 70. Tran, C. T., Wolz, P., Egensperger, R., Kosel, S., Imai, Y., Bise, K., Kohsaka, S., Mehraein, P., and Graeber, M. B. Diffe rential expression of MHC class II molecules by microglia and neoplastic astroglia: relevance for the escape of astrocytoma cells from immune survei llance. Neuropathol Appl Neurobiol, 24: 293-301, 1998.
125 71. Ford, A. L., Goodsall, A. L., Hickey, W. F., and Sedgwick, J. D. Normal adult ramified microglia separated from othe r central nervous system macrophages by flow cytometric sorting. Phenotypic differen ces defined and direct ex vivo antigen presentation to myelin basic protein-r eactive CD4+ T cells compared. J Immunol, 154: 4309-4321, 1995. 72. Frei, K., Siepl, C., Groscurth, P., Bodm er, S., Schwerdel, C., and Fontana, A. Antigen presentation and tumor cytot oxicity by interfer on-gamma-treated microglial cells. Eur J Immunol, 17: 1271-1278, 1987. 73. Cash, E. and Rott, O. Microglial cells qualify as the stimulators of unprimed CD4+ and CD8+ T lymphocytes in th e central nervous system. Clin Exp Immunol, 98: 313-318, 1994. 74. Williams, K., Jr., Ulvestad, E., Cragg, L., Bl ain, M., and Antel, J. P. Induction of primary T cell responses by human glial cells. J Neurosci Res, 36: 382-390, 1993. 75. Aloisi, F. Immune function of microglia. Glia, 36: 165-179, 2001. 76. Gehrmann, J., Matsumoto, Y., and Kreu tzberg, G. W. Microglia: intrinsic immuneffector cell of the br ain. Brain Res Brain Res Rev, 20: 269-287, 1995. 77. Matsumoto, Y., Ohmori, K., and Fujiwara, M. Immune regulation by brain cells in the central nervous system: microglia but not astrocytes present myelin basic protein to encephalitogenic T cells under in vivo-mimicking conditions. Immunology, 76: 209-216, 1992. 78. Streit, W. J., Graeber, M. B., and Kreu tzberg, G. W. Functional plasticity of microglia: a review. Glia, 1: 301-307, 1988. 79. Becher, B., Prat, A., and Antel, J. P. Brain-immune connection: immunoregulatory properties of CN S-resident cells. Glia, 29: 293-304, 2000. 80. Aloisi, F., Ria, F., Columba-Cabezas, S., Hess, H., Penna, G., and Adorini, L. Relative efficiency of microglia, astrocyt es, dendritic cells and B cells in naive CD4+ T cell priming and Th1/Th2 cell restimulation. Eur J Immunol, 29: 27052714, 1999. 81. Flugel, A., Labeur, M. S., Grasbon-Frodl, E. M., Kreutzberg, G. W., and Graeber, M. B. Microglia only weakly present g lioma antigen to cytotoxic T cells. Int J Dev Neurosci, 17: 547-556, 1999. 82. Burnet, F. M. Cancer a bi ological approach. Br Med J, 1: 841-847, 1957.
126 83. Burnet, F. M. The concept of immunol ogical surveillance. Prog Exp Tumor Res, 13: 1-27, 1970. 84. Thomas, L. In: H. S. Lawrence (ed.), Cellular and Humoral Aspects of the Hypersensitive States, pp. 529-532. New York: Hoeber-Harper, 1959. 85. Dunn, G. P., Bruce, A. T., Ikeda, H., Old, L. J., and Schreiber, R. D. Cancer immunoediting: from immunosurveilla nce to tumor escape. Nat Immunol, 3: 991998, 2002. 86. Kielian, T., van Rooijen, N., and Hick ey, W. F. MCP-1 expression in CNS-1 astrocytoma cells: implications for macrophage infiltration into tumors in vivo. J Neurooncol, 56: 1-12, 2002. 87. Apuzzo, M. L. and Mitchell, M. S. Immunological aspects of intrinsic glial tumors. J Neurosurg, 55: 1-18, 1981. 88. Fossati, G., Ricevuti, G., Edwards, S. W., Walker, C., Dalton, A., and Rossi, M. L. Neutrophil infiltration into human gliomas. Acta Neuropathol (Berl), 98: 349354, 1999. 89. Badie, B. and Schartner, J. Role of microglia in glioma biology. Microsc Res Tech, 54: 106-113., 2001. 90. Graeber, M. B., Scheithauer, B. W., and Kreutzberg, G. W. Microglia in brain tumors. Glia, 40: 252-259., 2002. 91. Hitchcock, E. R. and Morris, C. S. Mononuclear cell infiltration in central portions of human astrocytomas. J Neurosurg, 68: 432-437, 1988. 92. Morioka, T., Baba, T., Black, K. L., and Streit, W. J. Imm unophenotypic analysis of infiltrating leukocytes and microglia in an experimental rat glioma. Acta Neuropathol, 83: 590-597, 1992. 93. Badie, B. and Schartner, J. M. Flow cytometric charac terization of tumorassociated macrophages in experi mental gliomas. Neurosurgery, 46: 957-961; discussion 961-952, 2000. 94. Kulprathipanja, N. V. and Kruse, C. A. Microglia phagocytose alloreactive CTLdamaged 9L gliosarcoma cells. J Neuroimmunol, 153: 76-82, 2004. 95. Rossi, M. L., Hughes, J. T., Esiri, M. M., Coakham, H. B., and Brownell, D. B. Immunohistological study of mononuclear cell infiltrate in malignant gliomas. Acta Neuropathol (Berl), 74: 269-277, 1987.
127 96. Morimura, T., Neuchrist, C., Kitz, K., Budka, H., Scheiner, O., Kraft, D., and Lassmann, H. Monocyte subpopulations in hum an gliomas: expression of Fc and complement receptors and correlation with tumor proliferation. Acta Neuropathol (Berl), 80: 287-294, 1990. 97. Morris, C. S. and Esiri, M. M. I mmunocytochemical study of macrophages and microglial cells and extracellular matrix components in human CNS disease. 1. Gliomas. J Neurol Sci, 101: 47-58, 1991. 98. Roggendorf, W., Strupp, S., and Paulus, W. Distribution and characterization of microglia/macrophages in human brain tumors. Acta Neuropathol (Berl), 92: 288293, 1996. 99. Wang, X. C., Kochi, N., Tani, E., Kaba, K., Matsumoto, T., and Shindo, H. Lectin histochemistry of human gliomas. Acta Neuropathol (Berl), 79: 176-182, 1989. 100. Klein, R. and Roggendorf, W. Increase d microglia proliferation separates pilocytic astrocytomas from diffuse as trocytomas: a double labeling study. Acta Neuropathol (Berl), 101: 245-248, 2001. 101. Leung, S. Y., Wong, M. P., Chung, L. P., Ch an, A. S., and Yuen, S. T. Monocyte chemoattractant protein-1 expression a nd macrophage infiltration in gliomas. Acta Neuropathol (Berl), 93: 518-527, 1997. 102. Stan, A. C., Walter, G. F., Welte, K ., and Pietsch, T. I mmunolocalization of granulocyte-colony-stimulating fact or in human glial and primitive neuroectodermal tumors. Int J Cancer, 57: 306-312, 1994. 103. Badie, B., Schartner, J., Klaver, J., and Vorpahl, J. In vitro modulation of microglia motility by glioma cells is mediat ed by hepatocyte grow th factor/scatter factor. Neurosurgery, 44: 1077-1082; discussion 1082-1073, 1999. 104. Huettner, C., Czub, S., Kerkau, S., Rogge ndorf, W., and Tonn, J. C. Interleukin 10 is expressed in human gliomas in vivo and increases glioma cell proliferation and motility in vitro. Anticancer Res, 17: 3217-3224, 1997. 105. Wagner, S., Czub, S., Greif, M., Vince, G. H., Suss, N., Kerkau, S., Rieckmann, P., Roggendorf, W., Roosen, K., and Tonn, J. C. Microglial/macrophage expression of interleukin 10 in human glioblastomas. Int J Cancer, 82: 12-16, 1999. 106. Constam, D. B., Philipp, J., Malipiero, U. V., ten Dijke, P., Schachner, M., and Fontana, A. Differential expression of tr ansforming growth f actor-beta 1, -beta 2, and -beta 3 by glioblastoma cells, as trocytes, and microglia. J Immunol, 148: 1404-1410, 1992.
