Cellular components of TME
The CNS has long been considered an « immuno-privileged » organ due to the presence of the BBB limiting the exchanges between the brain and the blood vessels. Leukocytes are the main cellular contributor in protection against tumors and infections. Prior to their release to the blood stream, they are developed in hematopoietic organs such as the bone marrow or/and thymus. Leukocytes migrate to the specific infected tissues following an activation process. However, due to the BBB, their passage to the CNS is limited through few passage ways: (i) through the choroid plexus, (ii) across superficial leptomeningeal vessels into the subarachnoid space and (iii) through the perivascular space into the brain parenchyma (Ratnam et al., 2019). The passage of leukocytes to the brain parenchyma is partially restricted by the BBB. Indeed, endothelial cell (ECs) within the BBB and their tight junctions prevent the passage of cells to the brain. Furthermore, pericytes around the vascular structures of the parenchyma maintain and support the integrity of the BBB. In physiological conditions, leukocytes are not detected in the brain parenchyma. However, under pathological conditions i.e., GBM, the integrity of the BBB is compromised allowing lymphocytes to reach the GBM TME (Weenink et al., 2020)
Accumulating evidence have suggested that tumor development is not only due to the accumulation of intrinsic abnormalities but also to extrinsic signals from the TME. Indeed, the TME, which is defined as a cellular (i.e. blood vessels, immune cells and fibroblasts), molecular (i.e. intercellular signaling molecules, extracellular matrix –ECM-), and dynamic network surrounding tumor cells, plays a significant role in tumor biology (Broekman et al., 2018). The tumor and the TME are linked in a highly interactive manner (Figure 6). They influence each other through extracellular signals (Huang et al., 2020).
Microglia and macrophages
Microglia, the brain-specific immune cells, ensure immunosurveillance of the brain parenchyma. Since the discovery of a cerebral lymphatic system (De Leo et al., 2020) and leukocyte extravasation to the GBM TME, microglia gained interest as an immune cellular component in the brain. Macrophages are the most abundant immune cells found amongst circulating leukocytes. Microglia corresponds to the resident macrophages in the CNS, while circulating macrophages originate from monocytes recruited from the blood.
Immunosuppression within GBM is characterized by enhancement of immune-suppressive cytokines and inhibition of T-lymphocytes proliferation. The presence of several immune-suppressive cytokines characterizes the GBM TME (e.g., Interleukins -IL-, IL-6 and IL-10, prostaglandin E2, IL-1, and transforming growth factor-beta -TGF-β). Each mediator affects the GBM immune TME in a specific matter. For example, TGF-β blocks the activation of T-lymphocytes, inhibits IL-2 production, and decreases NK-T lymphocytes activity. IL-2, a known immunosuppressive cytokine, is secreted mainly by macrophages and GBM cells within the TME. IL-2 enhances GBM cell growth and inhibits interferon-gamma (IFNγ) and tumor necrosis factor-alpha (TNF-). It is also associated with a downregulation of major histocompatibility complex (MHC) class II and enhancement of CD80/CD86 expression on the surface of infiltrating T-lymphocytes as well as on GBM cells (Scheffel et al., 2020).
Antigen presentation and tumor-associated macrophages (TAM)
Antigens released from tumor cells are processed by antigen presenting cells (APC) on MHC class I and presented to cytotoxic T-lymphocytes. Microglia cells have been identified with the ability to present antigens to T-lymphocytes within the CNS. However, the downregulation of APCs within the TME decreases microglia’s ability to exert this role. In GBM, macrophages derived from monocyte precursors polarize into two distinct categories within the GBM TME. Exposure to INF-γ polarizes monocytes to M1 macrophages. The role of M1 macrophages are pro-inflammatory cytokines and chemokines secretion. Therefore, participate in the positive immune response and function as an immune monitor. On the other hand, M2 macrophages are involved in the anti-inflammatory cytokine’s secretion therefore, reducing inflammation and contributing to immunosuppressive function and tumor growth (Grégoire et al., 2020).
