Exploring glioblastoma stem cell heterogeneity: Immune microenvironment modulation and therapeutic opportunities

Abstract

Despite its growing use in cancer treatment, immunotherapy has been virtually ineffective in clinical trials for gliomas. The inherently cold tumor immune microenvironment (TIME) in gliomas, characterized by a high ratio of pro-tumor to anti-tumor immune cell infiltrates, acts as a seemingly insurmountable barrier to immunotherapy. Glioma stem cells (GSCs) within these tumors are key contributors to this cold TIME, often functioning indirectly through activation and recruitment of pro-tumor immune cell types. Furthermore, drivers of GSC plasticity and heterogeneity (e.g., reprogramming transcription factors, epigenetic modifications) are associated with induction of immunosuppressive cell states. Recent studies have identified GSC-intrinsic mechanisms, including functional mimicry of immune suppressive cell types, as key determinants of anti-tumor immune escape. In this review, we cover recent advancements in our understanding of GSC-intrinsic mechanisms that modulate GSC-TIME interactions and discuss cutting-edge techniques and bioinformatics platforms available to study immune modulation at high cellular resolution with exploration of both malignant (i.e., GSC) and non-malignant (i.e., immune) cell fractions. Finally, we provide insight into the therapeutic opportunities for targeting immunomodulatory GSC-intrinsic mechanisms to potentiate immunotherapy response in gliomas.

1 Glioblastoma stem cells

High-grade gliomas, including glioblastoma (GBM), are highly heterogeneous with a complex oncogenic microenvironment consisting of distinct tumor niches and remarkable cell heterogeneity (1, 2). A critical component of glioma malignancy derives from the distinct population of glioma stem cells (GSCs) that function to promote and maintain oncogenicity through their capacity for self-renewal, cellular adaptation, and multipotency (3–5). These stem-like cells engage in a synergistic relationship with the surrounding tumor microenvironment (TME) to promote tumor progression and contribute to the vast degree of intratumoral heterogeneity, immune-suppression, and therapy resistance encountered in gliomas (5). The plasticity of GSCs facilitates shifts between distinct tumorigenic stem-like states and allows for transitions along the spectrum of differentiation that characterizes glioma cells (6–8). These state transitions are brought on by various stemness-inducing stimuli (e.g., hypoxia, interaction with non-neoplastic cells, immune exposure, therapeutics, etc.) and carried out by reprogramming mechanisms that involve specific transcription factors and dynamic epigenetic modifications that alter transcriptional profiles and consequently, cellular and molecular phenotypes (9) (Figure 1).

Figure 1

A unique facet of cancer stem cell biology is the capacity of a tumor-propagating stem-like cancer cells to evolve from a non-stem-like state through a process known as dedifferentiation. Reprogramming transcription factors, collectively referred to as the Yamanaka factors (Oct3/4, Sox2, Klf4, c-Myc), play critical driving roles in this process (10, 11). In particular, Oct4 and Sox2 are sufficient to induce stem cell properties and in vivo tumor-propagating potential in differentiated and non-tumor-propagating GBM cells (12, 13). Additionally, our lab has identified multiple mechanisms downstream of Oct4 and Sox2 underlying GSC stemness and tumorigenicity (12–17). Furthermore, epigenetic modifications mediated by Sox2 and Oct4 have implications in GSC-immune interactions that contribute to an immunosuppressive TME (17).

In spite of the progress in our understanding of GSC biology, the inherent heterogeneity and plasticity of the GSC population and resulting phenotypic consequences have restricted the impact of ongoing therapeutic efforts. In this review, we cover the advancements in our understanding of GSC plasticity, GSC-intrinsic immunomodulatory mechanisms, and the capacity of GSCs to co-opt immunosuppressive cell phenotypes. Additionally, we discuss the application of evaluable and emerging cutting-edge techniques to study immunosuppressive GSC states and cell-cell interactions at high cellular resolution. Finally, we discuss potential therapeutic approaches aimed at exploiting epigenetic mechanisms and metabolic vulnerabilities associated with acquired immunosuppressive GSC states to enhance the efficacy of GBM immunotherapy.

2 Contribution of GSCs to the immunosuppressive microenvironment in glioblastoma

2.1 Immunosuppressive TME in GBM

A hallmark of GBM and significant barrier to immunotherapies is the immunosuppressive TME defined by relatively high numbers of suppressive, pro-tumor immune cell infiltrates (e.g., regulatory T cells (Tregs), tumor-associated microglia, tumor-associated macrophages (TAMs), myeloid-derived suppressor cells (MDSCs)) and high prevalence of dysfunctional T cell states (e.g., exhaustion) (18–26). High infiltration of immunosuppressive myeloid and lymphoid cell populations negatively correlates with patient prognosis and therapy response in GBM (19, 25, 27–29). Notably, most infiltrating immune cells in GBM are of myeloid origin (e.g., monocyte-derived macrophages, brain-resident microglia, and MDSCs) and are defined by a variety of pro-tumor phenotypes that function to suppress the anti-tumor immune response (20, 24, 26, 30–32). Beyond their direct immune suppressive functions, pro-tumor myeloid populations (i.e., TAMs, microglia, and MDSCs) promote GBM growth, invasion, and angiogenesis (33–39). Moreover, TAMs have the capacity to induce a mesenchymal state in GBM, which is associated with therapy resistance and poor patient survival (40–42).

As mentioned above, GBM is highly heterogeneous and can be broadly classified by three molecularly defined subtypes – classical, proneural, or mesenchymal (43) – each with unique immunosuppressive characteristics. The diversity of myeloid cell states is demonstrated by the existence of myeloid phenotypes that preferentially associate with specific GBM subtypes and specific tumor niches. For example, mesenchymal tumors are enriched in blood-derived macrophages that possess a transcriptionally distinct immunosuppressive profile marked by upregulation of genes involved in chemokine signaling and lymphocyte chemotaxis and reside in microvascular and peri-necrotic regions (31).

GSCs also avoid recognition by the immune system through expression of checkpoint inhibitors, including PD-L1, CD70, A2aR and TDO, and downregulation of antigen-presentation molecules, specifically MHC class I (MHC-1) molecules (17, 44–46). Furthermore, GSCs can regulate the immune TME by recruiting and polarizing myeloid cells to an immunosuppressive state, producing T cell-suppressing cytokines, and monopolizing nutrients necessary for proper T cell function (47–51). For example, Wu et al. demonstrated GSC-mediated recruitment and polarization of macrophages and microglia to an M2-like phenotype through production of TGF-β1, MIC-1 and CSF-1 (47). From a metabolic standpoint, GSCs preferentially take up glucose from the microenvironment, leading to impaired T cell function due to glucose deprivation (49–51).

There are multiple contributors to the lack of immunotherapy efficacy in GBM. These include a relatively low gene mutation rate resulting in relatively few tumor-associated neoantigens (52), a unique TME comprised of an abundance of tumor-associated immunosuppressive M2-like macrophages (comprising up to 60% of immune cells) (20) and the immune-privileged brain environment (53). However, the fact that standard-of-care chemo/radiation for newly diagnosed GBM can lead to hypermutated recurrent tumors (54) unresponsive to immunotherapy and the efficacy of immunotherapy in cases of CNS metastases (e.g., melanoma) emphasize the potential dominant role for the GBM TME in driving immunotherapy resistance (55–58). For example, the TME of brain metastases and primary CNS tumors, including GBM, have distinct immune landscapes whereby brain metastases contain higher infiltration of leukocytes and fewer cells of monocytic-lineage (19).

2.2 Molecular regulation of the immunosuppressive GSC transcriptome

Understanding the therapeutic consequences and vulnerabilities of GBM cell heterogeneity and plasticity is an ongoing effort in the scientific community. Despite the association between genetic aberrations and cell states in GBM (59), genomic alterations alone are insufficient to explain the highly adaptive and heterogeneous nature of GSCs (60). Epigenetic modifications mediate the acquisition of a stem-like state in cancer cells, including GBM, and recent studies have implicated epigenetic variability as a driving force in the adaptation of GBM to external stimuli throughout tumor evolution (4, 12, 14–16, 60–62).

Epigenetic mechanisms, including the addition and removal of activating or repressive marks to histone proteins, DNA, and RNA, serve to regulate gene expression in a reversible manner (62, 63). Notably, the balance between transcriptionally repressive methylation and activating demethylation of DNA, controlled by DNA methyltransferases (DNMTs) and ten-eleven translocation proteins (TETs), respectively, governs acquisition and maintenance of the GSC phenotype (12, 16). On the other hand, RNA methylation occurs post-transcriptionally to alter mRNA stability and impact stem cell differentiation (64–66). Histone proteins, which are essential for forming and stabilizing chromatin structure in either accessible (euchromatin) or inaccessible (heterochromatin) states, are also susceptible to modifications (e.g., methylation, acetylation, phosphorylation) which consequently alter chromatin architecture and lead to activation or repression of gene transcription (63). The progression from chromatin remodeling to transcription activation involves the activity of epigenetic readers, such as bromodomain and extraterminal domain (BET) proteins, that detect and bind histone marks. Readers recruit transcription regulatory factors and/or chromatin remodeling protein complexes to control gene expression (67). Collectively, dysregulation of epigenetic factors underlies the acquisition, maintenance, and plasticity of the GSC phenotype and represents a promising avenue for GSC-targeted therapeutics in GBM.

A relatively underappreciated aspect of dynamic GSC states is their contribution to the immunosuppressive tumor microenvironment. The role of the tumor microenvironment, specifically the immune compartment, in driving GBM and GSC states is being increasingly elucidated (7, 31, 41, 68, 69). However, improved understanding of how GSC adaptations influence the immune TME is crucial to overcome the immunosuppressive pressures that diminish the efficacy of immunotherapy in gliomas. As discussed above, the cell-intrinsic factors that modify and drive GSC states and transitions are abundant and include specialized reprogramming transcription factors, epigenetic modifiers, transcriptional regulators, cellular cycles, etc. In this section, we discuss findings that highlight the effects of GSC-intrinsic transcriptional variation and cellular transitions on their immunomodulatory role.

2.2.1 Transcription factors

As mentioned previously, the most notable transcription factors associated with acquisition of a stem-like phenotype are the Yamanaka factors. Of particular interest in the context of glioma, especially related to the immune influence of GSCs, are Sox2 and Oct4. Sox2, which is a ubiquitously expressed transcription factor in cancer stem cells, has a direct effect on the immunosuppressive capacity of GSCs through induction of CD39 (ENTPD1), an ectonucleotidase responsible for hydrolyzing ATP towards adenosine and a critical mediator of immune response in cancer (70, 71). In GSCs, knockdown of Sox2 increased the extracellular ATP concentration and enhanced dendritic cell recruitment and phagocytic-ability as well as T-cell-mediated GSC lysis in co-cultures in direct association with a reduction in CD39 (70).

Another noteworthy stem cell-driving transcription factor, Oct4 (POU5F1), cooperates with Sox2 to induce an immunosuppressive transcriptome in GSCs, defined by induction of immune checkpoint inhibitory molecules (PD-L1, CD70, A2aR and TDO) alongside dysregulation of immune modulatory cytokines and chemokines including upregulation of SSP1, IL8, CXCL3, and CCL20 and downregulation of CCL5, CXCL9, and CXCL10. This GSC immune-suppressive phenotype was found to be mediated by and dependent on the BET protein BRD4 which is involved in directing chromatin remodeling in response to histone modifications (17).

Aside from the classical cancer stem cell reprogramming transcription factors, other transcription factors are involved in both the induction of a glioma stem-like state and acquisition of an immunosuppressive GSC phenotype. Yin Yang 1 (YY1), a zinc-finger transcription factor involved in polycomb protein recruitment and transcription regulation (72, 73), is necessary for maintenance of the stem cell phenotype in GBM, is associated with Sox2 and Oct4 expression across cancers (74–77), and mediates both chemotherapy and radiation resistance in GBM (74, 78). The YY1-CDK9 transcriptional complex in GBM cells promotes Treg infiltration, inhibits RNA methylation-dependent interferon responses, and reduces the efficacy of immune checkpoint inhibitor therapy in GBM (77), highlighting how transcription factors and epigenetic modifications cooperate to impart immunomodulatory function in GBM cells.