128 107. Kiefer, R., Supler, M. L., Toyka, K. V., and Streit, W. J. In situ detection of transforming growth factor-beta mRNA in experimental rat glioma and reactive glial cells. Neurosci Lett, 166: 161-164, 1994. 108. Platten, M., Wick, W., and Weller, M. Malignant glioma biology: role for TGFbeta in growth, motility, angiogenesis, and immune escape. Microsc Res Tech, 52: 401-410, 2001. 109. Badie, B., Schartner, J., Prabakaran, S., Paul, J., and Vorpahl, J. Expression of Fas ligand by microglia: possible role in g lioma immune evasion. J Neuroimmunol, 120: 19-24., 2001. 110. Brooks, W. H., Markesbery, W. R., Gupta, G. D., and Roszman, T. L. Relationship of lymphocyte invasion and survival of brain tumor patients. Ann Neurol, 4: 219-224, 1978. 111. Di Lorenzo, N., Palma, L., and Nicole , S. Lymphocytic in filtration in longsurvival glioblastomas: possible host's resistance. Acta Neurochir (Wien), 39: 2733, 1977. 112. Palma, L., Di Lorenzo, N., and Guidetti, B. Lymphocytic infiltrates in primary glioblastomas and recidivous gliomas. Inci dence, fate, and relevance to prognosis in 228 operated cases. J Neurosurg, 49: 854-861, 1978. 113. Tzeng, J. J., Barth, R. F., Orosz, C. G., and James, S. M. Phenotype and functional activity of tumo r-infiltrating lymphocytes isolated from immunogenic and nonimmunogenic rat brain tumors. Cancer Res, 51: 2373-2378, 1991. 114. Burger, P. C., Vogel, F. S., Green, S. B., and Strike, T. A. Glioblastoma multiforme and anaplastic astrocytom a. Pathologic cr iteria and prognostic implications. Cancer, 56: 1106-1111, 1985. 115. Badie, B. and Schartner, J. M. Flow cytometric charac terization of tumorassociated macrophages in experi mental gliomas. Neurosurgery, 46: 957-961; discussion 961-952., 2000. 116. Parney, I. F., Hao, C., and Pe truk, K. C. Glioma immunology and immunotherapy. Neurosurgery, 46: 778-791; discussion 791-772, 2000. 117. Huettner, C., Paulus, W., and Roggendorf, W. Messenger RNA expression of the immunosuppressive cytokine IL-10 in human gliomas. Am J Pathol, 146: 317322, 1995. 118. Merlo, A., Juretic, A., Zuber, M., Filgue ira, L., Luscher, U., Caetano, V., Ulrich, J., Gratzl, O., Heberer, M., and Spagnoli, G. C. Cytokine gene expression in
129 primary brain tumours, metastases and meningiomas suggests specific transcription patterns. Eur J Cancer, 29A: 2118-2125, 1993. 119. Schneider, J., Hofman, F. M., Apuzzo, M. L., and Hinton, D. R. Cytokines and immunoregulatory molecules in mali gnant glial neoplasms. J Neurosurg, 77: 265273, 1992. 120. Kuppner, M. C., Hamou, M. F., Bodmer, S., Fontana, A., and de Tribolet, N. The glioblastoma-derived T-cell suppressor factor/transformi ng growth factor beta 2 inhibits the generation of lymphokine-activated killer (LAK) cells. Int J Cancer, 42: 562-567, 1988. 121. Kuppner, M. C., Sawamura, Y., Hamou, M. F., and de Tribolet, N. Influence of PGE2and cAMP-modulating agents on hu man glioblastoma cell killing by interleukin-2-activated ly mphocytes. J Neurosurg, 72: 619-625, 1990. 122. O'Keefe, G. M., Nguyen, V. T., and Benveni ste, E. N. Class II transactivator and class II MHC gene expression in micr oglia: modulation by the cytokines TGFbeta, IL-4, IL-13 and IL-10. Eur J Immunol, 29: 1275-1285, 1999. 123. Suzumura, A., Sawada, M., Yamamot o, H., and Marunouchi, T. Transforming growth factor-beta suppresse s activation and proliferati on of microglia in vitro. J Immunol, 151: 2150-2158, 1993. 124. Zou, J. P., Morford, L. A., Chougnet, C., Dix, A. R., Brooks, A. G., Torres, N., Shuman, J. D., Coligan, J. E., Brooks, W. H., Roszman, T. L., and Shearer, G. M. Human glioma-induced immunosuppression invo lves soluble factor(s) that alters monocyte cytokine profile and surface markers. J Immunol, 162: 4882-4892, 1999. 125. Weller, M. and Fontana, A. The failure of current immunotherapy for malignant glioma. Tumor-derived TGF-beta, T-cell a poptosis, and the immune privilege of the brain. Brain Res Brain Res Rev, 21: 128-151, 1995. 126. Wischhusen, J., Jung, G., Radovanovic, I., Be ier, C., Steinbach, J. P., Rimner, A., Huang, H., Schulz, J. B., Ohgaki, H., Agu zzi, A., Rammensee, H. G., and Weller, M. Identification of CD70-mediated apoptos is of immune effector cells as a novel immune escape pathway of huma n glioblastoma. Cancer Res, 62: 2592-2599, 2002. 127. O'Connell, J., Bennett, M. W., O'Sullivan, G. C., Collins, J. K., and Shanahan, F. The Fas counterattack: cancer as a site of immune privilege. Immunol Today, 20: 46-52, 1999. 128. Saas, P., Walker, P. R., Hahne, M., Qui querez, A. L., Schnuriger, V., Perrin, G., French, L., Van Meir, E. G., de Tribolet , N., Tschopp, J., and Dietrich, P. Y. Fas
130 ligand expression by astrocytoma in vivo: maintaining immune privilege in the brain? J Clin Invest, 99: 1173-1178, 1997. 129. Overwijk, W. W., Theoret, M. R., Finkels tein, S. E., Surman, D. R., de Jong, L. A., Vyth-Dreese, F. A., Dellemijn, T. A., Antony, P. A., Spiess, P. J., Palmer, D. C., Heimann, D. M., Klebanoff, C. A., Yu, Z., Hwang, L. N., Feigenbaum, L., Kruisbeek, A. M., Rosenberg, S. A., and Restifo, N. P. Tumor regression and autoimmunity after reversal of a functiona lly tolerant state of self-reactive CD8+ T cells. J Exp Med, 198: 569-580, 2003. 130. Eberlein, T. J., Rosenstein, M., Spiess , P., Wesley, R., and Rosenberg, S. A. Adoptive chemoimmunotherapy of a syngene ic murine lymphoma with long-term lymphoid cell lines expanded in T ce ll growth factor. Cancer Immunol Immunother, 13: 5-13, 1982. 131. Hanson, H. L., Donermeyer, D. L., Ikeda, H., White, J. M., Shankaran, V., Old, L. J., Shiku, H., Schreiber, R. D., and Allen, P. M. Eradication of established tumors by CD8+ T cell adoptive immunotherapy. Immunity, 13: 265-276, 2000. 132. Dudley, M. E., Wunderlich, J. R., Robbins, P. F., Yang, J. C., Hwu, P., Schwartzentruber, D. J., Topalian, S. L., Sherry, R., Restifo, N. P., Hubicki, A. M., Robinson, M. R., Raffeld, M., Duray, P., Seipp, C. A., Rogers-Freezer, L., Morton, K. E., Mavroukakis, S. A., White, D. E., and Rosenberg, S. A. Cancer regression and autoimmunity in patients after clonal repopulation with antitumor lymphocytes. Science, 298: 850-854, 2002. 133. Rosenberg, S. A., Yang, J. C., and Res tifo, N. P. Cancer immunotherapy: moving beyond current vaccines. Nat Med, 10: 909-915, 2004. 134. Plautz, G. E., Barnett, G. H., Miller, D. W., Cohen, B. H., Prayson, R. A., Krauss, J. C., Luciano, M., Kangisser, D. B., and Shu, S. Systemic T cell adoptive immunotherapy of malignant gliomas. J Neurosurg, 89: 42-51, 1998. 135. Plautz, G. E., Miller, D. W ., Barnett, G. H., Stevens, G. H., Maffett, S., Kim, J., Cohen, P. A., and Shu, S. T cell adoptiv e immunotherapy of newly diagnosed gliomas. Clin Cancer Res, 6: 2209-2218, 2000. 136. Plautz, G. E. and Shu, S. Systemic T-Cell Immunotherapy for Brain Tumors. In: L. M. Liau, D. P. Becker, T. F. Cloughe sy, and D. D. Bigner (eds.), Brain Tumor Immunotherapy, pp. 133-148. Tototwa: Humana Press, 2001. 137. Ridgway, D. The first 1000 dendri tic cell vaccinees. Cancer Invest, 21: 873-886, 2003.