M2 macrophages polarize through exposure to IL-4. TAMs are known to be capable of cross-presenting tumor antigens to T-lymphocytes and priming anti-tumor immune response (anti-inflammatory response). There is no definite answer about the importance of TAM in GBM antigen presentation. However, the presence of TAM is linked to GBM progression. Indeed, results from published articles reported that modulation of macrophage polarization has a regulatory effect on the GBM TME (Saha et al., 2017). The inhibitory effect of GBM TME rises from a regulatory link between M2 macrophages and tumor cells. Several factors such as colony-stimulating factor 1, TGF-1, macrophages inhibitory cytokines-1, and IL-10 can polarize TAMs to M2 phenotype and inhibiting their phagocytic capacity (Grégoire et al., 2020).
Tumor vessels are structurally and functionally abnormal. The tumor vascularization is highly disorganized and shows several anomalies which are responsible for functional defects. These structural abnormalities include endothelial cell hyperplasia, a decrease in the number of pericytes in contact with endothelial cells, and tortuous vessel organization (Li et al., 2021a), all factors leading to increased vascular permeability. Glioma-associated neovascularization (GAN) is a complex and regulated process and is highly dependent on the balance between five separate pathways: (i) vascular co-option and (ii) angiogenesis, followed by (iii) vasculogenesis and (iv) vascular mimicry and finally (v) GBM-endothelial cell trans-differentiation.
Vascular co-option was reported for the first time in 1999 and was described as the first process involved in the organization of tumor cells around normal tissue vasculature. Holash et al. (1999) was the first person to report vascular co-option in a rat model of glioma. Early tumors were well vascularized, and it took at least four weeks for an angiogenic response to be observed at the tumor’s edge. Winkler et al. (2009) discussed the invasive potential of glioma cells after being in close contact with the surrounding micro-vessels. Vascularization occurred via vascular co-option (Figure 7-A) but not angiogenesis. (Hardee and Zagzag, 2012).
Angiogenesis, a step following the vascular co-option, is known as new vessels developing from pre-existing micro-vessels. Angiogenesis processes were described in 1976 when Brem (1976) observed a high neovascularization process in GBM animal models. Hypoxic glioma cells around necrosis release proangiogenic factors, and other hypoxia independent mechanisms shift the angiogenic balance toward proangiogenic phenotype (Figure 7-B). GBM angiogenic phase is characterized by the formation of an irregular vascular network, with dilated and distorted arteries, abnormal branching, and shunts, contributing to abnormal perfusion. GBMs have immature vasculature with excessive leakage. GAN is driven by many key pathways identified (e.g., erythropoietin and their receptor, macrophages migration inhibitory factor, basic fibroblast growth factor, and placental growth factor) (Xue et al., 2017)
Vasculogenesis has been identified to include mobilization, differentiation, and recruitment of marrow-derived cells known as endothelial progenitor cells (Figure 7-C). Similarly, to the angiogenic process, vasculogenesis is induced by both hypoxia-dependent and independent mechanisms. The most well-known factors are the SDF-1 and CXCR4 pathways (Sun et al., 2019) Vascular mimicry characterizes tumor cells that organize themselves with ECM to mimic the structure of a vessel (Figure 7-D). Thus, the cells forming these structures do not express endothelial cell markers (CD31, CD34) but may show gene alterations specific to GBM cells e.g., EGFR amplification. This vascular mimicry appears to be connected to functional blood vessels and, although permeable, would increase nutrient delivery to the tumor. However, it is accepted that the neovessels formed exhibit altered structures and functionalities.
History and concept of immune system’s modulation
The relationship between immune functions and cancer cells was reported for the first time by Rudolf Virchow 150 years ago. He observed the presence of leukocytes within tumor tissue. Therefore, he suggested that the leukocyte infiltrate reflected that cancer’s origin lies in chronic inflammation (Balkwill and Mantovani, 2001). William Coley hypothesized the concept that our immune system can effectively recognize and eliminate cancer cells. He injected living or inactivated bacteria in the intra-tumor regions. The idea of Coley’s toxins generated several discussions between researchers and scientists. His hypothesis that activated phagocytes would kill both living bacteria and adjunctive tumor cells was accepted at that time, following evidence that injection of bacteria in the intratumor region led to cancer shrinkage. Although the concept showed an innovative idea regarding cancer treatment, the responses were heterogeneous, and the success rate was not promising (Carlson et al., 2020). Cancer is characterized by alterations in molecular pathways and cellular processes. These alterations result in diverse neoantigens presented by MHC class I on tumor cells’ surface. These complexes can be recognized by CD8+ T- lymphocytes in cancer patients.