While we typically think of external pressures when discussing induction of stemness, pre-programmed endogenous processes such as the circadian rhythm may also drive and maintain the stem cell phenotype in cancer cells (79, 80). The circadian rhythm is a cell-autonomous, cyclical process composed of transcriptional-translational feedback loops that govern carefully timed adjustments in gene expression (81). The transcriptional variation attributed to circadian oscillations has a demonstrated capacity to alter cancer stem cell states, their interaction with the TME, and their susceptibility to therapeutic agents (79, 82–86). Disruption of the circadian rhythm leads to altered immune cell infiltration and T-cell exhaustion in a variety of cancers (85, 87, 88). In glioma, the CLOCK-BMAL1 complex, the main transcriptional component in the circadian rhythm, modulates the stem cell phenotype by enhancing self-renewal capacity and migration and regulating GSC metabolism (89–91). Furthermore, CLOCK, in partnership with BMAL1, upregulates LGMN via induction of OLFML3, a novel microglia-recruiting factor, in GSCs to enhance infiltration of microglia and polarize them to an immunosuppressive phenotype (89, 92). This finding, along with our understanding of circadian dysfunction in cancer, demonstrates the possibility of dynamic immunomodulatory GSC phenotypes mediated by cell-intrinsic processes.

2.2.2 Histone modifications

As discussed previously, readers of histone modifications, e.g., BRDs, can direct acquisition of an immunosuppressive phenotype in GSCs (17). BRDs also have an immunomodulatory role in ATRX-deficient, IDH-mutant gliomas. Deficiency of ATRX, a histone protein involved in the chromatin remodeling SWI/SNF family, occurs in >80% of IDH-mutant grade II/III astrocytomas and >50% of secondary GBMs (93, 94). Hu et al. reports a BRD3/4-dependent induction of an immunosuppressive transcriptome, marked by upregulation of PD-L1/2, IL-33, CXCL8/9, CSF2, and IL-6, in ATRX-deficient glioma cells (95). Furthermore, this ATRX loss resulted in decreased T-cell-mediated glioma cell lysis, increased macrophage immune-suppressive M2-like polarization, and enhanced Treg tumor infiltration. Independently, Babikir et al. found increased infiltration of immunosuppressive monocyte-lineage cells in ATRX-mutant versus ATRX-wildtype IDH-mutant glioma (96). While the former findings were not in the context of cancer stem cells, ATRX-mutant IDH1-mutant GSCs have been shown to possess increased cell propagation capacity and upregulation of TGF-beta-associated pathways compared to ATRX-wildtype GSCs (97) suggesting that ATRX-deficiency enhances certain stemness characteristics in glioma in accordance with increased immunosuppressive effects.

Enzymes that remove histone acetylation marks, known collectively as histone deacetylases (HDACs), also regulate GSC-intrinsic immunomodulatory mechanisms (44, 98). A major factor in the poor immunogenicity of gliomas derives from the innate ability of GSCs to evade the immune response through dysregulation of major histocompatibility complex (MHC) molecule expression (44). Specifically, Yang et al. demonstrated HDAC-dependent MHC-1 downregulation in GSCs that consequently suppressed the function of tumor-infiltrating lymphocytes (TILs). Furthermore, pan-HDAC inhibition not only enhanced MHC-1 expression and the T-cell response, but it downregulated Sox2 and Oct4 protein expression and decreased the self-renewal capacity of GSCs. While the knowledge that HDAC inhibitors upregulate MHC-1 molecule expression in cancer is not new (99–103), this study extended the findings to gliomas and, more importantly, detailed the association between HDAC regulation of MHC1 expression and maintenance of the stem cell phenotype. Aside from MHC-1 expression, histone deacetylation via HDAC8 in glioma cells represses NKG2D ligand expression to evade natural killer (NK) cell-mediated tumor cell death (104). A study by Zhan et al. found that HDAC1 and HDAC2 promote GSC evasion of the suppressive interferon-mediated immune response through their involvement in the nucleosome remodeling and deacetylase (NuRD) complex (105). Zhan and colleagues show that methyl-CpG-binding domain 3 (MBD3) recruits and assembles the HDAC-containing NuRD complex at the promoter of STAT1 where deacetylation of H3K27 leads to repression of STAT1 transcription. From there, STAT1 repression desensitizes GSCs to interferon treatment. The driver of this mechanism, MBD3, also regulates proliferation, viability, and self-renewal capacity of GSCs and positively correlates with expression of Sox2, Olig2, and Nestin in GBM specimens (106). Together, these findings directly link glioma cell-intrinsic HDAC-mediated immunosuppression to the acquisition and maintenance of the stemness phenotype of GSCs.

2.2.3 DNA and RNA methylation

The dynamics and consequences of DNA methylation in cancer biology have been vastly explored. Patterns of DNA methylation mediate repression of gene transcription thereby altering cell states and phenotypes and contributing to tumor progression (12, 16, 60). In the context of gliomas, levels of DNA methylation are associated with the IDH-mutation status of the tumor with IDH-mutant gliomas tending to acquire a hypermethylated phenotype, commonly referred to as the glioma CpG island methylator phenotype or G-CIMP (107). This CpG island hypermethylation has been implicated in controlling immunomodulatory effects of GSCs in both IDH-mutant and IDH-wildtype gliomas. Specifically, in IDH-mutant GSCs, increased methylation of the NKG2D ligand gene promoter suppresses its expression leading to the ability of GSCs to evade NK cell-mediated cytotoxicity (108). In line with evidence that decreasing DNA methylation in cancers enhances immunogenicity through upregulation of MHC-1 expression (109–112), GSCs are found to contain increased methylation along the regulatory regions of HLA genes (113). Similarly, a separate study discovered that methylation-dependent regulation of MHC-1 and ICAM1 expression regulates T cell-mediated GSC cytotoxicity in IDH-wildtype glioma (46). Gangoso et al. demonstrated emergence of a DNA methylation-dependent immune evasive phenotype highly associated with mesenchymal and mesenchymal-like GBM states (68). These GSCs were found to have extensive hypomethylation, especially along genes involved in immune-suppressive processes. Furthermore, immune evasive GSCs fostered an immunosuppressive TME in vivo characterized by enhanced infiltration of macrophages and myeloid-derived suppressor cells (MDSCs) and increased TIL dysfunction (68).

Unlike DNA methylation, RNA methylation is a relatively underappreciated regulator of the stem cell phenotype. In embryonic stem cells (ESCs), low levels of N (6)-methyladenosine (m6A methylation) promote pluripotency and protect against differentiation (64–66). Many transcripts of pluripotency genes, including Nanog, Sox2 and Klf2, are regulated by m6A methylation in both human and mouse ESCs (64, 65). In GSCs, Cui et al. found that METTL3, a catalytic subunit of the m6A methyltransferase complex, reduces cell proliferation and self-renewal capacity along with CD44 expression, through methylation of pluripotency-driving transcripts (114). However, conflicting roles of METTL3 in GSCs have been demonstrated by other studies (115–117). Visvanathan et al. showed that GSCs are enriched for METTL3 which functions to stabilized SOX2 expression and enhance self-renewal capacity (117). Moreover, METTL3 has been linked to TMZ resistance in GBM cells (115, 116). On the other hand, ALKBH5, an m6A RNA demethylase, is highly expressed in GSCs and associated with poor patient prognosis in GBM (118). Through demethylation, ALKBH5 enhances both self-renewal capacity and expression of stemness genes (i.e., Nestin, Sox2, Nanog, and Oct4) in GSCs (118). Furthermore, ALKBH5 can indirectly regulate the GBM cell-intrinsic immunosuppressive phenotype by demethylating the lncRNA NEAT1 which allows for NEAT1-mediated paraspeckle assembly (119). From there, relocation of transcriptional repressors to the newly assembled paraspeckles results in upregulation of CXCL8 and IL8. Consequently, ALKBH5-initiated CXCL8/IL8 induction improves tumor-associated macrophage infiltration and promotes tumor progression, demonstrating the impact of post-transcriptional epigenetic modifications in regulating the immunosuppressive GSC phenotype. These findings suggest the need for a more refined investigation of the role of m6A methylation in controlling the GSC phenotype, taking special care to account for and address the heterogeneous and dynamic states of GSCs.

2.3 Immunosuppressive influence of metabolic plasticity in GSCs

A hallmark of cancer, metabolic remodeling permits cells to adapt and persist in conditions of depleted oxygen and/or nutrients by utilizing alternative metabolic pathways (120). Cancer stem cells in particular possess a great potential for metabolic reprogramming compared to more differentiated counterparts (121–123). In GBM, GSCs readily adjust their metabolic preferences to meet their energy demands (49, 123–125). Remodeling in response to hypoxic or low glucose environments occurs through various mechanisms including epigenetic modification (3, 126, 127), HIF-1α signaling (125, 128, 129), and altered expression of metabolic enzymes and transporters (49, 130). In general, metabolic profiles in GBM cells are spatially distinct and attributed to TME interactions and restricted nutrient availability (122, 125, 131, 132) with different GSC subsets having distinct metabolic preferences and dependencies (121, 133, 134).

Notably, metabolic reprogramming in GSCs modulates the anti-tumor immune response in favor of tumor progression. As mentioned previously, GSCs suppress T cell function and recruit and polarize microglia through glucose monopolization and circadian dysregulation, respectively (49, 89, 90). Additionally, GBM cells in spatial proximity to immune cells have distinct metabolic profiles associated with immune-suppressive transcriptional signatures. Similar to immune-responsive cells, hypoxia-responsive GBM cells, which are enriched for GSCs, reside in regions enriched in TAMs and T cells (including exhausted CD8⁺ T cells), linking hypoxia-induced metabolic remodeling to tumor-promoting immune interactions (135). Furthermore, Coy et al. identified increased levels of extracellular immune-suppressive adenosine in GBM attributed to high expression of CD73, which functions alongside CD39 to metabolize ATP into adenosine, in hypoxia-responsive GBM cells (131). Collectively, these findings demonstrate a connection between TME-driven metabolic reprogramming and GSC-mediated immune suppression, and prompt further investigation into therapeutically relevant metabolic vulnerabilities in these immune-suppressive cells.

2.4 GSC plasticity and immune cell mimicry

Over the last decade or so, our understanding of the phenotypic plasticity of cancer cells has expanded to include the ability of cancer stem cells to behave like, or mimic, the function of other cells (Figure 2) (136). For example, GSCs can behave like vascular endothelial cells and pericytes to promote angiogenesis and increase oxygen and nutrient availability in a phenomenon referred to as vasculogenic mimicry (137–139). In many respects, this epitomizes the inherent plasticity of GSCs and is mediated by epigenetic mechanisms and influenced by conditions in the TME (136–141). Additionally, subsets of GSCs resemble neuronal and glial progenitor cells and engage in synaptic interactions to promote tumor invasion and progression (59, 142–146). Notably, a similar process affecting the GBM immune microenvironment has been identified whereby GSCs mimic immune cell function by co-opting transcriptional profiles typically associated with myeloid cells (68, 147). The aforementioned study by Gangoso et al. demonstrated acquisition of a myeloid-related immunosuppressive transcriptional profile in GSCs mediated by a DNA methylation-dependent immunoediting mechanism in response to repeated immune exposure (68). In particular, Irf8, Nt5e, and Cd274 were among the genes upregulated in immune evasive, mesenchymal-like GSCs due to DNA demethylation. Due to the often myeloid-specific expression of Irf8 and transcriptome-level induction of myeloid-related signatures, they concluded that altered DNA methylation governs acquisition of a myeloid-mimicking phenotype in GSCs (68). Furthermore, our lab has identified a Treg-like transcriptional profile in TGFBR2^high GSCs driven by Oct4 and Sox2 that regulates the immunosuppressive TME by repressing T cell function through the action of effector genes canonically expressed by Tregs (e.g., CD274, ENTPD1, NT5E, LGALS1, TGFB1) (148). Collectively, these findings suggest that GSCs co-opt functions of non-neoplastic cells from the TME through dynamic state transitions to facilitate tumor progression and suppress the immune response.