131 138. Ribas, A., Butterfield, L. H., Glaspy, J. A., and Economou, J. S. Current developments in cancer vaccines and cellular immunotherapy. J Clin Oncol, 21: 2415-2432, 2003. 139. Carpentier, A. F. and Meng, Y. Recent advances in immunotherapy for human glioma. Curr Opin Oncol, 18: 631-636, 2006. 140. Yamanaka, R., Yajima, N., Abe, T., Tsuchiya, N., Homma, J., Narita, M., Takahashi, M., and Tanaka, R. Dendrit ic cell-based glioma immunotherapy (review). Int J Oncol, 23: 5-15, 2003. 141. Coley, W. B. The treatment of malignant tumors by repeated inoculations of erysipelas: with a repo rt of ten original cases. Am J Med Sci, 105: 487-511, 1893. 142. Coley, W. B. Treatment of inoperable malignant tumors with the toxines of erysipelas and the bacillus prodigiosus. Am J Med Sci, 108: 50-66, 1894. 143. Coley, W. B. The treatment of inopera ble sarcoma with the mixed toxins of erysipelas and Bacillus prodigiosus. J Am Med Assoc, 31: 389-395, 1898. 144. Gratia, A. and Linz, R. Le phenomene de Shwartzman dans le sarcome du cobaye. Compt Rend Soc de Biol, 108: 427-428, 1931. 145. Shear, M. J. Studies on the chemical trea tment of tumors. II. The effect of disturbances in fluid exchange on transplanted mouse tumors. Am J Cancer, 25: 66-88, 1935. 146. Shwartzman, G. and Michailovsky, N. Phenomenon of local skin reactivity to bacterial filtrates in the treatment of mouse sa rcoma 180. Proc Soc Exper Biol Med, 29: 737-741, 1932. 147. Shear, M. J., Turner, F. C., Perrault, A ., and Shovelton, T. Chemical treatment of tumors. V. Isolation of the hemo rrhage-producing fraction from Serratia marcescens (Bacillus prodigiosus) cultu re filtrate. J Natl Cancer Inst, 4: 81-97, 1943. 148. Beutler, B. and Rietschel, E. T. Innate immune sensing and its roots: the story of endotoxin. Nat Rev Immunol, 3: 169-176, 2003. 149. Freudenberg, M. A. and Galanos, C. Bacterial lipopolysacchar ides: structure, metabolism and mechanisms of action. Int Rev Immunol, 6: 207-221, 1990. 150. Berendt, M. J., North, R. J., and Kirste in, D. P. The immunological basis of endotoxin-induced tumor regression. Require ment for T-cell-mediated immunity. J Exp Med, 148: 1550-1559, 1978.
132 151. Algire, G. H., Legallis, F. Y., and Par k, H. D. Vascular reactions of normal and malignant tissues in vivo. II. The vasc ular reactions of normal and neoplastic tissues of mice to a bacterial polysacch aride from Serratia marescens (Bacillus prodigeosus) culture filtrate s. J Natl Cancer Inst, 8: 53, 1948. 152. Havell, E. A., Fiers, W., and North, R. J. The antitumor function of tumor necrosis factor (TNF), I. Therapeutic ac tion of TNF against an established murine sarcoma is indirect, immunologically depe ndent, and limited by severe toxicity. J Exp Med, 167: 1067-1085, 1988. 153. Carswell, E. A., Old, L. J., Kassel, R. L., Green, S., Fiore, N., and Williamson, B. An endotoxin-induced serum factor that causes necrosis of tumors. Proc Natl Acad Sci U S A, 72: 3666-3670, 1975. 154. Berendt, M. J., North, R. J., and Kirste in, D. P. The immunological basis of endotoxin-induced tumor regression. Requi rement for a pre-existing state of concomitant anti-tumor immunity. J Exp Med, 148: 1560-1569, 1978. 155. Bloksma, N. and Hofhuis, F. M. Synerg istic action of human recombinant tumor necrosis factor with endotoxins or nont oxic poly A:U against solid Meth A tumors in mice. Cancer Immunol Immunother, 24: 165-171, 1987. 156. Engelhardt, R., Mackensen, A., Gala nos, C., and Andreesen, R. Biological response to intravenously administered endotoxin in patients with advanced cancer. J Biol Response Mod, 9: 480-491, 1990. 157. Goto, S., Sakai, S., Kera, J., Suma, Y., Soma, G. I., and Takeuchi, S. Intradermal administration of lipopolysaccharide in treatment of human cancer. Cancer Immunol Immunother, 42: 255-261, 1996. 158. Otto, F., Schmid, P., Mackensen, A., Wehr , U., Seiz, A., Braun, M., Galanos, C., Mertelsmann, R., and Engelhardt, R. Phas e II trial of intravenous endotoxin in patients with colorectal and non-sm all cell lung cancer. Eur J Cancer, 32A: 17121718, 1996. 159. Galanos, C., Luderitz, O., Rietschel, E. T., Westphal, O., Brade, H., Brade, L., Freudenberg, M., Schade, U., Imoto, M., Yo shimura, H., and et al. Synthetic and natural Escherichia coli free lipid A expre ss identical endotoxic activities. Eur J Biochem, 148: 1-5, 1985. 160. Jeannin, J. F., Onier, N., Lagadec, P. , von Jeney, N., Stutz, P., and Liehl, E. Antitumor effect of synthetic derivatives of lipid A in an experimental model of colon cancer in the rat. Gastroenterology, 101: 726-733, 1991. 161. Kumazawa, E., Akimoto, T., Kita, Y., Jimbo, T., Joto, N., and Tohgo, A. Intratumoral production of tumor necr osis factor augmented by endogenous
133 interferons results in pot ent antitumor effects of DT-5461, a synthetic lipid A analog. J Immunother Emphasis Tumor Immunol, 17: 141-150, 1995. 162. Onier, N., Hilpert, S., Arnould, L., Saint-Gi orgio, V., Davies, J. G., Jeannin, J. F., and Jeannin, J. F. Cure of colon cancer me tastasis in rats with the new lipid A OM 174. Apoptosis of tumor cells and immuni zation of rats. Clin Exp Metastasis, 17: 299-306, 1999. 163. Won, E. K., Zahner, M. C., Grant, E. A ., Gore, P., and Chicoine, M. R. Analysis of the antitumoral mechanisms of lipopolysaccharide against glioblastoma multiforme. Anticancer Drugs, 14: 457-466, 2003. 164. de Bono, J. S., Dalgleish, A. G., Carmichael , J., Diffley, J., Loft s, F. J., Fyffe, D., Ellard, S., Gordon, R. J., Brindley, C. J ., and Evans, T. R. Phase I study of ONO4007, a synthetic analogue of the lipid A moiety of bacterial lipopolysaccharide. Clin Cancer Res, 6: 397-405, 2000. 165. Conley, F. K. and Remington, J. S. Nonsp ecific inhibition of tumor growth in the central nervous system: obser vations of intracerebral ependymoblastoma in mice with chronic Toxoplasma infection. J Natl Cancer Inst, 59: 963-973, 1977. 166. Kennedy, J. D. and Conley, F. K. E ffect of intracerebrally injected Corynebacterium parvum on implanted brain tumor in mice. J Neurooncol, 7: 89101, 1989. 167. Kennedy, J. D., Sutton, R. C., and Conley, F. K. Effect of intracerebrally injected Corynebacterium parvum on the developm ent and growth of metastatic brain tumor in mice. Neurosurgery, 25: 709-714, 1989. 168. Kiya, K., Toge, T., Harada, K., Uozumi, T ., and Hattori, T. Effect of intracerebral administration of Corynebacterium parvum on the growth of brain tumors in mice. Gann, 72: 446-450, 1981. 169. Mahaley, M. S., Jr., Bigner, D. D., Dudka , L. F., Wilds, P. R., Williams, D. H., Bouldin, T. W., Whitaker, J. N., and B ynum, J. M. Immunobiology of primary intracranial tumors. Part 7: Active imm unization of patients with anaplastic human glioma cells: a pilot study. J Neurosurg, 59: 201-207, 1983. 170. Mahaley, M. S., Jr., Gentry, R. E., and Bigner, D. D. Immunobiology of primary intracranial tumors. J Neurosurg, 47: 35-43, 1977. 171. Mahaley, M. S., Jr., Steinbok, P., Aronin, P., Dudka, L., and Zinn, D. Immunobiology of primary intrac ranial tumors. Part 4: levamisole as an immune stimulant in patients and in the ASV glioma model. J Neurosurg, 54: 220-227, 1981.