Although cancer progression involves a variety of methods to overcome the host’s immunity, immunotherapy can restore and even improve the patient’s immune system. Many immunotherapeutic approaches have already shown efficacy in patients, while other new therapeutic approaches remain under development. Immune checkpoint blocking antibodies (ICBs) are currently under clinical investigation(Persico et al., 2021). To date, 1st of April 2021, 4 042 clinical trials of immunotherapy in all types of cancer are listed on ClinicalTrials.gov. One of the most attractive immunotherapy features is its ability to target cancer cells and thus spare healthy tissue. This characteristic differentiates immunotherapy from other « traditional » therapies such as radiation therapy and chemotherapy. The efficacy of immunotherapy was first demonstrated in the treatment of melanoma and renal cell carcinoma with high doses of IL-2 and is now spreading to other haematological and solid cancers (Ventola, 2017).
Antitumor immune response
Our knowledge of fundamental cellular and molecular mechanisms of the immune system’s innate and adaptive components has evolved. Cells from the innate system have receptors that can detect foreign microorganisms and dying cells. Macrophages and neutrophils provide early defense against microorganisms, while dendritic cells (DC) provide a linkage to the immune system’s adaptive components. Immunological reactions against a growing tumor require an integrated response between the innate and adaptive immune responses. Based on our current knowledge of immune responses, several distinct steps must be completed, endogenously or therapeutically, to produce an effective antitumor response (Figure 9). Oncogenesis processes in tumor cells generate neoantigens that start the initial step in antitumor immune response when DCs detect such neoantigens. Additionally, pro-inflammatory molecules, together with chemokines released by the tumor cells themselves, will recruit innate immune cells to this local source of « danger » (Pio et al., 2019). Initiation of the antitumor response occurs when innate immunity cells are alerted to the presence of a growing tumor. The following two steps occur when the captured antigens on MHC class I and MHC class II molecules are presented to T-lymphocytes by DCs, triggering the activation and the priming of effector T-lymphocytes against tumor-specific antigens. At this stage, the immune response is initiated, with the ratio of T effector lymphocytes to T-regulatory lymphocytes presenting a critical determinant in this response.
Monoclonal antibodies (MAbs) approved to be a significant strategy used in the treatment of solid tumors and hematological malignancies. They have a unique specificity for a specific antigen which allow them to bind to epitopes on the surface of cancer cells or immune cells. Therapeutic Abs are related to the immunoglobulin G family and composed of fragments that bind to their antigen. Furthermore, they are known as « naked, » as they are not conjugated with another active principle such as chemotherapy or radiotherapeutic agent. The primary mechanisms of action of the majority of naked mAbs are antibody-dependent cellular cytotoxicity and complement-dependent cytotoxicity. Other mechanisms are also reported, such as the direct triggering of cell death or the blocking of angiogenesis and cell survival signaling pathways. All-human MAbs show lower immunogenicity compared to murine, chimeric, or humanized mAbs.
Therapeutic MAbs with a non-human sequence is more easily identified as foreign subjects and induce host immune responses. Reduced efficacy was observed in non-human MAbs mainly due to increased clearance and more adverse reactions, such as injection site reactions. The use of naked MAbs has significantly improved the treatment of certain solid tumors (Zahavi and Weiner, 2020).