Figure 2

In summary, GSCs alter their immunomodulatory phenotype in response to stemness-regulating stimuli (Figure 1). A variety of gene expression regulators carry out these transformations, from pluripotency-driving transcription factors and circadian transcriptional complexes to epigenetic modifications (Table 1). While improving, our understanding of the immunosuppressive functions of GSCs as they relate to their heterogeneous and ever-changing states is limited. Unfortunately, many studies conducted in the context of GSC-immune interactions are restricted from the start to certain subpopulations of GSCs (e.g., gatekeeping GSCs by their CD133 expression). This not only sets the stage for potentially context-dependent findings, but inaccurately represents the diverse spectrum of GSC phenotypes found in patient specimens. Moving forward, it is essential to approach studies of GSC-intrinsic mechanisms from an unbiased starting point, especially in the context of TME interactions whereby both cell intrinsic and extrinsic influences constantly drive GSC adaptation and state plasticity. In the next section, we discuss cutting-edge techniques and bioinformatics tools available to study heterogeneous and dynamic GSC states at a high cellular resolution allowing for deep exploration of the bidirectional interaction between GSCs and cells of the TME that influence GSC-mediated immunomodulation (Table 2).

3 Analytical tools to explore GBM heterogeneity

3.1 Deconvolution of bulk tumor sequencing

Whole tumor sequencing modalities, such as whole-genome and whole-exome sequencing, RNA-sequencing (RNA-seq), bisulfite sequencing, and assays for chromatin accessibility sequencing (ATAC-seq), have been widely used to classify tumors and explore context-specific cancer biology. However, these sequencing analyses conducted on dissociated bulk tumor specimens output an average value across all cells, placing limitations on the ability to distinguish which cell type or compartment is contributing to said output and, therefore, interfering with exploration of heterogeneous cell populations and phenotypes. Computational tools to deconvolute bulk-sequencing (bulk-seq) data have been developed, allowing one to take full advantage of large, pre-existing datasets of bulk-seq patient tumor specimens. These tools present an opportunity to infer cellular-level data from a vast number of patient tumors when the use of more advanced technologies is unfeasible or cost-prohibitive.

Computational deconvolution of bulk RNA-seq data has garnered interest in the realm of cancer immunology by granting insight into the composition of the immune compartment (e.g., cell type proportions and overall extent of immune infiltration) and transcriptional phenotypes within immune cell types based on associated expression of cell-specific markers (149–156). The degree of immune cell infiltration and tumor purity (i.e., proportion of tumor composed of neoplastic cells), calculated through ESTIMATE (149), have been associated with molecular phenotypes and trends in patient survival, with high immune cell infiltration and low tumor purity corresponding to mesenchymal tumors and shorter survival (193, 194). More in-depth analytic tools, such as CIBERSORTx (150), xCell (151), MCP-Counter (152), EPIC (153, 154), quanTIseq (155) and TIMER (156), go beyond the extent of immune infiltration and tumor purity to deduce absolute and relative cell proportions based on pre-set or user-defined cell signatures. The CIBERSORTx algorithm takes it a step further by allowing users to impute gene expression profiles (GEPs) for the estimated cell populations. However, this additional step is not recommended for low abundance cell types, such as tumor-infiltrating lymphocytes in GBM, due to insufficient statistical power. In populations with sufficient abundance, imputed GEPs can define cell-type specific transcriptional phenotypes in immune cells (e.g., M1/M2 macrophages, functional or exhausted T cells, activated natural killer or dendritic cells, etc.) and permit further exploration into cell-cell interactions based on ligand/receptor expression within respective cell types.

The value of bulk deconvolution algorithms is ultimately dependent upon two factors: (i) use of a validated, context-specific cell signature matrix and (ii) supportive findings through in vitro and/or in vivo experimentation. Varn et al. applied the CIBERSORTx platform to construct and validate a glioma cell signature matrix encompassing both neoplastic and non-neoplastic cell types derived from single-cell RNA-seq of IDH-wildtype and IDH-mutant patient gliomas (31). These efforts defined the temporal transitions in the composition of neoplastic cell populations and the associated changes in infiltrating immune cell types in newly diagnosed and recurrent tumors (31, 195). Moreover, a second CIBERSORTx cell signature matrix was constructed to infer proportions of GBM histological features within tumors by using gene expression profiles specific to Ivy GAP-defined histological features (e.g., leading-edge, infiltrating tumor, cellular tumor, microvascular proliferation, and pseudopalisading cells around necrosis) (196). After pathologist-based validation of the inferred proportions, Varn and colleagues were able to calculate correlations between cellular states and histological features. Together, utilization of CIBERSORTx software to estimate proportions of distinct cellular states and phenotypes and the extent of hallmark histological features (196, 197) allowed for comprehensive analysis of tumor cell interactions with and in relationship to the TME in a computational manner, setting the stage for downstream investigation of potentially therapeutically actionable cell states and mechanisms.

While tools for deconvolution of bulk sequencing still have their place in cancer research, especially due to their low cost and accessibility and the large volumes of publicly available data from patient tumors, the advent of cutting-edge single-cell sequencing technologies have turned what was once purely computational predictions into actual omics information at cellular-resolution.

3.2 Single-cell tools to explore glioma-immune cell interactions

The development of single-cell technologies to explore cellular states at the genomic, epigenomic, transcriptomic, proteomic, and metabolomic levels has led to unprecedented understanding of the inner workings of cancer (198–200). The resolution of single-cell sequencing allows for in-depth evaluation of rare cancer cell populations (e.g., GSC subsets and tumor-infiltrating immune cell subsets) within tumors that are otherwise lost in bulk analyses. The growing repertoire of single-cell technologies allows us to explore not only the constantly adapting cancer cell population but also the non-neoplastic cells in the surrounding TME, providing insight into the fluctuating & bidirectional communication between cell compartments.

In glioma, single-cell RNA-seq (scRNA-seq) analyses have broadened our knowledge of dynamic cellular and molecular phenotypes that represent the heterogeneous nature of gliomas, confer therapy resistance, and drive tumor recurrence in patients (1, 59, 134). Previously, GBMs were classified by their molecular subtype as determined by bulk sequencing of tumors (43). However, classification of tumor subtypes based on bulk transcriptome analyses masks the inherently high degree of intra-tumoral heterogeneity in GBM, demonstrated in early single-cell analyses by the representation of multiple molecular subtypes within individual tumors (1). Since then, scRNA-seq analyses conducted in patient tumors have defined distinct cellular states that vary in respect to their gene expression profiles (59, 60, 201), interaction with the TME (31, 41, 59, 68), and metabolic states (59, 135). Notably, Neftel et al. demonstrated that glioma cells can exist in and transition between four cellular states, oligodendrocyte-progenitor-like (OPC-like), neural progenitor-like (NPC-like), astrocyte-like (AC-like), and mesenchymal-like (MES-like), that differentially express genes involved in cell cycle, development, metabolism, and immune response, and in which all but one state (MES-like) resemble normal neurodevelopmental cell types (59). Cells in the MES-like state, which expresses higher levels of hypoxia-response and glycolytic genes relative to the other states, form synergistic relationships with TAMs to promote their respective mesenchymal-like, immunosuppressive phenotypes, demonstrated by Hara et al. (41).

Aside from scRNA-seq, one of the more commonly used single-cell techniques in the study of gliomas is cytometry by time-of-flight (CyTOF) (159, 160). CyTOF provides valuable information on cell proportions based on surface marker expression and has been used to characterize variation in immune cell landscapes between primary and metastatic CNS cancers (19), newly diagnosed and recurrent GBMs (20), and immune checkpoint inhibitor (ICI) refractory and responsive cancers and GBM mouse models (202). For example, Fu et al. identified a significant decrease in the proportion of tumor-promoting macrophages and microglia with recurrence whereas the frequencies of T cells, B cells, and NK cells were relatively unchanged (20). In another study, Simonds and colleagues compared immune cell proportions between ICI-refractory (GBM and sarcoma) and ICI-sensitive (renal cell carcinoma) cancers and found decreased T cells and DCs, but increased TAMs in ICI-refractory tumors (202). This aligns with the previously discussed finding that brain metastases, which are more sensitive to immunotherapy than GBM, have relatively more leukocytes and less myeloid cells in their TME, also assessed through CyTOF analysis (19).

Additional insight into cancer-immune interactions from the perspective of infiltrating immune cells can be obtained through techniques like CITE-seq (164, 165) and T cell receptor (TCR) sequencing (161) that provide information on surface marker-defined cell expression profiles and TCR variability and clonality, respectively. The use of CITE-seq, which simultaneously quantifies epitope-specific cell types and their associated transcriptomes via RNA sequencing, in glioma has identified novel markers for various transcriptionally defined subsets of TAMs with demonstrated susceptibility to macrophage-directed therapeutics (203). This exploration of the TAM population also unveiled a shift from microglia-derived to monocyte-derived TAMs in the TME with tumor recurrence (203). Notably, high infiltration of monocytic macrophages relative to resident microglia corresponds to increased malignancy in gliomas (19, 204). As for the tumor-infiltrating lymphoid population, TCR sequencing of newly diagnosed and recurrent GBMs revealed a diminished TCR repertoire with recurrence (22), which leads to suppressed T cell-mediated immune response via altered antigen-recognition machinery. Based on our understanding of how dynamic GSC states differentially express MHC molecules (44, 46, 113) and the fundamental knowledge that TCR repertoires are shaped in response to antigen exposure, one could infer that immunoediting mechanisms within GSCs directly alter T cell clonality during tumor progression, adding another facet to the immune evasive nature of glioma cells. However, this has yet to be conclusively demonstrated in the context of gliomas. Alternatively, a restricted TCR repertoire could signify T cell specification against the most actionable tumor antigens; however, those T cells are rendered ineffective for the reasons discussed in this review. Hypotheticals aside, these single-cell analyses expand on our understanding of immune TME changes with recurrence that suppress the immune response and facilitate glioma progression.

As discussed previously, epigenetic changes contribute significantly to the acquisition of an immune evasive phenotype in glioma cells in response to immune exposure, a mechanism widely referred to as immunoediting (68). These single-cell modalities to measure epigenetic changes can be combined with single-cell transcriptomics to provide unparalleled information on adaptive cell transitions within tumors. For example, a multimodal approach involving scRNA-seq and single-cell reduced representation bisulfite sequencing (scRRBS) in gliomas by Johnson et al. revealed DNA methylation-mediated induction of cell states associated with activation of stress response pathways and increased therapy resistance (60). Defined by distinct transcriptome and DNA methylome profiles, Johnson and colleagues identified three cell states referred to as stem-like, proliferating stem-like, and differentiated-like. Multi omic analysis revealed key transcriptional regulators in each cell state with differentiated-like cells having enhanced activity in transcription factors involved in hypoxia and stress response, demonstrating the value of epigenetic modifiers in stress-induced cell state plasticity (60).

Another technology amenable to single-cell implementation is the Assay for Transposase-Accessible Chromatin via Sequencing (ATAC-Seq (157)). Using this approach Guilhamon and colleagues found three GSC states, reactive, constructive, and invasive, distinguished by their chromatin accessibility profiles. Cells in the reactive state had increased accessibility to the promoters of immune-related genes, include genes associated with Tregs. Alternatively, the invasive GSC state had a demonstrated negative correlation with survival in orthotopic xenografts (201).