134 172. Farkkila, M., Jaaskelainen, J., Kallio, M., Blomstedt, G., Raininko, R., Virkkunen, P., Paetau, A., Sarelin, H., and Mantyla, M. Randomised, controlled study of intratumoral recombinant gamma-inte rferon treatment in newly diagnosed glioblastoma. Br J Cancer, 70: 138-141, 1994. 173. Mahaley, M. S., Jr., Bertsch, L., Cush, S., and Gillespie, G. Y. Systemic gammainterferon therapy for recurre nt gliomas. J Neurosurg, 69: 826-829, 1988. 174. Carpentier, A. F., Xie, J., Mokhtari, K., and Delattre, J. Y. Successful treatment of intracranial gliomas in ra t by oligodeoxynucleotides containing CpG motifs. Clin Cancer Res, 6: 2469-2473, 2000. 175. Carpentier, A., Laigle-Donadey, F., Zohar, S., Capelle, L., Behin, A., Tibi, A., Martin-Duverneuil, N., Sanson, M., Lacomb lez, L., Taillibert, S., Puybasset, L., Van Effenterre, R., Delattre, J. Y., and Ca rpentier, A. F. Phase 1 trial of a CpG oligodeoxynucleotide for patients with recurrent glioblastoma. Neuro-oncol, 8: 60-66, 2006. 176. Bsibsi, M., Ravid, R., Gveric, D., and va n Noort, J. M. Broad expression of Tolllike receptors in the human central ner vous system. J Neuropathol Exp Neurol, 61: 1013-1021, 2002. 177. Olson, J. K. and Miller, S. D. Microglia initiate central nervous system innate and adaptive immune responses through multiple TLRs. J Immunol, 173: 3916-3924, 2004. 178. Fitch, M. T., Doller, C., Combs, C. K., La ndreth, G. E., and Silver, J. Cellular and molecular mechanisms of glial scarring and progressive cavitation: in vivo and in vitro analysis of inflammation-induced secondary injury after CNS trauma. J Neurosci, 19: 8182-8198, 1999. 179. Popovich, P. G., Guan, Z., McGaughy, V., Fisher, L., Hickey, W. F., and Basso, D. M. The neuropathological and beha vioral consequences of intraspinal microglial/macrophage activati on. J Neuropathol Exp Neurol, 61: 623-633, 2002. 180. Colton, C. A., Yao, J., Keri, J. E., an d Gilbert, D. Regul ation of microglial function by interferons. J Neuroimmunol, 40: 89-98., 1992. 181. Corradin, S. B., Mauel, J., Donini, S. D., Quattrocchi, E., and RicciardiCastagnoli, P. Inducible nitr ic oxide synthase activity of cloned muri ne microglial cells. Glia, 7: 255-262., 1993. 182. Czapiga, M. and Colton, C. A. Function of microglia in organo typic slice cultures. J Neurosci Res, 56: 644-651., 1999.
135 183. Sankarapandi, S., Zweier, J. L., Mukherjee, G., Quinn, M. T., and Huso, D. L. Measurement and characterization of supe roxide generation in microglial cells: evidence for an NADPH oxidase-depende nt pathway. Arch Biochem Biophys, 353: 312-321., 1998. 184. Murata, J., Ricciardi-Castagnoli, P., De ssous L'Eglise Mange, P., Martin, F., and Juillerat-Jeanneret, L. Microglial cells induce cytotoxic effects toward colon carcinoma cells: measurement of tumor cytotoxicity with a gamma-glutamyl transpeptidase assay. Int J Cancer, 70: 169-174, 1997. 185. Sutter, A., Hekmat, A., and Luckenbach, G. A. Antibody-mediated tumor cytotoxicity of microglia. Pathobiology, 59: 254-258, 1991. 186. Walter, S. A. Anti-tumor response of glioma-associated microglia: modulation by lipopolysaccharide, interferon-gamma and transforming growth factor-beta. Neuroscience, pp. 160. Gainesville: University of Florida, 1998. 187. Chang, G. H., Barbaro, N. M., and Piep er, R. O. Phosphatidylserine-dependent phagocytosis of apoptotic glioma cells by normal human microglia, astrocytes, and glioma cells. Neuro-oncol, 2: 174-183, 2000. 188. Sawada, M., Kondo, N., Suzumura, A., and Marunouchi, T. Production of tumor necrosis factor-alpha by microglia and astrocytes in culture. Brain Res, 491: 394397., 1989. 189. Tansley, K. The development of th e rat eye in graft. J Exp Biol, 22: 221-223, 1946. 190. Cserr, H. F., Harling-Berg, C. J., and K nopf, P. M. Drainage of brain extracellular fluid into blood and deep cervical lym ph and its immunological significance. Brain Pathol, 2: 269-276, 1992. 191. Aloisi, F., Ria, F., and Adorini, L. Regulation of T-cell responses by CNS antigen-presenting cells: different role s for microglia and astrocytes. Immunol Today, 21: 141-147, 2000. 192. Graeber, M. B., Scheithauer, B. W., and Kreutzberg, G. W. Microglia in brain tumors. Glia, 40: 252-259, 2002. 193. Barth, R. F. Rat brain tumor models in experimental neuro-on cology: the 9L, C6, T9, F98, RG2 (D74), RT-2 and CNS-1 gliomas. J Neurooncol, 36: 91-102, 1998. 194. Parsa, A. T., Chakrabarti, I., Hurley, P. T., Chi, J. H., Hall, J. S., Kaiser, M. G., and Bruce, J. N. Limitations of the C6/Wistar rat intracer ebral glioma model: implications for evaluating immunotherapy. Neurosurgery, 47: 993-999; discussion 999-1000, 2000.
136 195. Ko, L., Koestner, A., and Wechsler, W. Characterization of cell cycle and biological parameters of transplantable glioma cell lines and clones. Acta Neuropathol (Berl), 51: 107-111, 1980. 196. Stojiljkovic, M., Piperski, V., Dacevic, M., Rakic, L., Ruzdijic, S., and Kanazir, S. Characterization of 9L glioma m odel of the Wistar rat. J Neurooncol, 63: 1-7, 2003. 197. Saini, M., Bellinzona, M., Meyer, F., Cali, G., and Samii, M. Morphometrical characterization of two glioma models in the brain of immunocompetent and immunodeficient rats. J Neurooncol, 42: 59-67, 1999. 198. Albright, A. L., Gill, T. J., 3rd, and Ge yer, S. J. Immunogenetic control of brain tumor growth in rats. Cancer Res, 37: 2512-2521, 1977. 199. Lodin, Z., Hasek, M., Chutna, J., Sladecek, M., and Holan, V. Transplantation immunity in the brain. J Neurosci Res, 3: 275-280, 1977. 200. Scheinberg, L. C., Edelman, F. L ., and Levy, W. A. Is the Brain "an Immunologically Privileged Site"?I. Studies Based on Intracerebral Tumor Homotransplantation and Isotransplantati on to Sensitized Hosts. Arch Neurol, 11: 248-264, 1964. 201. Scheinberg, L. C., Levy, A., and Edelma n, F. Is the brain an "immunologically privileged site"? 2. Studies in induced host resistance to transplantable mouse glioma following irradiation of prior implants. Arch Neurol, 13: 283-286, 1965. 202. Chrisman, C. L., Mariani, C.L., Platt, S.R., Clemmons, R.C. Neurology for the Small Animal Practitioner, p. 353. Jackson: Teton New Media, 2003. 203. Hasek, M., Chutna, J., Sladecek, M., a nd Lodin, Z. Immunological tolerance and tumor allografts in the brain. Nature, 268: 68-69, 1977. 204. Weller, R. O., Kida, S., and Zhang, E. T. Pathways of fluid drainage from the brain--morphological aspects and immunol ogical significance in rat and man. Brain Pathol, 2: 277-284, 1992. 205. Watters, J. J., Schartner, J. M., and Badi e, B. Microglia function in brain tumors. J Neurosci Res, 81: 447-455, 2005. 206. Kreutzberg, G. W. Microglia: a sensor for pathological events in the CNS. Trends Neurosci, 19: 312-318, 1996.
137 207. Perry, V. H. A revised view of the cen tral nervous system microenvironment and major histocompatibility complex class II antigen presentation. J Neuroimmunol, 90: 113-121, 1998. 208. Calzascia, T., Di Berardino-Besson, W ., Wilmotte, R., Masson, F., de Tribolet, N., Dietrich, P. Y., and Walker, P. R. Cutting edge: cross-presentation as a mechanism for efficient recruitment of tumor-specific CTL to the brain. J Immunol, 171: 2187-2191, 2003. 209. Okada, H., Tsugawa, T., Sato, H., Kuwashima, N., Gambotto, A., Okada, K., Dusak, J. E., Fellows-Mayle, W. K., Papwor th, G. D., Watkins, S. C., Chambers, W. H., Potter, D. M., Storkus, W. J., a nd Pollack, I. F. Delivery of interferonalpha transfected dendritic cells into centr al nervous system tumors enhances the antitumor efficacy of peripheral peptide-based vaccines. Cancer Res, 64: 58305838, 2004. 210. Jander, S., Schroeter, M., D'Urso, D., G illen, C., Witte, O. W., and Stoll, G. Focal ischaemia of the rat brain elicits an unusual inflammatory response: early appearance of CD8+ macrophage s/microglia. Eur J Neurosci, 10: 680-688, 1998. 211. Perry, V. H. and Gordon, S. Modulation of CD4 antigen on macrophages and microglia in rat brain. J Exp Med, 166: 1138-1143, 1987. 212. Popovich, P. G., van Rooijen, N., Hicke y, W. F., Preidis, G., and McGaughy, V. Hematogenous macrophages express CD8 a nd distribute to regions of lesion cavitation after spinal co rd injury. Exp Neurol, 182: 275-287, 2003. 213. Schroeter, M., Jander, S., Huitinga, I., a nd Stoll, G. CD8+ phagocytes in focal ischemia of the rat brain: predominan t origin from hematogenous macrophages and targeting to areas of pannecr osis. Acta Neuropathol (Berl), 101: 440-448, 2001. 214. Weissenbock, H., Hornig, M., Hickey, W. F., and Lipkin, W. I. Microglial activation and neuronal apoptosis in Born avirus infected neonatal Lewis rats. Brain Pathol, 10: 260-272, 2000. 215. Hirji, N., Lin, T. J., and Befus, A. D. A novel CD8 molecule expressed by alveolar and peritoneal macrophages stimulates nitric oxide production. J Immunol, 158: 1833-1840, 1997. 216. Hirji, N., Lin, T. J., Bissonnette, E., Belo sevic, M., and Befus, A. D. Mechanisms of macrophage stimulation through CD8: macrophage CD8alpha and CD8beta induce nitric oxide producti on and associated killing of the parasite Leishmania major. J Immunol, 160: 6004-6011, 1998.