Monoclonal antibodies against checkpoint proteins
MAbs that block immune checkpoints represent up-and-coming treatments for various cancer types as they have remarkable and long-lasting responses in some patients. Unlike chemotherapies, MAbs are well tolerated and provide long-term benefits on patient survival. A notable example is pembrolizumab’s success, an anti-PD-1 antibody combined with surgery and radiation therapy have eradicated all melanoma traces in former President Jimmy Carter. The mechanism of action of ICBs was a breakthrough in the conception of cancer treatment and led for a Nobel Prize in Physiology (Huang and Chang, 2019). Chemotherapy and radiation therapy are directed to destroy cancer cells, while ICBs target the tumor-induced immunosuppression. These MAbs block checkpoint proteins on the surface of T-lymphocytes that are responsible for the immune response, resulting in prolonged antitumor responses (Desland and Hormigo, 2020).
Immunomodulatory antibodies can prevent checkpoint ligand/receptor interactions. They bind either to immune checkpoint proteins on T-lymphocytes, such as : (i) cytotoxic T-lymphocyte-associated antigen–4 (CTLA-4) and its ligands CD80/CD86 or (ii) PD-1 and its ligands programmed cell death ligand 1 (PD-L1). These ICBs have demonstrated clinical efficacy, but many other ICBs have been identified and under developments (Figure 11). Stimulation of the immune system with ICBs i.e., Ipilimumab, anti-CTLA-4, and atezolizumab, anti-PD-L1, showed promising effects alone or with other chemotherapies on treating multiple cancers. Ipilimumab was the first humanized anti-CTLA-4 approved by the American federal drug administration (FDA) to treat inoperable melanoma (Tarhini, 2013). Five years later, atezolizumab was the first humanized anti-PDL1 to be approved by the FDA to treat advanced or metastatic urothelial carcinoma (Hsu et al., 2017). However, the combination of nivolumab and ipilimumab with GBM in clinical trials ended with immune-related severe adverse effects and avelumab monotherapy, anti-PD-L1 has a small effect on progression-free survival. In preclinical settings, ICBs efficacy was enhanced when antibodies were delivered to brain tumors. (Guo et al., 2020). PD-L1 proteins are expressed as surface molecules by cancerous cells as GBM cells (Hao et al., 2020) and provide a tumor escape mechanism when binds to PD-1 proteins at the surface of activated T-lymphocytes leading to T lymphocytes exhaustion (Azoury et al., 2015).
Table of contents :
LIST OF TABLES
LIST OF FIGURES
1. CLASSIFICATION OF PRIMARY BRAIN TUMORS IN ADULTS
1.1 Primary brain tumors in adults
1.2 Glioblastoma IDH wildtype
1.2.2 Clinical presentation and diagnosis of GBM
1.2.3 Standard treatment of GBM
1.2.4 Prognostic factors in GBM patients
2. GBM CELL BIOLOGY AND THEIR MICROENVIRONMENT
2.1 GBM cells and heterogeneity
2.1.1 Tumor cells origin, and intratumor cell heterogeneity
2.1.2 Four signaling pathways are disrupted in GBM
2.2 Tumor cell microenvironment
2.2.1 Niches and stem cells of GBM
2.2.2 Cellular components of TME
2.2.3 Glioma-associated neovascularization
2.2.4 The blood-brain barrier
3. THERAPEUTIC STRATEGIES TO MODULATE THE TUMOR MICROENVIRONMENT
3.2 Modulation of the immune system
3.2.1 History and concept of immune system’s modulation
3.2.2 Antitumor immune response
3.2.3 Tumor escape mechanisms
3.2.5 Monoclonal antibodies
3.2.6 Monoclonal antibodies against checkpoint proteins
3.3 Overcoming, disrupting, or bypassing the BBB
3.3.1 The BBB limits drug penetration into normal and tumor tissue
3.3.2 Role of ultrasound-mediated BBB opening
3.3.3 UMBO and antibodies delivery to the brain
3.3.4 Ultrasound mediated BBB opening in clinical settings
Role of Multi-Drug Resistance in Glioblastoma Chemoresistance: Focus on ABC Transporters
CD80 and CD86: expression and prognostic value in newly diagnosed glioblastoma
Temporary blood-brain barrier disruption by low-intensity pulsed ultrasound increases anti-PD-L1 delivery and efficacy in glioblastoma mouse models
GENERAL DISCUSSION 152 GENERAL CONCLUSION