Currently, several technologies are being developed for multimodal single-cell application (Table 2) in addition to sophisticated computational tools for downstream analyses following single-cell sequencing. Often used in developmental biology but increasingly applied to cancer studies, these analyses can inform cellular hierarchies through pseudo-lineage tracing (e.g., pseudotime analysis (30, 205–207)), predicted cell state transitions (e.g., RNA Velocity (142)), and even metabolic phenotypes (e.g., Compass (208)). Since single-cell sequencing provides a point-in-time view of cell states, these algorithm-based insights into cellular transitions may reveal previously undetectable adaptive mechanisms in GSCs. Multi-plexing these state-of-the-art tools will permit interrogation of highly dimensional cell phenotypes at an unparalleled depth and can provide insight into clonal architecture and lineage tracing, regulators of GSC states, links between genetic and epigenetic profiles, drivers of adaptive responses to therapeutic or immune stimuli, and more (199). We envision these approaches being combined with scRNA-Seq to explore the role of the TME in promoting reversible, epigenetic-mediated GSC state transitions and, thus, build upon similar studies utilizing bulk multimodal analysis in patient-derived GSCs (61). Similar analyses at single-cell resolution would be invaluable in examining specific cell subsets and assessing their reciprocal impact on the TME.

Unfortunately, every rose has its thorn, and a major one for these cutting-edge single-cell sequencing technologies is the complete absence of spatial information. When it comes to understanding the influence of endogenous cell interactions, specifically GSC interactions with the surrounding TME, this information is critical, especially for pre-clinical studies relying on accurate representation of patient tumors. However, this information is lost with dissociative single-cell techniques, which is where newly generated spatially resolved omics technologies come into play.

3.3 Spatial omics

While cell-cell interactions can be inferred to a degree through computational manipulation of scRNA-seq data (209–212), spatial omics provide next-level understanding of these interactions by accounting for endogenous tissue organization and cell-cell proximity (213). Interrogation of RNA localization in tissue specimens using spatial transcriptomics techniques, broadly classified as either high-plex RNA imaging or spatial barcoding (199, 213), is arguably the most used spatial analysis as of now. High-plex imaging techniques include in situ sequencing (e.g., STARmap (170), FISSEQ (171)) and fluorescent in situ hybridization (e.g. RNAScope (172), MERFISH (173, 174), and seqFISH (175–177)) that provide single-cell resolution spatial gene expression with the limitation of relying on predetermined target transcripts and quality imaging instrumentation. On the other hand, spatial barcoding or in situ capture techniques, including Slide-seq (178, 179), 10X Visium, and high-definition spatial transcriptomics (180), provide unbiased, transcriptome-wide expression output. However, data is collected at multi-cellular spot resolution which requires additional deconvolution analysis to appropriately distinguish individual cell expression.

Aside from gene expression data, spatially resolved proteomic, metabolomic, and lipidomic information can be gathered from tissue specimens using imaging mass spectrometry or cytometry techniques. Unbiased coverage of biochemical features can be achieved through semi-quantitative mass spectrometry imaging (MSI) methods including time-of-flight secondary ion mass spectrometry (ToF-SIMS) (181, 182) and matrix-assisted laser desorption/ionization (MALDI) time-of-flight (MALDI-TOF) or fourier transform ion cyclotron resonance (MALDI-FTICR) mass spectrometry (183). Alternatively, imaging mass cytometry (IMC) can detect and quantify surface protein profiles using isotope-coupled antibody panels (184) in a method analogous to a spatially resolved CyTOF analysis. While many of the above spatial technologies have yet to be applied in the context of GBM, RNAScope and MSI methods have been utilized with the former identifying expression-based prognostic markers from GBM tissues (214, 215). On the other hand, MSI methods have investigated GBM heterogeneity based on metabolic and proteomic profiles and drug distribution in tumors (131, 216–220).

Integration of spatial and single-cell omics data grants unprecedented understanding of cellular and molecular phenotypes of cancer cells within their native environment. Ravi et. Al. define spatially distinct cellular phenotypes distinguished by transcriptional and metabolic programs through integration of spatial transcriptomics, MALDI-FTICR mass spectrometry, and mass cytometry proteomics (132). These five phenotypes were classified as radial glia, spatial OPC, neural development, reactive hypoxia, and reactive immune. Furthermore, cell phenotypes overlapped with previously established GSC states and were further distinguished by their spatial localization and proximity to immune cell infiltrates. Specifically, the reactive immune cell state, which has similarities to both the MES1 and AC-like states defined by Neftel et al. (59), had significant interaction with both myeloid and lymphoid infiltrating immune cells. Moreover, T cells near reactive-immune cells expressed higher levels of PD-1, representative of T cell exhaustion (132). Overall, multi-modal in situ sequencing of sequential tumor tissue slices paints a spatially resolved, complete picture of native cellular states, paving the way for mechanistic validation and identification of potentially translatable therapeutic vulnerabilities within specified cell types.

While single-cell and spatial analyses cannot classify GSCs by their defining functional traits (i.e., self-renewal capacity, multipotency, and tumor propagation capacity), these technologies are advantageous in their own rights for characterization of GSCs. Notably, chromatin accessibility profiles and validated GSC-specific transcriptomes can be used to delineate between GSCs and differentiated GBM cells in silico (134, 157, 201, 221–223). Additionally, high resolution omics data provides information at a remarkable depth necessary for proper investigation of dynamic and heterogeneous GSC states. Overall, the degree of insight into dynamic GSC biology afforded by multi-dimensional single-cell and spatial technologies has the potential to elucidate not only the factors that coordinate adaptive and immunosuppressive GSC mechanisms but their associated therapeutic vulnerabilities. Meticulous experimental design and utilization of spatially resolved single-cell technologies could conceivably inform novel treatment approaches capable of eliciting substantial anti-tumor response, a feat yet to be achieved in recurrent GBM.

4 Therapeutic opportunities

The existence of dynamic GSC states shaped by cell-intrinsic factors, the tumor microenvironment, and therapeutic pressures demand a revised approach to GBM therapy development that’s directed at disrupting the bidirectional interactions between GSCs and the TME that promote GBM malignancy, immune evasion, and tumor recurrence. Given the pervasive involvement of epigenetic modifications in GSC plasticity and acquisition of immunosuppressive phenotypes, using epigenetic inhibitors to augment the response of GBM to available immunotherapies warrants investigation (Figure 3). Notably, clinical attempts to treat newly diagnosed and recurrent GBM with epigenetic inhibitors (plus chemoradiation and surgery) have had negligible anti-tumor effects compared to standard-of-care treatment alone (186–191). However, preclinical studies investigating the immune effects of epigenetic inhibitors demonstrate increased immunogenicity, indicating that combining epigenetic and immunostimulatory treatment modalities may have potent anti-tumor effects (Table 3) (17, 44, 46, 104, 185). Two early phase clinical trials combining anti-PD1 checkpoint inhibitors with azacytidine (DNA methyltransferase inhibitor) or vorinostat (HDAC inhibitor) plus standard chemoradiation are currently underway in IDH-mutant gliomas (NCT03684811) and newly diagnosed GBM (NCT03426891), respectively. Additionally, several clinical trials testing combination epigenetic and immune therapy have been conducted or are currently underway in a variety of cancers, however, durable clinical effects are rare and often restricted to small subsets of patients (224–236). However, ongoing clinical trials may provide insight into mechanisms of resistance and opportunities for therapeutic improvement. Given the knowledge that immune cell states are also intimately controlled by epigenetic modifications (237), the lack of response from combination epigenetic-immune therapies may be due in part to unwanted effects on systemic immune cell pools that alter their phenotypic profiles and potentially induce dysfunctional states. Thus, strategies for targeting immune suppressive mechanism specifically within the TME while avoiding counterproductive systemic effects are needed. Despite these early phase results, the immunogenic effects of epigenetic inhibitors and combo epigenetic-immune therapy in preclinical GBM studies are promising (17, 44, 46, 104, 185, 192) and deserve to be explored to their full extent in clinical trials. Notably, increased immune recognition and resultant cytolysis of tumor cells by T cells and NK cells has been achieved through HDAC inhibitors and DNA hypomethylating agents (e.g., decitabine) (44, 46, 104, 185). Additionally, BRD4 inhibition attenuated an immunosuppressive GSC phenotype to enhance the anti-tumor immune response (17). Furthermore, BRD4 was implicated in an acquired immunosuppressive gene expression profile in GSCs refractory to CAR-T cell treatment. Targeting BRD4 in a mouse model of GBM effectively potentiated CAR-T cell therapy, highlighting the potential for clinical success with combination epigenetic and immunotherapy in GBM (192).

Figure 3

The involvement of GSCs in modulation of the anti-tumor immune response has prompted investigation into GSC-targeted immune-related therapies (Figure 3). Proposed methods include chimeric antigen recognition T cells (CAR-T cells) or dendritic cell vaccines directed at GSC-specific surface markers and antigens (53). However, the immunosuppressive effects of GSCs may attenuate the functionality of these engineered immune cells, thereby impeding therapeutic efficacy. Alternatively, forced differentiation of GSCs into a cellular state responsive to current therapeutic regimens holds promise (238), especially in combination with immunotherapy, but fails to consider how controlled differentiation would oppose plasticity mechanisms induced by standard-of-care treatments and/or potentially select for a more immunosuppressive GSC state. Therefore, we propose combining immunotherapy with approaches that disrupt mechanisms governing the transition of GSCs to immunosuppressive states and/or exploit cellular and molecular vulnerabilities inherent in these cell states (Figure 3). Identifying actionable therapeutic targets by incorporating high-dimensional, spatially aware single-cell analyses of patient tumor specimens and patient-derived GSC models into preclinical studies allows for development of therapeutics with high potential for clinical success.

Funding

This study is funded by the National Institute of Neurological Disorders and Stroke (NINDS) grants 1R01NS110087, 1R01NS096754 (JL), 1R01NS120949 (HLB).

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

• 1

  PatelAPTiroshITrombettaJJShalekAKGillespieSMWakimotoHet al. Single-cell rna-seq highlights intratumoral heterogeneity in primary glioblastoma. Science (2014) 344(6190):1396–401. doi: 10.1126/science.1254257

  • CrossRef
  • Google Scholar

• 2

  SottorivaASpiteriIPiccirilloSGTouloumisACollinsVPMarioniJCet al. Intratumor heterogeneity in human glioblastoma reflects cancer evolutionary dynamics. Proc Natl Acad Sci U S A (2013) 110(10):4009–14. doi: 10.1073/pnas.1219747110

  • CrossRef
  • Google Scholar

• 3

  GimpleRCBhargavaSDixitDRichJN. Glioblastoma stem cells: Lessons from the tumor hierarchy in a lethal cancer. Genes Dev (2019) 33(11-12):591–609. doi: 10.1101/gad.324301.119

  • CrossRef
  • Google Scholar

• 4

  LiauBBSieversCDonohueLKGillespieSMFlavahanWAMillerTEet al. Adaptive chromatin remodeling drives glioblastoma stem cell plasticity and drug tolerance. Cell Stem Cell (2017) 20(2):233–46.e7. doi: 10.1016/j.stem.2016.11.003

  • CrossRef
  • Google Scholar

• 5

  SchonbergDLLubelskiDMillerTERichJN. Brain tumor stem cells: Molecular characteristics and their impact on therapy. Mol Aspects Med (2014) 39:82–101. doi: 10.1016/j.mam.2013.06.004

  • CrossRef
  • Google Scholar

• 6

  ChenRNishimuraMCBumbacaSMKharbandaSForrestWFKasmanIMet al. A hierarchy of self-renewing tumor-initiating cell types in glioblastoma. Cancer Cell (2010) 17(4):362–75. doi: 10.1016/j.ccr.2009.12.049

  • CrossRef
  • Google Scholar

• 7

  DirkseAGolebiewskaABuderTNazarovPVMullerAPoovathingalSet al. Stem cell-associated heterogeneity in glioblastoma results from intrinsic tumor plasticity shaped by the microenvironment. Nat Commun (2019) 10(1):1787. doi: 10.1038/s41467-019-09853-z

  • CrossRef
  • Google Scholar

• 8

  LanXJorgDJCavalliFMGRichardsLMNguyenLVVannerRJet al. Fate mapping of human glioblastoma reveals an invariant stem cell hierarchy. Nature (2017) 549(7671):227–32. doi: 10.1038/nature23666