138 217. Aloisi, F., De Simone, R., Columba-Ca bezas, S., Penna, G., and Adorini, L. Functional maturation of adult mouse resti ng microglia into an APC is promoted by granulocyte-macrophage colony-stimula ting factor and interaction with Th1 cells. J Immunol, 164: 1705-1712, 2000. 218. Dobrovolskaia, M. A. and Vogel, S. N. Toll receptors, CD14, and macrophage activation and deactivation by LPS. Microbes Infect, 4: 903-914, 2002. 219. Williams, M. A., Kelsey, S. M., and Newland, A. C. GM-CSF and stimulation of monocyte/macrophage function in vivo rele vance and in vitro observations. Eur J Cancer, 35 Suppl 3: S18-22, 1999. 220. Young, S. H., Ye, J., Frazer, D. G., Shi, X., and Castranova, V. Molecular mechanism of tumor necrosis factor-a lpha production in 1-->3-beta-glucan (zymosan)-activated macrophages. J Biol Chem, 276: 20781-20787, 2001. 221. Miura, T., Ohno, N., Miura, N. N., Adachi, Y., Shimada, S., and Yadomae, T. Antigen-specific response of murine imm une system toward a yeast beta-glucan preparation, zymosan. FEMS Immunol Med Microbiol, 24: 131-139, 1999. 222. Pistoia, V. Granulocyte-macrophage colony stimulating factor (GM-CSF); sources, targets and mechanism of action. Leukemia, 5 Suppl 1: 114-118, 1991. 223. Wallenfriedman, M. A., Conr ad, J. A., DelaBarre, L., Graupman, P. C., Lee, G., Garwood, M., Gregerson, D. S., Jean, W. C ., Hall, W. A., and Low, W. C. Effects of continuous localized infusion of gr anulocyte-macrophage colony-stimulating factor and inoculations of irradiat ed glioma cells on tumor regression. J Neurosurg, 90: 1064-1071, 1999. 224. Dranoff, G. GM-CSF-based cancer vaccines. Immunol Rev, 188: 147-154, 2002. 225. Dranoff, G. GM-CSF-secreting melanoma vaccines. Oncogene, 22: 3188-3192, 2003. 226. Reinisch, W., Holub, M., Katz, A., He rneth, A., Lichtenberger, C., SchonigerHekele, M., Waldhoer, T., Oberhuber, G., Fe renci, P., Gangl, A., and Mueller, C. Prospective pilot study of recombin ant granulocyte-macrophage colonystimulating factor and interferon-ga mma in patients with inoperable hepatocellular carcinoma. J Immunother, 25: 489-499, 2002. 227. Salgia, R., Lynch, T., Skarin, A., Lucca, J., Lynch, C., Jung, K., Hodi, F. S., Jaklitsch, M., Mentzer, S., Swanson, S ., Lukanich, J., Bueno, R., Wain, J., Mathisen, D., Wright, C., Fidias, P., D onahue, D., Clift, S., Hardy, S., Neuberg, D., Mulligan, R., Webb, I., Sugarbaker, D., Mihm, M., and Dranoff, G. Vaccination with irradiated autologous tumor cells engineered to secrete granulocyte-macrophage colony-stimulating factor augments antitumor immunity
139 in some patients with metastatic non-sm all-cell lung carcinoma. J Clin Oncol, 21: 624-630, 2003. 228. Verra, N., Jansen, R., Groenewegen, G., Mallo, H., Kersten, M. J., Bex, A., VythDreese, F. A., Sein, J., van de Kastee le, W., Nooijen, W. J., de Waal, M., Horenblas, S., and de Gast, G. C. Immunot herapy with concurrent subcutaneous GM-CSF, low-dose IL-2 and IFN-alpha in patients with progressive metastatic renal cell carcinoma. Br J Cancer, 88: 1346-1351, 2003. 229. Boehm, U., Klamp, T., Groot, M., and Howard, J. C. Cellular responses to interferon-gamma. Annu Rev Immunol, 15: 749-795, 1997. 230. Dutta, T., Spence, A., and Lampson, L. A. Robust ability of IFN-gamma to upregulate class II MHC antigen expre ssion in tumor bearing rat brains. J Neurooncol, 64: 31-44, 2003. 231. Tuttle, T. M., McCrady, C. W., Inge, T. H., Salour, M., and Bear, H. D. gammaInterferon plays a key role in T-cell-i nduced tumor regression. Cancer Res, 53: 833-839, 1993. 232. Winter, H., Hu, H. M., Urba, W. J., and Fox, B. A. Tumor regression after adoptive transfer of effector T cells is independent of perforin or Fas ligand (APO-1L/CD95L). J Immunol, 163: 4462-4472, 1999. 233. Farber, J. M. A collection of mRNA sp ecies that are inducible in the RAW 264.7 mouse macrophage cell line by gamma inte rferon and other agents. Mol Cell Biol, 12: 1535-1545, 1992. 234. Pace, J. L., Russell, S. W., Torres, B. A., Johnson, H. M., and Gray, P. W. Recombinant mouse gamma interferon indu ces the priming step in macrophage activation for tumor ce ll killing. J Immunol, 130: 2011-2013, 1983. 235. Held, T. K., Weihua, X., Yuan, L., Kalv akolanu, D. V., and Cross, A. S. Gamma interferon augments macrophage activa tion by lipopolysaccharide by two distinct mechanisms, at the signal transduction level and via an autocrine mechanism involving tumor necrosis factor alph a and interleukin-1. Infect Immun, 67: 206212, 1999. 236. Carpentier, A. F., Chen, L., Maltonti, F., and Delattre, J. Y. Oligodeoxynucleotides containing CpG motifs can induce rejection of a neuroblastoma in mice. Cancer Res, 59: 5429-5432, 1999. 237. Bobo, R. H., Laske, D. W., Akbasak, A., Morrison, P. F., Dedrick, R. L., and Oldfield, E. H. Convection-enhanced de livery of macromolecules in the brain. Proc Natl Acad Sci U S A, 91: 2076-2080, 1994.
140 238. Bruce, J. N., Falavigna, A., Johnson, J. P ., Hall, J. S., Birch, B. D., Yoon, J. T., Wu, E. X., Fine, R. L., and Parsa, A. T. In tracerebral clysis in a rat glioma model. Neurosurgery, 46: 683-691, 2000. 239. Degen, J. W., Walbridge, S., Vortmeyer, A. O., Oldfield, E. H., and Lonser, R. R. Safety and efficacy of convection-enhanced delivery of gemcitabine or carboplatin in a malignant glioma model in rats. J Neurosurg, 99: 893-898, 2003. 240. Mamot, C., Nguyen, J. B., Pourdehnad, M ., Hadaczek, P., Saito, R., Bringas, J. R., Drummond, D. C., Hong, K ., Kirpotin, D. B., McKni ght, T., Berger, M. S., Park, J. W., and Bankiewicz, K. S. Extens ive distribution of liposomes in rodent brains and brain tumors following convec tion-enhanced delivery. J Neurooncol, 68: 1-9, 2004. 241. Boiardi, A., Bartolomei, M., Silvani, A., Eoli, M., Salmaggi, A., Lamperti, E., Milanesi, I., Botturi, A., Rocca, P., Bode i, L., Broggi, G., and Paganelli, G. Intratumoral delivery of mitoxant rone in association with 90-Y radioimmunotherapy (RIT) in recu rrent glioblastoma. J Neurooncol, 72: 125-131, 2005. 242. McDermott, M. W., Berger, M. S., Kunwar, S., Parsa, A. T., Sneed, P. K., and Larson, D. A. Stereotactic radiosurgery and interstitial br achytherapy for glial neoplasms. J Neurooncol, 69: 83-100, 2004. 243. Wikstrand, C. J., Zalutsky, M. R., and Bi gner, D. D. Radiolabeled Antibodies for Therapy of Brain Tumors. In: L. M. Liau, D. P. Becker, T. F. Cloughesy, and D. D. Bigner (eds.), Brain Tumor Immu notherapy, pp. 205-229. Totowa: Humana Press, 2001. 244. Hall, W. A. Immunotoxin Therapy of Brain Tumors. In: L. M. Liau, D. P. Becker, T. F. Cloughesy, and D. D. Bigner (e ds.), Brain Tumor Immunotherapy, pp. 231247. Totowa: Humana Press, 2001. 245. Kruse, C. A. and Merchant, R. E. Cellu lar Therapy of Brain Tumors: Clinical Trials. In: P. L. Kornblith and M. D. Walker (eds.), Advances in NeuroOncology, pp. 487-504. Armonk: Futura, 1997. 246. Kruse, C. A. and Rubinstein, D. Cyto toxic T-lymphocytes Reactive to Patient Major Histocompatibility Complex Prot eins for Therapy of Brain Tumors. In: L. M. Liau, D. P. Becker, T. F. Cloughes y, and D. D. Bigner (eds.), Brain Tumor Immunotherapy, pp. 149-170. Totowa: Humana Press, 2001. 247. Steinbok, P., Thomas, J. P., Grossman, L., and Dolman, C. L. Intratumoral autologous mononuclear cells in the tr eatment of recurrent glioblastoma multiforme. A phase 1 (toxicity) study. J Neurooncol, 2: 147-151, 1984.
141 248. Vaquero, J., Martinez, R., Oya, S., Coca, S., Barbolla, L., Ramiro, J., and Salazar, F. G. Intratumoural injection of autologous lymphocytes plus human lymphoblastoid interferon for the treatment of glioblastoma. Acta Neurochir (Wien), 98: 35-41, 1989. 249. Chicoine, M. R., Won, E. K., and Zahne r, M. C. Intratumoral injection of lipopolysaccharide causes regression of subcutaneously implanted mouse glioblastoma multiforme. Neurosurgery, 48: 607-614; discussion 614-605, 2001. 250. Bruland, T., Dai, H. Y., Lavik, L. A., Kristiansen, L. I., and Dalen, A. Genderrelated differences in susceptibil ity, early virus dissemination and immunosuppression in mice infected with Fr iend murine leukaemia virus variant FIS-2. J Gen Virol, 82: 1821-1827, 2001. 251. Remoue, F., To Van, D., Schacht, A. M., Picquet, M., Garraud, O., Vercruysse, J., Ly, A., Capron, A., and Riveau, G. Gende r-dependent specific immune response during chronic human Schistosomiasi s haematobia. Clin Exp Immunol, 124: 6268, 2001. 252. Eidinger, D. and Garrett, T. J. Studies of the regulatory effects of the sex hormones on antibody formation and stem cell differentiation. J Exp Med, 136: 1098-1116, 1972. 253. Grossman, C. Possible underlying mechan isms of sexual dimorphism in the immune response, fact and hypot hesis. J Steroid Biochem, 34: 241-251, 1989. 254. Whitacre, C. C., Reingold, S. C., and O'Looney, P. A. A gender gap in autoimmunity. Science, 283: 1277-1278, 1999. 255. Yang, L., Hu, Y., and Hou, Y. Effects of 17beta-estradiol on the maturation, nuclear factor kappa B p65 and functions of murine spleen CD11c-positive dendritic cells. Mol Immunol, 43: 357-366, 2006. 256. Nakamura, H., Kawasaki, N., Hagiwara, M., Saito, M., Konaka, C., and Kato, H. Cellular immunologic parameters related to age, gender, and stage in lung cancer patients. Lung Cancer, 28: 139-145, 2000. 257. Gahrton, G., Iacobelli, S., Apperley, J., Bandini, G., Bjorkstrand, B., Blade, J., Boiron, J. M., Cavo, M., Cornelissen, J., Corradini, P., Kroger, N., Ljungman, P., Michallet, M., Russell, N. H., Samson, D ., Schattenberg, A., Sirohi, B., Verdonck, L. F., Volin, L., Zander, A., and Nieder wieser, D. The impact of donor gender on outcome of allogeneic hematopoietic stem cell transplantation for multiple myeloma: reduced relapse risk in fema le to male transplants. Bone Marrow Transplant, 35: 609-617, 2005.
142 258. Gratwohl, A., Hermans, J., Niederwies er, D., van Biezen, A., van Houwelingen, H. C., and Apperley, J. Female donors infl uence transplant-related mortality and relapse incidence in male recipients of sibling blood and marrow transplants. Hematol J, 2: 363-370, 2001. 259. Aas, A. T., Brun, A., Blennow, C., Stromb lad, S., and Salford, L. G. The RG2 rat glioma model. J Neurooncol, 23: 175-183, 1995. 260. Bailey, D. W. and Hoste, J. A gene gove rning the female immune response to the male antigen in mice. Transplantation, 11: 404-407, 1971. 261. Eichwald, E. J. and Silmser, C. R. Skin. Transplant Bull, 2: 148-149, 1955. 262. Vogt, M. H., de Paus, R. A., Voogt, P. J., Willemze, R., and Falkenburg, J. H. DFFRY codes for a new human male-sp ecific minor transplantation antigen involved in bone marrow graft rejection. Blood, 95: 1100-1105, 2000. 263. Shear, M. J. and Turner, F. C. Chemical treatment of tumors. V. Isolation of the hemorrhage-producing fraction from Serrati a marcescens (Bacil lus prodigiosus) culture filtrate. J Natl Cancer Inst, 4: 81-97., 1943. 264. Hoeller, C., Jansen, B., Heere-Ress, E., Pustelnik, T., Mossbacher, U., Schlagbauer-Wadl, H., Wolff, K., and Pehamb erger, H. Perilesion al injection of rGM-CSF in patients with cutaneous mela noma metastases. J Invest Dermatol, 117: 371-374, 2001. 265. Pasare, C. and Medzhitov, R. Toll path way-dependent blockade of CD4+CD25+ T cell-mediated suppression by dendritic cells. Science, 299: 1033-1036, 2003. 266. Matzinger, P. The danger model: a renewed sense of self. Science, 296: 301-305, 2002. 267. Zeisberger, E. and Roth, J. Tolera nce to pyrogens. Ann N Y Acad Sci, 856: 116131, 1998. 268. Yang, Y., Huang, C. T., Huang, X., and Pa rdoll, D. M. Persistent Toll-like receptor signals are required for revers al of regulatory T cell-mediated CD8 tolerance. Nat Immunol, 5: 508-515, 2004. 269. Celluzzi, C. M., Mayordomo, J. I., Storkus , W. J., Lotze, M. T., and Falo, L. D., Jr. Peptide-pulsed dendritic cells induce antigen-specific CTL-mediated protective tumor immunity. J Exp Med, 183: 283-287, 1996. 270. Chauvenet, P. H., McArthur, C. P., and Smith, R. T. Demonstration in vitro of cytotoxic T cells with apparent specific ity toward tumor-specific transplantation antigens on chemically induced tumors. J Immunol, 123: 2575-2581, 1979.
143 271. Fernandez-Cruz, E., Halliburton, B., and Feldman, J. D. In vivo elimination by specific effector cells of an establis hed syngeneic rat moloney virus-induced sarcoma. J Immunol, 123: 1772-1777, 1979. 272. Barth, R. J., Jr., Mule, J. J., Spiess, P. J., and Rosenberg, S. A. Interferon gamma and tumor necrosis factor have a role in tumor regressions mediated by murine CD8+ tumor-infiltrating lymphocytes. J Exp Med, 173: 647-658, 1991. 273. Shrikant, P. and Mescher, M. F. Contro l of syngeneic tumor growth by activation of CD8+ T cells: efficacy is limited by migr ation away from the site and induction of nonresponsiveness. J Immunol, 162: 2858-2866, 1999. 274. Pardoll, D. M. Cancer vaccines. Nat Med, 4: 525-531, 1998. 275. Jiang, H., Stewart, C. A., Fast, D. J., a nd Leu, R. W. Tumor target-derived soluble factor synergizes with IFN-gamma and IL-2 to activate macrophages for tumor necrosis factor and nitric oxide producti on to mediate cytotoxicity of the same target. J Immunol, 149: 2137-2146, 1992. 276. Bonnotte, B., Larmonier, N., Favre, N., Fromentin, A., Moutet, M., Martin, M., Gurbuxani, S., Solary, E., Chauffert, B., and Martin, F. Identification of tumorinfiltrating macrophages as the killers of tumor cells after immunization in a rat model system. J Immunol, 167: 5077-5083, 2001. 277. Evans, R. and Alexander, P. Cooperation of immune lymphoid cells with macrophages in tumour immunity. Nature, 228: 620-622, 1970. 278. Piessens, W. F., Churchill, W. H., Jr ., and David Macrophages activated in vitro with lymphocyte mediators kill neopla stic but not normal cells. J Immunol, 114: 293-299, 1975. 279. Berendt, M. J. and North, R. J. Tcell-mediated suppression of anti-tumor immunity. An explanation for progressive growth of an immunogenic tumor. J Exp Med, 151: 69-80, 1980. 280. Gordon, S. and Taylor, P. R. Monocyt e and macrophage heterogeneity. Nat Rev Immunol, 5: 953-964, 2005. 281. Leon, B., Lopez-Bravo, M., and Ardavin, C. Monocyte-derived dendritic cells. Semin Immunol, 17: 313-318, 2005. 282. Dijkstra, C. D., Dopp, E. A., Joling, P ., and Kraal, G. The heterogeneity of mononuclear phagocytes in lymphoid organs : distinct macrophage subpopulations in the rat recognized by monoclonal antibodies ED1, ED2 and ED3. Immunology, 54: 589-599, 1985.