  • CrossRef
  • Google Scholar

• 9

  LiYLaterraJ. Cancer stem cells: Distinct entities or dynamically regulated phenotypes? Cancer Res (2012) 72(3):576–80. doi: 10.1158/0008-5472.CAN-11-3070

  • CrossRef
  • Google Scholar

• 10

  LiuXHuangJChenTWangYXinSLiJet al. Yamanaka factors critically regulate the developmental signaling network in mouse embryonic stem cells. Cell Res (2008) 18(12):1177–89. doi: 10.1038/cr.2008.309

  • CrossRef
  • Google Scholar

• 11

  TakahashiKYamanakaS. Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell (2006) 126(4):663–76. doi: 10.1016/j.cell.2006.07.024

  • CrossRef
  • Google Scholar

• 12

  Lopez-BertoniHLalBLiACaplanMGuerrero-CazaresHEberhartCGet al. Dnmt-dependent suppression of microrna regulates the induction of gbm tumor-propagating phenotype by Oct4 and Sox2. Oncogene (2015) 34(30):3994–4004. doi: 10.1038/onc.2014.334

  • CrossRef
  • Google Scholar

• 13

  LiYLiAGlasMLalBYingMSangYet al. C-met signaling induces a reprogramming network and supports the glioblastoma stem-like phenotype. Proc Natl Acad Sci U S A (2011) 108(24):9951–6. doi: 10.1073/pnas.1016912108

  • CrossRef
  • Google Scholar

• 14

  Lopez-BertoniHKotchetkovISMihelsonNLalBRuiYAmesHet al. A Sox2:Mir-486-5p axis regulates survival of gbm cells by inhibiting tumor suppressor networks. Cancer Res (2020) 80(8):1644–55. doi: 10.1158/0008-5472.CAN-19-1624

  • CrossRef
  • Google Scholar

• 15

  Lopez-BertoniHLalBMichelsonNGuerrero-CazaresHQuinones-HinojosaALiYet al. Epigenetic modulation of a mir-296-5p:Hmga1 axis regulates Sox2 expression and glioblastoma stem cells. Oncogene (2016) 35(37):4903–13. doi: 10.1038/onc.2016.22

  • CrossRef
  • Google Scholar

• 16

  Lopez-BertoniHJohnsonARuiYLalBSallSMalloyMet al. Sox2 induces glioblastoma cell stemness and tumor propagation by repressing Tet2 and deregulating 5hmc and 5mc DNA modifications. Signal Transduct Target Ther (2022) 7(1):37. doi: 10.1038/s41392-021-00857-0

  • CrossRef
  • Google Scholar

• 17

  MaTHuCLalBZhouWMaYYingMet al. Reprogramming transcription factors Oct4 and Sox2 induce a brd-dependent immunosuppressive transcriptome in gbm-propagating cells. Cancer Res (2021) 81(9):2457–69. doi: 10.1158/0008-5472.CAN-20-2489

  • CrossRef
  • Google Scholar

• 18

  OhueYNishikawaHRegulatory T (Treg). Cells in cancer: Can treg cells be a new therapeutic target? Cancer Sci (2019) 110(7):2080–9. doi: 10.1111/cas.14069

  • CrossRef
  • Google Scholar

• 19

  FriebelEKapolouKUngerSNunezNGUtzSRushingEJet al. Single-cell mapping of human brain cancer reveals tumor-specific instruction of tissue-invading leukocytes. Cell (2020) 181(7):1626–42.e20. doi: 10.1016/j.cell.2020.04.055

  • CrossRef
  • Google Scholar

• 20

  FuWWangWLiHJiaoYHuoRYanZet al. Single-cell atlas reveals complexity of the immunosuppressive microenvironment of initial and recurrent glioblastoma. Front Immunol (2020) 11:835. doi: 10.3389/fimmu.2020.00835

  • CrossRef
  • Google Scholar

• 21

  WangQHuBHuXKimHSquatritoMScarpaceLet al. Tumor evolution of glioma-intrinsic gene expression subtypes associates with immunological changes in the microenvironment. Cancer Cell (2017) 32(1):42–56.e6. doi: 10.1016/j.ccell.2017.06.003

  • CrossRef
  • Google Scholar

• 22

  MohmeMSchliffkeSMaireCLRungerAGlauLMendeKCet al. Immunophenotyping of newly diagnosed and recurrent glioblastoma defines distinct immune exhaustion profiles in peripheral and tumor-infiltrating lymphocytes. Clin Cancer Res (2018) 24(17):4187–200. doi: 10.1158/1078-0432.CCR-17-2617

  • CrossRef
  • Google Scholar

• 23

  WoronieckaKChongsathidkietPRhodinKKemenyHDechantCFarberSHet al. T-Cell exhaustion signatures vary with tumor type and are severe in glioblastoma. Clin Cancer Res (2018) 24(17):4175–86. doi: 10.1158/1078-0432.CCR-17-1846

  • CrossRef
  • Google Scholar

• 24

  DubinskiDWolferJHasselblattMSchneider-HohendorfTBogdahnUStummerWet al. Cd4+ T effector memory cell dysfunction is associated with the accumulation of granulocytic myeloid-derived suppressor cells in glioblastoma patients. Neuro Oncol (2016) 18(6):807–18. doi: 10.1093/neuonc/nov280

  • CrossRef
  • Google Scholar

• 25

  FecciPEMitchellDAWhitesidesJFXieWFriedmanAHArcherGEet al. Increased regulatory T-cell fraction amidst a diminished Cd4 compartment explains cellular immune defects in patients with malignant glioma. Cancer Res (2006) 66(6):3294–302. doi: 10.1158/0008-5472.CAN-05-3773

  • CrossRef
  • Google Scholar

• 26

  RaychaudhuriBRaymanPIrelandJKoJRiniBBordenECet al. Myeloid-derived suppressor cell accumulation and function in patients with newly diagnosed glioblastoma. Neuro Oncol (2011) 13(6):591–9. doi: 10.1093/neuonc/nor042

  • CrossRef
  • Google Scholar

• 27

  ElliottLHBrooksWHRoszmanTL. Activation of immunoregulatory lymphocytes obtained from patients with malignant gliomas. J Neurosurg (1987) 67(2):231–6. doi: 10.3171/jns.1987.67.2.0231

  • CrossRef
  • Google Scholar

• 28

  HaddadAFChenJSOhTPereiraMPJoshiRSAghiMK. Higher cytolytic score correlates with an immunosuppressive tumor microenvironment and reduced survival in glioblastoma. Sci Rep (2020) 10(1):17580. doi: 10.1038/s41598-020-73793-8

  • CrossRef
  • Google Scholar

• 29

  MarinariEAllardMGustaveRWidmerVPhilippinGMerklerDet al. Inflammation and lymphocyte infiltration are associated with shorter survival in patients with high-grade glioma. Oncoimmunology (2020) 9(1):1779990. doi: 10.1080/2162402X.2020.1779990

  • CrossRef
  • Google Scholar

• 30

  LandryAPBalasMAlliSSpearsJZadorZ. Distinct regional ontogeny and activation of tumor associated macrophages in human glioblastoma. Sci Rep (2020) 10(1):19542. doi: 10.1038/s41598-020-76657-3

  • CrossRef
  • Google Scholar

• 31

  VarnFSJohnsonKCMartinekJHuseJTNasrallahMPWesselingPet al. Glioma progression is shaped by genetic evolution and microenvironment interactions. Cell (2022) 185(12):2184–99.e16. doi: 10.1016/j.cell.2022.04.038

  • CrossRef
  • Google Scholar

• 32

  GrabowskiMMSankeyEWRyanKJChongsathidkietPLorreySJWilkinsonDSet al. Immune suppression in gliomas. J Neurooncol (2021) 151(1):3–12. doi: 10.1007/s11060-020-03483-y

  • CrossRef
  • Google Scholar

• 33

  BayikDZhouYParkCHongCVailDSilverDJet al. Myeloid-derived suppressor cell subsets drive glioblastoma growth in a sex-specific manner. Cancer Discovery (2020) 10(8):1210–25. doi: 10.1158/2159-8290.CD-19-1355

  • CrossRef
  • Google Scholar

• 34

  BettingerIThanosSPaulusW. Microglia promote glioma migration. Acta Neuropathol (2002) 103(4):351–5. doi: 10.1007/s00401-001-0472-x

  • CrossRef
  • Google Scholar

• 35

  Carvalho da FonsecaACWangHFanHChenXZhangIZhangLet al. Increased expression of stress inducible protein 1 in glioma-associated Microglia/Macrophages. J Neuroimmunol (2014) 274(1-2):71–7. doi: 10.1016/j.jneuroim.2014.06.021

  • CrossRef
  • Google Scholar

• 36

  ChenXZhangLZhangIYLiangJWangHOuyangMet al. Rage expression in tumor-associated macrophages promotes angiogenesis in glioma. Cancer Res (2014) 74(24):7285–97. doi: 10.1158/0008-5472.CAN-14-1240

  • CrossRef
  • Google Scholar

• 37

  OtvosBSilverDJMulkearns-HubertEEAlvaradoAGTuragaSMSorensenMDet al. Cancer stem cell-secreted macrophage migration inhibitory factor stimulates myeloid derived suppressor cell function and facilitates glioblastoma immune evasion. Stem Cells (2016) 34(8):2026–39. doi: 10.1002/stem.2393

  • CrossRef
  • Google Scholar

• 38

  WesolowskaAKwiatkowskaASlomnickiLDembinskiMMasterASliwaMet al. Microglia-derived tgf-beta as an important regulator of glioblastoma invasion–an inhibition of tgf-Beta-Dependent effects by shrna against human tgf-beta type ii receptor. Oncogene (2008) 27(7):918–30. doi: 10.1038/sj.onc.1210683

  • CrossRef
  • Google Scholar

• 39

  YeXZXuSLXinYHYuSCPingYFChenLet al. Tumor-associated Microglia/Macrophages enhance the invasion of glioma stem-like cells Via tgf-Beta1 signaling pathway. J Immunol (2012) 189(1):444–53. doi: 10.4049/jimmunol.1103248

  • CrossRef
  • Google Scholar

• 40

  BhatKPLBalasubramaniyanVVaillantBEzhilarasanRHummelinkKHollingsworthFet al. Mesenchymal differentiation mediated by nf-kappab promotes radiation resistance in glioblastoma. Cancer Cell (2013) 24(3):331–46. doi: 10.1016/j.ccr.2013.08.001

  • CrossRef
  • Google Scholar

• 41

  HaraTChanoch-MyersRMathewsonNDMyskiwCAttaLBussemaLet al. Interactions between cancer cells and immune cells drive transitions to mesenchymal-like states in glioblastoma. Cancer Cell (2021) 39(6):779–92.e11. doi: 10.1016/j.ccell.2021.05.002

  • CrossRef
  • Google Scholar

• 42

  PhillipsHSKharbandaSChenRForrestWFSorianoRHWuTDet al. Molecular subclasses of high-grade glioma predict prognosis, delineate a pattern of disease progression, and resemble stages in neurogenesis. Cancer Cell (2006) 9(3):157–73. doi: 10.1016/j.ccr.2006.02.019

  • CrossRef
  • Google Scholar

• 43

  VerhaakRGHoadleyKAPurdomEWangVQiYWilkersonMDet al. Integrated genomic analysis identifies clinically relevant subtypes of glioblastoma characterized by abnormalities in pdgfra, Idh1, egfr, and Nf1. Cancer Cell (2010) 17(1):98–110. doi: 10.1016/j.ccr.2009.12.020

  • CrossRef
  • Google Scholar

• 44

  YangWLiYGaoRXiuZSunT. Mhc class I dysfunction of glioma stem cells escapes from ctl-mediated immune response Via activation of Wnt/Beta-catenin signaling pathway. Oncogene (2020) 39(5):1098–111. doi: 10.1038/s41388-019-1045-6