144 283. Hubert, F. X., Voisine, C., Louvet, C., Heslan, J. M., Ouabed, A., Heslan, M., and Josien, R. Differential pattern recogniti on receptor expression but stereotyped responsiveness in rat spleen de ndritic cell subsets. J Immunol, 177: 1007-1016, 2006. 284. Trinite, B., Voisine, C., Yagita, H., and Josien, R. A subset of cytolytic dendritic cells in rat. J Immunol, 165: 4202-4208, 2000. 285. Voisine, C., Hubert, F. X., Trinite, B., Heslan, M., and Josien, R. Two phenotypically distinct subset s of spleen dendritic cells in rats exhibit different cytokine production and T cell s timulatory activity. J Immunol, 169: 2284-2291, 2002. 286. Chan, C. W., Crafton, E., Fan, H. N., Flook, J., Yoshimura, K., Skarica, M., Brockstedt, D., Dubensky, T. W., Stins, M. F., Lanier, L. L., Pardoll, D. M., and Housseau, F. Interferon-producing killer dendritic cells provide a link between innate and adaptive immunity. Nat Med, 12: 207-213, 2006. 287. Taieb, J., Chaput, N., Menard, C., Apetoh, L., Ullrich, E., Bonmort, M., Pequignot, M., Casares, N., Terme, M., Flament, C., Opolon, P., Lecluse, Y., Metivier, D., Tomasello, E., Vivier, E., Ghiringhelli, F., Martin, F., Klatzmann, D., Poynard, T., Tursz, T., Raposo, G., Yagita, H., Ryffel, B., Kroemer, G., and Zitvogel, L. A novel dendritic cell subset involved in tumor immunosurveillance. Nat Med, 12: 214-219, 2006. 288. Castro, M. G., Cowen, R., Williamson, I. K., David, A., Jimenez-Dalmaroni, M. J., Yuan, X., Bigliari, A., Williams, J. C., Hu, J., and Lowenstein, P. R. Current and future strategies for the treatment of malignant brain tumors. Pharmacol Ther, 98: 71-108, 2003. 289. Grossman, S. A. and Batara, J. F. Current management of glioblastoma multiforme. Semin Oncol, 31: 635-644, 2004. 290. Kaufmann, S. H. and Earnshaw, W. C. Induction of apoptosis by cancer chemotherapy. Exp Cell Res, 256: 42-49, 2000. 291. Lake, R. A. and Robinson, B. W. Imm unotherapy and chemotherapy--a practical partnership. Nat Rev Cancer, 5: 397-405, 2005. 292. Restifo, N. P. Building better vaccin es: how apoptotic cel l death can induce inflammation and activate innate and adaptive immunity. Curr Opin Immunol, 12: 597-603, 2000. 293. Nowak, A. K., Lake, R. A., Marzo, A. L., Scott, B., Heath, W. R., Collins, E. J., Frelinger, J. A., and Robinson, B. W. Induction of tumor cell apoptosis in vivo
145 increases tumor antigen cross-presenta tion, cross-priming rather than crosstolerizing host tumor-specific CD8 T cells. J Immunol, 170: 4905-4913, 2003. 294. McBride, W. H. Combining Radiat ion Therapy with Immunotherapy for Treatment of Brain Tumors. In: L. M. Liau, D. P. Becker, T. F. Cloughesy, and D. D. Bigner (eds.), Brain Tumor Immu notherapy, pp. 345-361. Totowa: Humana Press, 2001. 295. Nowak, A. K., Robinson, B. W., and Lake , R. A. Synergy between chemotherapy and immunotherapy in the treatment of es tablished murine solid tumors. Cancer Res, 63: 4490-4496, 2003. 296. Nakata, H., Yoshimine, T., Murasawa, A ., Kumura, E., Harada, K., Ushio, Y., and Hayakawa, T. Early blood-br ain barrier disruption after high-dose sing le-fraction irradiation in rats. Ac ta Neurochir (Wien), 136: 82-86; discussion 86-87, 1995. 297. Qin, D. X., Zheng, R., Tang, J., Li, J. X., and Hu, Y. H. Influence of radiation on the blood-brain barrier a nd optimum time of chemotherapy. Int J Radiat Oncol Biol Phys, 19: 1507-1510, 1990. 298. Trnovec, T., Kallay, Z., and Bezek, S. Ef fects of ionizing radiation on the blood brain barrier permeability to pharmacologically active substances. Int J Radiat Oncol Biol Phys, 19: 1581-1587, 1990. 299. Hallahan, D. E., Staba-Hogan, M. J., Virudachalam, S., and Kolchinsky, A. Xray-induced P-selectin local ization to the lumen of tumor blood vessels. Cancer Res, 58: 5216-5220, 1998. 300. Hong, J. H., Chiang, C. S., Campbell, I. L., Sun, J. R., Withers, H. R., and McBride, W. H. Induction of acute phase gene expression by br ain irradiation. Int J Radiat Oncol Biol Phys, 33: 619-626, 1995. 301. Gridley, D. S., Loredo, L. N., Slater, J. D., Archambeau, J. O., Bedros, A. A., Andres, M. L., and Slater, J. M. Pilot ev aluation of cytokine levels in patients undergoing radiotherapy for brain tumor. Cancer Detect Prev, 22: 20-29, 1998. 302. Rubinfeld, B., Upadhyay, A., Clark, S. L., Fong, S. E., Smith, V., Koeppen, H., Ross, S., and Polakis, P. Identificati on and immunotherapeutic targeting of antigens induced by chemotherapy. Nat Biotechnol, 24: 205-209, 2006. 303. Milas, L., Mason, K. A., Ariga, H., H unter, N., Neal, R., Valdecanas, D., Krieg, A. M., and Whisnant, J. K. CpG oligodeoxynucleotide enhances tumor response to radiation. Cancer Res, 64: 5074-5077, 2004.
146 304. Meng, Y., Carpentier, A. F., Chen, L., Boi sserie, G., Simon, J. M., Mazeron, J. J., and Delattre, J. Y. Successful combinati on of local CpG-ODN and radiotherapy in malignant glioma. Int J Cancer, 116: 992-997, 2005. 305. Balsari, A., Tortoreto, M., Besusso, D., Petrangolini, G., Sfondrini, L., Maggi, R., Menard, S., and Pratesi, G. Combina tion of a CpG-oligodeoxynucleotide and a topoisomerase I inhibitor in the ther apy of human tumour xenografts. Eur J Cancer, 40: 1275-1281, 2004. 306. van der Most, R. G., Himbeck, R., Aarons , S., Carter, S. J., Larma, I., Robinson, C., Currie, A., and Lake, R. A. Antitumor efficacy of the novel chemotherapeutic agent coramsine is potentiated by cotreatment with CpG-containing oligodeoxynucleotides. J Immunother, 29: 134-142, 2006. 307. Weigel, B. J., Rodeberg, D. A., Krieg, A. M., and Blazar, B. R. CpG oligodeoxynucleotides potentiate the anti tumor effects of chemotherapy or tumor resection in an orthotopic murine model of rhabdomyosarcoma. Clin Cancer Res, 9: 3105-3114, 2003. 308. Moroson, H. and Schechter, M. Treatm ent of rat fibrosarcoma by radiotherapy plus immune adjuvant. Biomedicine, 25: 97-100, 1976. 309. Gridley, D. S., Baer, J. R., Cao, J. D., M iller, G. M., Kim, D. W., Timiryasova, T. M., Fodor, I., and Slater, J. M. TNF-al pha gene and proton radiotherapy in an orthotopic brain tumor model. Int J Oncol, 21: 251-259, 2002. 310. Roth, W., Fontana, A., Trepel, M., Ree d, J. C., Dichgans, J., and Weller, M. Immunochemotherapy of malignant glioma : synergistic activity of CD95 ligand and chemotherapeutics. Cancer Immunol Immunother, 44: 55-63, 1997. 311. Plautz, G. E., Inoue, M., and Shu, S. Defi ning the synergistic e ffects of irradiation and T-cell immunotherapy for murine intracranial tumors. Cell Immunol, 171: 277-284, 1996. 312. Graf, M. R., Prins, R. M., Hawkins, W. T., and Merchant, R. E. Irradiated tumor cell vaccine for treatment of an establis hed glioma. I. Successful treatment with combined radiotherapy and cellular v accination. Cancer Immunol Immunother, 51: 179-189, 2002. 313. Lumniczky, K., Desaknai, S., Mangel, L., Szende, B., Hamada, H., Hidvegi, E. J., and Safrany, G. Local tumor irradiation augments the antitumor effect of cytokine-producing autologous cancer cell vaccines in a murine glioma model. Cancer Gene Ther, 9: 44-52, 2002. 314. Shin, J. Y., Lee, S. K., Kang, C. D., Chung, J. S., Lee, E. Y., Seo, S. Y., Lee, S. Y., Baek, S. Y., Kim, B. S., Kim, J. B., and Yoon, S. Antitumor effect of
147 intratumoral administration of dendritic cell combination with vincristine chemotherapy in a murine fibrosar coma model. Hist ol Histopathol, 18: 435-447, 2003. 315. Teitz-Tennenbaum, S., Li, Q., Rynkiewicz, S., Ito, F., Davis, M. A., McGinn, C. J., and Chang, A. E. Radiotherapy pote ntiates the therap eutic efficacy of intratumoral dendritic cell administration. Cancer Res, 63: 8466-8475, 2003. 316. Yu, B., Kusmartsev, S., Cheng, F., Paolin i, M., Nefedova, Y., Sotomayor, E., and Gabrilovich, D. Effective combination of chemotherapy and dendritic cell administration for the treatment of adva nced-stage experimental breast cancer. Clin Cancer Res, 9: 285-294, 2003. 317. Ali, S., King, G. D., Curtin, J. F., Candolfi, M., Xiong, W., Liu, C., Puntel, M., Cheng, Q., Prieto, J., Ribas, A., Kupiec-Weglinski, J., van Rooijen, N., Lassmann, H., Lowenstein, P. R., and Castro, M. G. Combined immunostimulation and conditional cytotoxic gene therapy provide long-term survival in a large glioma model. Cancer Res, 65: 7194-7204, 2005. 318. Mason, K. A., Ariga, H., Neal, R., Va ldecanas, D., Hunter, N., Krieg, A. M., Whisnant, J. K., and Milas, L. Targeting toll-like receptor 9 with CpG oligodeoxynucleotides enhances tumor res ponse to fractionated radiotherapy. Clin Cancer Res, 11: 361-369, 2005. 319. Luxton, G., Petrovich, Z., Jozsef, G., Nedzi, L. A., and Apuzzo, M. L. Stereotactic radiosurgery: principles and comparison of treatment methods. Neurosurgery, 32: 241-259; discussion 259, 1993. 320. Niranjan, A. and Lunsford, L. D. Radios urgery: where we were, are, and may be in the third millennium. Neurosurgery, 46: 531-543, 2000. 321. Nwokedi, E. C., DiBiase, S. J., Jabbour, S., Herman, J., Amin, P., and Chin, L. S. Gamma knife stereotactic radiosurgery for patients with glioblastoma multiforme. Neurosurgery, 50: 41-46; discussion 46-47, 2002. 322. Sarkaria, J. N., Mehta, M. P., Loeffler, J. S., Buatti, J. M., Chappell, R. J., Levin, A. B., Alexander, E., 3rd, Friedman, W. A., and Kinsella, T. J. Radiosurgery in the initial management of malignant g liomas: survival comparison with the RTOG recursive partitioning analysis. Ra diation Therapy Oncology Group. Int J Radiat Oncol Biol Phys, 32: 931-941, 1995. 323. Tsao, M. N., Mehta, M. P., Whelan, T. J., Morris, D. E., Hayman, J. A., Flickinger, J. C., Mills, M., Rogers, C. L., and Souhami, L. The American Society for Therapeutic Radiology and Oncology (ASTRO) evidence-based review of the role of radiosurgery for malignant gl ioma. Int J Radiat Oncol Biol Phys, 63: 4755, 2005.
148 324. Souhami, L., Seiferheld, W., Brachman, D., Podgorsak, E. B., Werner-Wasik, M., Lustig, R., Schultz, C. J., Sause, W., Okunieff, P., Buckner, J., Zamorano, L., Mehta, M. P., and Curran, W. J., Jr. Randomized comparison of stereotactic radiosurgery followed by conventiona l radiotherapy with carmustine to conventional radiotherapy with carmus tine for patients with glioblastoma multiforme: report of Radiation Therapy Oncology Group 93-05 protocol. Int J Radiat Oncol Biol Phys, 60: 853-860, 2004. 325. Hsieh, P. C., Chandler, J. P., Bhangoo, S., Panagiotopoulos, K., Kalapurakal, J. A., Marymont, M. H., Cozzens, J. W., Levy, R. M., and Salehi, S. Adjuvant gamma knife stereotactic radiosurgery at the time of tumor progression potentially improves survival for patients with glioblastoma multiforme. Neurosurgery, 57: 684-692; discussion 684-692, 2005. 326. Ulm, A. J., 3rd, Friedman, W. A., Bradsh aw, P., Foote, K. D., and Bova, F. J. Radiosurgery in the treatment of maligna nt gliomas: the University of Florida experience. Neurosurgery, 57: 512-517; discussion 512-517, 2005. 327. Bowles, A. P., Jr. and Perkins, E. Longterm remission of malignant brain tumors after intracranial infection: a re port of four cases. Neurosurgery, 44: 636-642; discussion 642-633, 1999. 328. Lacroix, M., Abi-Said, D., Fourney, D. R., Gokaslan, Z. L., Shi, W., DeMonte, F., Lang, F. F., McCutcheon, I. E., Hassenbusch, S. J., Holland, E., Hess, K., Michael, C., Miller, D., and Sawaya, R. A multivariate analysis of 416 patients with glioblastoma multiforme: prognosis, extent of resection, and survival. J Neurosurg, 95: 190-198, 2001. 329. Westphal, M., Hilt, D. C., Bortey, E., De lavault, P., Olivares, R., Warnke, P. C., Whittle, I. R., Jaaskelainen, J., and Ram, Z. A phase 3 trial of local chemotherapy with biodegradable carmustine (BCNU) wafe rs (Gliadel wafers) in patients with primary malignant glioma. Neuro-oncol, 5: 79-88, 2003. 330. Kondziolka, D., Mori, Y., Martinez, A. J., McLaughlin, M. R., Flickinger, J. C., and Lunsford, L. D. Beneficial effects of the radioprotectant 21-aminosteroid U74389G in a radiosurgery rat malignant gl ioma model. Int J Radiat Oncol Biol Phys, 44: 179-184, 1999. 331. McIntosh, T. K., Thomas, M., Smith, D., and Banbury, M. The novel 21aminosteroid U74006F attenuates cerebr al edema and improves survival after brain injury in the rat. J Neurotrauma, 9: 33-46, 1992. 332. Carpentier, A. F., Auf, G., and Delattre, J. Y. CpG-oligonucleotides for cancer immunotherapy: review of the literature and potential applications in malignant glioma. Front Biosci, 8: e115-127, 2003.
149 333. Walker, P. R., Calzascia, T., and Dietri ch, P. Y. All in the head: obstacles for immune rejection of br ain tumours. Immunology, 107: 28-38, 2002. 334. Bigner, D. D., Pitts, O. M., and Wikstra nd, C. J. Induction of lethal experimental allergic encephalomyelitis in nonhuman primates and guinea pigs with human glioblastoma multiforme tissue. J Neurosurg, 55: 32-42, 1981.
150 BIOGRAPHICAL SKETCH Christopher Leonard Mariani was born in Toronto, Ontario, Canada, and grew up in nearby Oakville. He graduated from St. Ignatius of Loyola Secondary School in 1988 and was admitted into the biological sciences program at the University of Guelph. In 1991, he was accepted into the DVM program at the Ontario Veterinary College, University of Guelph. After graduating w ith honors in 1996, Chris completed a rotating internship in small animal medicine and surg ery at Michigan Veterinary Specialists in Southfield, Michigan. From 1997 to 1998, he wa s an associate veterinarian at Beverly Hills Veterinary Associates in Beverly Hill s, Michigan before being accepted into a residency program in veterinary neurology and ne urosurgery at the University of Florida. Chris completed this residency program in 2001, after which he became board-certified by the American College of Veterinary Inte rnal Medicine (neurology subspecialty). Chris enrolled in the Interdisciplinary Program in Biomedical Sciences at the College of Medicine, University of Florida in 2001. Af ter graduation, Chris and his wife Rita will move to Raleigh, North Carolina, where they have accepted positions as assistant professors at North Carolina State Univer sityâ€™s College of Veterinary Medicine.