  • CrossRef
  • Google Scholar

• 45

  RicklefsFLAlayoQKrenzlinHMahmoudABSperanzaMCNakashimaHet al. Immune evasion mediated by pd-L1 on glioblastoma-derived extracellular vesicles. Sci Adv (2018) 4(3):eaar2766. doi: 10.1126/sciadv.aar2766

  • CrossRef
  • Google Scholar

• 46

  RiccadonnaCYacoub MarounCVuillefroy de SillyRBoehlerMCalvo TardonMJueligerSet al. Decitabine treatment of glioma-initiating cells enhances immune recognition and killing. PLoS One (2016) 11(8):e0162105. doi: 10.1371/journal.pone.0162105

  • CrossRef
  • Google Scholar

• 47

  WuAWeiJKongLYWangYPriebeWQiaoWet al. Glioma cancer stem cells induce immunosuppressive Macrophages/Microglia. Neuro Oncol (2010) 12(11):1113–25. doi: 10.1093/neuonc/noq082

  • CrossRef
  • Google Scholar

• 48

  CossetEIlmjarvSDutoitVElliottKvon SchalschaTCamargoMFet al. Glut3 addiction is a druggable vulnerability for a molecularly defined subpopulation of glioblastoma. Cancer Cell (2017) 32(6):856–68.e5. doi: 10.1016/j.ccell.2017.10.016

  • CrossRef
  • Google Scholar

• 49

  FlavahanWAWuQHitomiMRahimNKimYSloanAEet al. Brain tumor initiating cells adapt to restricted nutrition through preferential glucose uptake. Nat Neurosci (2013) 16(10):1373–82. doi: 10.1038/nn.3510

  • CrossRef
  • Google Scholar

• 50

  HoPCBihuniakJDMacintyreANStaronMLiuXAmezquitaRet al. Phosphoenolpyruvate is a metabolic checkpoint of anti-tumor T cell responses. Cell (2015) 162(6):1217–28. doi: 10.1016/j.cell.2015.08.012

  • CrossRef
  • Google Scholar

• 51

  SiskaPJvan der WindtGJKishtonRJCohenSEisnerWMacIverNJet al. Suppression of Glut1 and glucose metabolism by decreased Akt/Mtorc1 signaling drives T cell impairment in b cell leukemia. J Immunol (2016) 197(6):2532–40. doi: 10.4049/jimmunol.1502464

  • CrossRef
  • Google Scholar

• 52

  HodgesTROttMXiuJGatalicaZSwensenJZhouSet al. Mutational burden, immune checkpoint expression, and mismatch repair in glioma: Implications for immune checkpoint immunotherapy. Neuro Oncol (2017) 19(8):1047–57. doi: 10.1093/neuonc/nox026

  • CrossRef
  • Google Scholar

• 53

  EsparzaRAzadTDFerozeAHMitraSSCheshierSH. Glioblastoma stem cells and stem cell-targeting immunotherapies. J Neurooncol (2015) 123(3):449–57. doi: 10.1007/s11060-015-1729-x

  • CrossRef
  • Google Scholar

• 54

  KimHZhengSAminiSSVirkSMMikkelsenTBratDJet al. Whole-genome and multisector exome sequencing of primary and post-treatment glioblastoma reveals patterns of tumor evolution. Genome Res (2015) 25(3):316–27. doi: 10.1101/gr.180612.114

  • CrossRef
  • Google Scholar

• 55

  GoldbergSBGettingerSNMahajanAChiangACHerbstRSSznolMet al. Pembrolizumab for patients with melanoma or non-Small-Cell lung cancer and untreated brain metastases: Early analysis of a non-randomised, open-label, phase 2 trial. Lancet Oncol (2016) 17(7):976–83. doi: 10.1016/S1470-2045(16)30053-5

  • CrossRef
  • Google Scholar

• 56

  KlugerHMChiangVMahajanAZitoCRSznolMTranTet al. Long-term survival of patients with melanoma with active brain metastases treated with pembrolizumab on a phase ii trial. J Clin Oncol (2019) 37(1):52–60. doi: 10.1200/JCO.18.00204

  • CrossRef
  • Google Scholar

• 57

  LongGVAtkinsonVLoSSandhuSGuminskiADBrownMPet al. Combination nivolumab and ipilimumab or nivolumab alone in melanoma brain metastases: A multicentre randomised phase 2 study. Lancet Oncol (2018) 19(5):672–81. doi: 10.1016/S1470-2045(18)30139-6

  • CrossRef
  • Google Scholar

• 58

  TawbiHAForsythPAHodiFSLaoCDMoschosSJHamidOet al. Safety and efficacy of the combination of nivolumab plus ipilimumab in patients with melanoma and asymptomatic or symptomatic brain metastases (Checkmate 204). Neuro Oncol (2021) 23(11):1961–73. doi: 10.1093/neuonc/noab094

  • CrossRef
  • Google Scholar

• 59

  NeftelCLaffyJFilbinMGHaraTShoreMERahmeGJet al. An integrative model of cellular states, plasticity, and genetics for glioblastoma. Cell (2019) 178(4):835–49.e21. doi: 10.1016/j.cell.2019.06.024

  • CrossRef
  • Google Scholar

• 60

  JohnsonKCAndersonKJCourtoisETGujarADBarthelFPVarnFSet al. Single-cell multimodal glioma analyses identify epigenetic regulators of cellular plasticity and environmental stress response. Nat Genet (2021) 53(10):1456–68. doi: 10.1038/s41588-021-00926-8

  • CrossRef
  • Google Scholar

• 61

  BaysanMWoolardKBozdagSRiddickGKotliarovaSCamMCet al. Micro-environment causes reversible changes in DNA methylation and mrna expression profiles in patient-derived glioma stem cells. PloS One (2014) 9(4):e94045. doi: 10.1371/journal.pone.0094045

  • CrossRef
  • Google Scholar

• 62

  ChaligneRGaitiFSilverbushDSchiffmanJSWeismanHRKluegelLet al. Epigenetic encoding, heritability and plasticity of glioma transcriptional cell states. Nat Genet (2021) 53(10):1469–79. doi: 10.1038/s41588-021-00927-7

  • CrossRef
  • Google Scholar

• 63

  KouzaridesT. Chromatin modifications and their function. Cell (2007) 128(4):693–705. doi: 10.1016/j.cell.2007.02.005

  • CrossRef
  • Google Scholar

• 64

  BatistaPJMolinieBWangJQuKZhangJLiLet al. M(6)a rna modification controls cell fate transition in mammalian embryonic stem cells. Cell Stem Cell (2014) 15(6):707–19. doi: 10.1016/j.stem.2014.09.019

  • CrossRef
  • Google Scholar

• 65

  GeulaSMoshitch-MoshkovitzSDominissiniDMansourAAKolNSalmon-DivonMet al. Stem cells. M6a mrna methylation facilitates resolution of naive pluripotency toward differentiation. Science (2015) 347(6225):1002–6. doi: 10.1126/science.1261417

  • CrossRef
  • Google Scholar

• 66

  WangYLiYTothJIPetroskiMDZhangZZhaoJC. N6-methyladenosine modification destabilizes developmental regulators in embryonic stem cells. Nat Cell Biol (2014) 16(2):191–8. doi: 10.1038/ncb2902

  • CrossRef
  • Google Scholar

• 67

  MarmorsteinRZhouMM. Writers and readers of histone acetylation: Structure, mechanism, and inhibition. Cold Spring Harb Perspect Biol (2014) 6(7):a018762. doi: 10.1101/cshperspect.a018762

  • CrossRef
  • Google Scholar

• 68

  GangosoESouthgateBBradleyLRusSGalvez-CancinoFMcGivernNet al. Glioblastomas acquire myeloid-affiliated transcriptional programs Via epigenetic immunoediting to elicit immune evasion. Cell (2021) 184(9):2454–70.e26. doi: 10.1016/j.cell.2021.03.023

  • CrossRef
  • Google Scholar

• 69

  PineARCiriglianoSMNicholsonJGHuYLinkousAMiyaguchiKet al. Tumor microenvironment is critical for the maintenance of cellular states found in primary glioblastomas. Cancer Discovery (2020) 10(7):964–79. doi: 10.1158/2159-8290.CD-20-0057

  • CrossRef
  • Google Scholar

• 70

  LiuBCaoYLiYMaHYangMZhangQet al. Glioma stem cells upregulate Cd39 expression to escape immune response through Sox2 modulation. Cancers (Basel) (2022) 14(3). doi: 10.3390/cancers14030783

  • CrossRef
  • Google Scholar

• 71

  DeaglioSDwyerKMGaoWFriedmanDUshevaAEratAet al. Adenosine generation catalyzed by Cd39 and Cd73 expressed on regulatory T cells mediates immune suppression. J Exp Med (2007) 204(6):1257–65. doi: 10.1084/jem.20062512

  • CrossRef
  • Google Scholar

• 72

  GordonSAkopyanGGarbanHBonavidaB. Transcription factor Yy1: Structure, function, and therapeutic implications in cancer biology. Oncogene (2006) 25(8):1125–42. doi: 10.1038/sj.onc.1209080

  • CrossRef
  • Google Scholar

• 73

  WilkinsonFHParkKAtchisonML. Polycomb recruitment to DNA in vivo by the Yy1 repo domain. Proc Natl Acad Sci U.S.A. (2006) 103(51):19296–301. doi: 10.1073/pnas.0603564103

  • CrossRef
  • Google Scholar

• 74

  JiaBLiuWGuJWangJLvWZhangWet al. Mir-7-5p suppresses stemness and enhances temozolomide sensitivity of drug-resistant glioblastoma cells by targeting yin yang 1. Exp Cell Res (2019) 375(1):73–81. doi: 10.1016/j.yexcr.2018.12.016

  • CrossRef
  • Google Scholar

• 75

  KaufholdSGarbanHBonavidaB. Yin yang 1 is associated with cancer stem cell transcription factors (Sox2, Oct4, Bmi1) and clinical implication. J Exp Clin Cancer Res (2016) 35:84. doi: 10.1186/s13046-016-0359-2

  • CrossRef
  • Google Scholar

• 76

  LiJSongJGuoF. Mir-186 reverses cisplatin resistance and inhibits the formation of the glioblastoma-initiating cell phenotype by degrading yin yang 1 in glioblastoma. Int J Mol Med (2019) 43(1):517–24. doi: 10.3892/ijmm.2018.3940

  • CrossRef
  • Google Scholar

• 77

  QiuZZhaoLShenJZLiangZWuQYangKet al. Transcription elongation machinery is a druggable dependency and potentiates immunotherapy in glioblastoma stem cells. Cancer Discovery (2022) 12(2):502–21. doi: 10.1158/2159-8290.CD-20-1848

  • CrossRef
  • Google Scholar

• 78

  GuJMuNJiaBGuoQPanLZhuMet al. Targeting radiation-tolerant persister cells as a strategy for inhibiting radioresistance and recurrence in glioblastoma. Neuro Oncol (2021) 24(7):1056–70. doi: 10.1093/neuonc/noab288

  • CrossRef
  • Google Scholar

• 79

  DeABeligalaDHSharmaVPBurgosCALeeAMGeuszME. Cancer stem cell generation during epithelial-mesenchymal transition is temporally gated by intrinsic circadian clocks. Clin Exp Metastasis (2020) 37(5):617–35. doi: 10.1007/s10585-020-10051-1

  • CrossRef
  • Google Scholar

• 80

  PuramRVKowalczykMSde BoerCGSchneiderRKMillerPGMcConkeyMet al. Core circadian clock genes regulate leukemia stem cells in aml. Cell (2016) 165(2):303–16. doi: 10.1016/j.cell.2016.03.015

  • CrossRef
  • Google Scholar

• 81

  PartchCLGreenCBTakahashiJS. Molecular architecture of the mammalian circadian clock. Trends Cell Biol (2014) 24(2):90–9. doi: 10.1016/j.tcb.2013.07.002

  • CrossRef
  • Google Scholar

• 82

  GeuszMEMalikADeA. Insights into oncogenesis from circadian timing in cancer stem cells. Crit Rev Oncog (2021) 26(4):1–17. doi: 10.1615/CritRevOncog.2021041960

  • CrossRef
  • Google Scholar

• 83

  HadadiEAcloqueH. Role of circadian rhythm disorders on emt and tumour-immune interactions in endocrine-related cancers. Endocr Relat Cancer (2021) 28(2):R67–80. doi: 10.1530/ERC-20-0390

  • CrossRef
  • Google Scholar

• 84

  HassanSAAliAAHSohnDFlogelUJanickeRUKorfHWet al. Does timing matter in radiotherapy of hepatocellular carcinoma? an experimental study in mice. Cancer Med (2021) 10(21):7712–25. doi: 10.1002/cam4.4277

  • CrossRef
  • Google Scholar

• 85

  HuangZHeAWangJLuHZhangRWuLet al. The circadian clock is associated with prognosis and immune infiltration in stomach adenocarcinoma. Aging (Albany NY) (2021) 13(12):16637–55. doi: 10.18632/aging.203184

  • CrossRef
  • Google Scholar

• 86

  LeeY. Roles of circadian clocks in cancer pathogenesis and treatment. Exp Mol Med (2021) 53(10):1529–38. doi: 10.1038/s12276-021-00681-0

  • CrossRef
  • Google Scholar

• 87

  de AssisLVMKinkerGSMoraesMNMarkusRPFernandesPACastrucciAML. Expression of the circadian clock gene Bmal1 positively correlates with antitumor immunity and patient survival in metastatic melanoma. Front Oncol (2018) 8:185. doi: 10.3389/fonc.2018.00185

  • CrossRef
  • Google Scholar

• 88

  HeWHoltkampSHergenhanSMKrausKde JuanAWeberJet al. Circadian expression of migratory factors establishes lineage-specific signatures that guide the homing of leukocyte subsets to tissues. Immunity (2018) 49(6):1175–90.e7. doi: 10.1016/j.immuni.2018.10.007

  • CrossRef
  • Google Scholar

• 89

  ChenPHsuWHChangATanZLanZZhouAet al. Circadian regulator clock recruits immune-suppressive microglia into the gbm tumor microenvironment. Cancer Discovery (2020) 10(3):371–81. doi: 10.1158/2159-8290.CD-19-0400

  • CrossRef
  • Google Scholar

• 90

  DongZZhangGQuMGimpleRCWuQQiuZet al. Targeting glioblastoma stem cells through disruption of the circadian clock. Cancer Discovery (2019) 9(11):1556–73. doi: 10.1158/2159-8290.CD-19-0215

  • CrossRef
  • Google Scholar

• 91

  LiALinXTanXYinBHanWZhaoJet al. Circadian gene clock contributes to cell proliferation and migration of glioma and is directly regulated by tumor-suppressive mir-124. FEBS Lett (2013) 587(15):2455–60. doi: 10.1016/j.febslet.2013.06.018

  • CrossRef
  • Google Scholar

• 92

  XuanWHsuWHKhanFDuntermanMPangLWainwrightDAet al. Circadian regulator clock drives immunosuppression in glioblastoma. Cancer Immunol Res (2022) 10(6):770–84. doi: 10.1158/2326-6066.CIR-21-0559

  • CrossRef
  • Google Scholar

• 93

  Eckel-PassowJELachanceDHMolinaroAMWalshKMDeckerPASicotteHet al. Glioma groups based on 1p/19q, idh, and tert promoter mutations in tumors. N Engl J Med (2015) 372(26):2499–508. doi: 10.1056/NEJMoa1407279

  • CrossRef
  • Google Scholar

• 94

  SuzukiHAokiKChibaKSatoYShiozawaYShiraishiYet al. Mutational landscape and clonal architecture in grade ii and iii gliomas. Nat Genet (2015) 47(5):458–68. doi: 10.1038/ng.3273

  • CrossRef
  • Google Scholar

• 95

  HuCWangKDamonCFuYMaTKratzLet al. Atrx loss promotes immunosuppressive mechanisms in Idh1 mutant glioma. Neuro Oncol (2022) 24(6):888–900. doi: 10.1093/neuonc/noab292

  • CrossRef
  • Google Scholar

• 96

  BabikirHWangLShamardaniKCatalanFSudhirSAghiMKet al. Atrx regulates glial identity and the tumor microenvironment in idh-mutant glioma. Genome Biol (2021) 22(1):311. doi: 10.1186/s13059-021-02535-4

  • CrossRef
  • Google Scholar

• 97

  ShinKShinHChoHJKangHLeeJKSeoYJet al. Sphere-forming culture for expanding genetically distinct patient-derived glioma stem cells by cellular growth rate screening. Cancers (Basel) (2020) 12(3). doi: 10.3390/cancers12030549

  • CrossRef
  • Google Scholar

• 98

  YeltonCJRaySK. Histone deacetylase enzymes and selective histone deacetylase inhibitors for antitumor effects and enhancement of antitumor immunity in glioblastoma. Neuroimmunol Neuroinflamm (2018) 5. doi: 10.20517/2347-8659.2018.58

  • CrossRef
  • Google Scholar

• 99

  FalkenbergKJJohnstoneRW. Histone deacetylases and their inhibitors in cancer, neurological diseases and immune disorders. Nat Rev Drug Discovery (2014) 13(9):673–91. doi: 10.1038/nrd4360

  • CrossRef
  • Google Scholar

• 100

  SetiadiAFDavidMDSeippRPHartikainenJAGopaulRJefferiesWA. Epigenetic control of the immune escape mechanisms in malignant carcinomas. Mol Cell Biol (2007) 27(22):7886–94. doi: 10.1128/MCB.01547-07

  • CrossRef
  • Google Scholar

• 101

  SetiadiAFOmilusikKDavidMDSeippRPHartikainenJGopaulRet al. Epigenetic enhancement of antigen processing and presentation promotes immune recognition of tumors. Cancer Res (2008) 68(23):9601–7. doi: 10.1158/0008-5472.CAN-07-5270

  • CrossRef
  • Google Scholar

• 102

  TopperMJVazMChiappinelliKBDeStefano ShieldsCENiknafsNYenRCet al. Epigenetic therapy ties myc depletion to reversing immune evasion and treating lung cancer. Cell (2017) 171(6):1284–300.e21. doi: 10.1016/j.cell.2017.10.022

  • CrossRef
  • Google Scholar

• 103

  WoanKVSahakianESotomayorEMSetoEVillagraA. Modulation of antigen-presenting cells by hdac inhibitors: Implications in autoimmunity and cancer. Immunol Cell Biol (2012) 90(1):55–65. doi: 10.1038/icb.2011.96

  • CrossRef
  • Google Scholar

• 104

  MorminoACocozzaGFontemaggiGValenteSEspositoVSantoroAet al. Histone-deacetylase 8 drives the immune response and the growth of glioma. Glia (2021) 69(11):2682–98. doi: 10.1002/glia.24065

  • CrossRef
  • Google Scholar

• 105

  HuGWadePA. Nurd and pluripotency: A complex balancing act. Cell Stem Cell (2012) 10(5):497–503. doi: 10.1016/j.stem.2012.04.011

  • CrossRef
  • Google Scholar

• 106

  ZhanXGuoSLiYRanHHuangHMiLet al. Glioma stem-like cells evade interferon suppression through Mbd3/Nurd complex-mediated Stat1 downregulation. J Exp Med (2020) 217(5). doi: 10.1084/jem.20191340

  • CrossRef
  • Google Scholar

• 107

  TurcanSRohleDGoenkaAWalshLAFangFYilmazEet al. Idh1 mutation is sufficient to establish the glioma hypermethylator phenotype. Nature (2012) 483(7390):479–83. doi: 10.1038/nature10866

  • CrossRef
  • Google Scholar

• 108

  ZhangXRaoASettePDeibertCPomerantzAKimWJet al. Idh mutant gliomas escape natural killer cell immune surveillance by downregulation of Nkg2d ligand expression. Neuro Oncol (2016) 18(10):1402–12. doi: 10.1093/neuonc/now061

  • CrossRef
  • Google Scholar

• 109

  AdairSJHoganKT. Treatment of ovarian cancer cell lines with 5-Aza-2'-Deoxycytidine upregulates the expression of cancer-testis antigens and class I major histocompatibility complex-encoded molecules. Cancer Immunol Immunother (2009) 58(4):589–601. doi: 10.1007/s00262-008-0582-6

  • CrossRef
  • Google Scholar

• 110

  FalahatRBerglundAPutneyRMPerez-VillarroelPAoyamaSPilon-ThomasSet al. Epigenetic reprogramming of tumor cell-intrinsic sting function sculpts antigenicity and T cell recognition of melanoma. Proc Natl Acad Sci U S A (2021) 118(15). doi: 10.1073/pnas.2013598118

  • CrossRef
  • Google Scholar

• 111

  FonsattiENicolayHJSigalottiLCalabroLPezzaniLColizziFet al. Functional up-regulation of human leukocyte antigen class I antigens expression by 5-Aza-2'-Deoxycytidine in cutaneous melanoma: Immunotherapeutic implications. Clin Cancer Res (2007) 13(11):3333–8. doi: 10.1158/1078-0432.CCR-06-3091

  • CrossRef
  • Google Scholar

• 112

  SerranoATanzarellaSLionelloIMendezRTraversariCRuiz-CabelloFet al. Rexpression of hla class I antigens and restoration of antigen-specific ctl response in melanoma cells following 5-Aza-2'-Deoxycytidine treatment. Int J Cancer (2001) 94(2):243–51. doi: 10.1002/ijc.1452

  • CrossRef
  • Google Scholar

• 113

  SeseBIniguez-MunozSEnsenyat-MendezMLlinas-AriasPRamisGOrozcoJIJet al. Glioblastoma embryonic-like stem cells exhibit immune-evasive phenotype. Cancers (Basel) (2022) 14(9). doi: 10.3390/cancers14092070

  • CrossRef
  • Google Scholar

• 114

  CuiQShiHYePLiLQuQSunGet al. M(6)a rna methylation regulates the self-renewal and tumorigenesis of glioblastoma stem cells. Cell Rep (2017) 18(11):2622–34. doi: 10.1016/j.celrep.2017.02.059

  • CrossRef
  • Google Scholar

• 115

  LiFChenSYuJGaoZSunZYiYet al. Interplay of M(6) a and histone modifications contributes to temozolomide resistance in glioblastoma. Clin Transl Med (2021) 11(9):e553. doi: 10.1002/ctm2.553

  • CrossRef
  • Google Scholar

• 116

  ShiJChenGDongXLiHLiSChengSet al. Mettl3 promotes the resistance of glioma to temozolomide Via increasing mgmt and anpg in a M(6)a dependent manner. Front Oncol (2021) 11:702983. doi: 10.3389/fonc.2021.702983

  • CrossRef
  • Google Scholar

• 117

  VisvanathanAPatilVAroraAHegdeASArivazhaganASantoshVet al. Essential role of Mettl3-mediated M(6)a modification in glioma stem-like cells maintenance and radioresistance. Oncogene (2018) 37(4):522–33. doi: 10.1038/onc.2017.351

  • CrossRef
  • Google Scholar

• 118

  ZhangSZhaoBSZhouALinKZhengSLuZet al. M(6)a demethylase Alkbh5 maintains tumorigenicity of glioblastoma stem-like cells by sustaining Foxm1 expression and cell proliferation program. Cancer Cell (2017) 31(4):591–606.e6. doi: 10.1016/j.ccell.2017.02.013

  • CrossRef
  • Google Scholar

• 119

  DongFQinXWangBLiQHuJChengXet al. Alkbh5 facilitates hypoxia-induced paraspeckle assembly and Il8 secretion to generate an immunosuppressive tumor microenvironment. Cancer Res (2021) 81(23):5876–88. doi: 10.1158/0008-5472.CAN-21-1456

  • CrossRef
  • Google Scholar

• 120

  HanahanDWeinbergRA. Hallmarks of cancer: The next generation. Cell (2011) 144(5):646–74. doi: 10.1016/j.cell.2011.02.013

  • CrossRef
  • Google Scholar

• 121

  Hoang-MinhLBSiebzehnrublFAYangCSuzuki-HatanoSDajacKLocheTet al. Infiltrative and drug-resistant slow-cycling cells support metabolic heterogeneity in glioblastoma. EMBO J (2018) 37(23). doi: 10.15252/embj.201798772

  • CrossRef
  • Google Scholar

• 122

  ShakyaSGromovskyADHaleJSKnudsenAMPragerBWallaceLCet al. Altered lipid metabolism marks glioblastoma stem and non-stem cells in separate tumor niches. Acta Neuropathol Commun (2021) 9(1):101. doi: 10.1186/s40478-021-01205-7

  • CrossRef
  • Google Scholar

• 123

  VlashiELagadecCVergnesLMatsutaniTMasuiKPoulouMet al. Metabolic state of glioma stem cells and nontumorigenic cells. Proc Natl Acad Sci U S A (2011) 108(38):16062–7. doi: 10.1073/pnas.1106704108

  • CrossRef
  • Google Scholar

• 124

  JaniszewskaMSuvaMLRiggiNHoutkooperRHAuwerxJClement-SchatloVet al. Imp2 controls oxidative phosphorylation and is crucial for preserving glioblastoma cancer stem cells. Genes Dev (2012) 26(17):1926–44. doi: 10.1101/gad.188292.112

  • CrossRef
  • Google Scholar

• 125

  KathagenASchulteABalckeGPhillipsHSMartensTMatschkeJet al. Hypoxia and oxygenation induce a metabolic switch between pentose phosphate pathway and glycolysis in glioma stem-like cells. Acta Neuropathol (2013) 126(5):763–80. doi: 10.1007/s00401-013-1173-y

  • CrossRef
  • Google Scholar

• 126

  DongZCuiH. Epigenetic modulation of metabolism in glioblastoma. Semin Cancer Biol (2019) 57:45–51. doi: 10.1016/j.semcancer.2018.09.002

  • CrossRef
  • Google Scholar

• 127

  HeddlestonJMWuQRiveraMMinhasSLathiaJDSloanAEet al. Hypoxia-induced mixed-lineage leukemia 1 regulates glioma stem cell tumorigenic potential. Cell Death Differ (2012) 19(3):428–39. doi: 10.1038/cdd.2011.109

  • CrossRef
  • Google Scholar

• 128

  HeilandDHGaebeleinABorriesMWornerJPompeNFrancoPet al. Microenvironment-derived regulation of hif signaling drives transcriptional heterogeneity in glioblastoma multiforme. Mol Cancer Res (2018) 16(4):655–68. doi: 10.1158/1541-7786.MCR-17-0680

  • CrossRef
  • Google Scholar

• 129

  JosephJVConroySPavlovKSontakkePTomarTEggens-MeijerEet al. Hypoxia enhances migration and invasion in glioblastoma by promoting a mesenchymal shift mediated by the Hif1alpha-Zeb1 axis. Cancer Lett (2015) 359(1):107–16. doi: 10.1016/j.canlet.2015.01.010

  • CrossRef
  • Google Scholar

• 130

  XuXWangLZangQLiSLiLWangZet al. Rewiring of purine metabolism in response to acidosis stress in glioma stem cells. Cell Death Dis (2021) 12(3):277. doi: 10.1038/s41419-021-03543-9

  • CrossRef
  • Google Scholar

• 131

  CoySWangSStopkaSALinJRYappCRitchCCet al. Single cell spatial analysis reveals the topology of immunomodulatory purinergic signaling in glioblastoma. Nat Commun (2022) 13(1):4814. doi: 10.1038/s41467-022-32430-w

  • CrossRef
  • Google Scholar

• 132

  RaviVMWillPKueckelhausJSunNJosephKSalieHet al. Spatially resolved multi-omics deciphers bidirectional tumor-host interdependence in glioblastoma. Cancer Cell (2022) 40(6):639–55.e13. doi: 10.1016/j.ccell.2022.05.009

  • CrossRef
  • Google Scholar

• 133

  MaoPJoshiKLiJKimSHLiPSantana-SantosLet al. Mesenchymal glioma stem cells are maintained by activated glycolytic metabolism involving aldehyde dehydrogenase 1a3. Proc Natl Acad Sci U S A (2013) 110(21):8644–9. doi: 10.1073/pnas.1221478110

  • CrossRef
  • Google Scholar

• 134

  RichardsLMWhitleyOKNMacLeodGCavalliFMGCoutinhoFJJaramilloJEet al. Gradient of developmental and injury response transcriptional states defines functional vulnerabilities underpinning glioblastoma heterogeneity. Nat Cancer (2021) 2(2):157–73. doi: 10.1038/s43018-020-00154-9

  • CrossRef
  • Google Scholar

• 135

  GarofanoLMigliozziSOhYTD'AngeloFNajacRDKoAet al. Pathway-based classification of glioblastoma uncovers a mitochondrial subtype with therapeutic vulnerabilities. Nat Cancer (2021) 2(2):141–56. doi: 10.1038/s43018-020-00159-4

  • CrossRef
  • Google Scholar

• 136

  WeiXChenYJiangXPengMLiuYMoYet al. Mechanisms of vasculogenic mimicry in hypoxic tumor microenvironments. Mol Cancer (2021) 20(1):7. doi: 10.1186/s12943-020-01288-1

  • CrossRef
  • Google Scholar

• 137

  ChengLHuangZZhouWWuQDonnolaSLiuJKet al. Glioblastoma stem cells generate vascular pericytes to support vessel function and tumor growth. Cell (2013) 153(1):139–52. doi: 10.1016/j.cell.2013.02.021

  • CrossRef
  • Google Scholar

• 138

  Ricci-VitianiLPalliniRBiffoniMTodaroMInverniciGCenciTet al. Tumour vascularization Via endothelial differentiation of glioblastoma stem-like cells. Nature (2010) 468(7325):824–8. doi: 10.1038/nature09557

  • CrossRef
  • Google Scholar

• 139

  WangRChadalavadaKWilshireJKowalikUHovingaKEGeberAet al. Glioblastoma stem-like cells give rise to tumour endothelium. Nature (2010) 468(7325):829–33. doi: 10.1038/nature09624

  • CrossRef
  • Google Scholar

• 140

  PastorinoOGentileMTManciniADel GaudioNDi CostanzoABajettoAet al. Histone deacetylase inhibitors impair vasculogenic mimicry from glioblastoma cells. Cancers (Basel) (2019) 11(6). doi: 10.3390/cancers11060747

  • CrossRef
  • Google Scholar

• 141

  TaoMLiXHeLRongXWangHPanJet al. Decreased rna M(6)a methylation enhances the process of the epithelial mesenchymal transition and vasculogenic mimicry in glioblastoma. Am J Cancer Res (2022) 12(2):893–906.

  • Google Scholar

• 142

  CouturierCPAyyadhurySLePUNadafJMonlongJRivaGet al. Single-cell rna-seq reveals that glioblastoma recapitulates a normal neurodevelopmental hierarchy. Nat Commun (2020) 11(1):3406. doi: 10.1038/s41467-020-17186-5

  • CrossRef
  • Google Scholar

• 143

  VenkataramaniVTanevDIStrahleCStudier-FischerAFankhauserLKesslerTet al. Glutamatergic synaptic input to glioma cells drives brain tumour progression. Nature (2019) 573(7775):532–8. doi: 10.1038/s41586-019-1564-x

  • CrossRef
  • Google Scholar

• 144

  VenkateshHSMorishitaWGeraghtyACSilverbushDGillespieSMArztMet al. Electrical and synaptic integration of glioma into neural circuits. Nature (2019) 573(7775):539–45. doi: 10.1038/s41586-019-1563-y

  • CrossRef
  • Google Scholar

• 145

  YuanJLevitinHMFrattiniVBushECBoyettDMSamanamudJet al. Single-cell transcriptome analysis of lineage diversity in high-grade glioma. Genome Med (2018) 10(1):57. doi: 10.1186/s13073-018-0567-9

  • CrossRef
  • Google Scholar

• 146

  VenkataramaniVYangYSchubertMCReyhanETetzlaffSKWissmannNet al. Glioblastoma hijacks neuronal mechanisms for brain invasion. Cell (2022) 185(16):2899–917.e31. doi: 10.1016/j.cell.2022.06.054

  • CrossRef
  • Google Scholar

• 147

  GaoRHeBHuangQWangZYanMLamEWet al. Cancer cell immune mimicry delineates onco-immunologic modulation. iScience (2021) 24(10):103133. doi: 10.1016/j.isci.2021.103133

  • CrossRef
  • Google Scholar

• 148

  Amanda JohnsonJKLaterraJLopez-BertoniH. Stem-03. Oct4/Sox2 drive an immunosuppressive gsc phenotype by inducing T-reg effector genes Via Tgfbr2 signaling. Neuro-Oncology (2021) 23:vi21–vi2. doi: 10.1093/neuonc/noab196.079

  • CrossRef
  • Google Scholar

• 149

  YoshiharaKShahmoradgoliMMartinezEVegesnaRKimHTorres-GarciaWet al. Inferring tumour purity and stromal and immune cell admixture from expression data. Nat Commun (2013) 4:2612. doi: 10.1038/ncomms3612

  • CrossRef
  • Google Scholar

• 150

  NewmanAMSteenCBLiuCLGentlesAJChaudhuriAASchererFet al. Determining cell type abundance and expression from bulk tissues with digital cytometry. Nat Biotechnol (2019) 37(7):773–82. doi: 10.1038/s41587-019-0114-2

  • CrossRef
  • Google Scholar

• 151

  AranDHuZButteAJ. Xcell: Digitally portraying the tissue cellular heterogeneity landscape. Genome Biol (2017) 18(1):220. doi: 10.1186/s13059-017-1349-1

  • CrossRef
  • Google Scholar

• 152

  BechtEGiraldoNALacroixLButtardBElarouciNPetitprezFet al. Estimating the population abundance of tissue-infiltrating immune and stromal cell populations using gene expression. Genome Biol (2016) 17(1):218. doi: 10.1186/s13059-016-1070-5

  • CrossRef
  • Google Scholar

• 153

  RacleJde JongeKBaumgaertnerPSpeiserDEGfellerD. Simultaneous enumeration of cancer and immune cell types from bulk tumor gene expression data. Elife (2017) 6. doi: 10.7554/eLife.26476

  • CrossRef
  • Google Scholar

• 154

  RacleJGfellerD. Epic: A tool to estimate the proportions of different cell types from bulk gene expression data. Methods Mol Biol (2020) 2120:233–48. doi: 10.1007/978-1-0716-0327-7_17

  • CrossRef
  • Google Scholar

• 155

  FinotelloFMayerCPlattnerCLaschoberGRiederDHacklHet al. Molecular and pharmacological modulators of the tumor immune contexture revealed by deconvolution of rna-seq data. Genome Med (2019) 11(1):34. doi: 10.1186/s13073-019-0638-6

  • CrossRef
  • Google Scholar

• 156

  LiTFuJZengZCohenDLiJChenQet al. Timer2.0 for analysis of tumor-infiltrating immune cells. Nucleic Acids Res (2020) 48(W1):W509–W14. doi: 10.1093/nar/gkaa407

  • CrossRef
  • Google Scholar

• 157

  BuenrostroJDCorcesMRLareauCAWuBSchepANAryeeMJet al. Integrated single-cell analysis maps the continuous regulatory landscape of human hematopoietic differentiation. Cell (2018) 173(6):1535–48.e16. doi: 10.1016/j.cell.2018.03.074

  • CrossRef
  • Google Scholar

• 158

  GuoHZhuPWuXLiXWenLTangF. Single-cell methylome landscapes of mouse embryonic stem cells and early embryos analyzed using reduced representation bisulfite sequencing. Genome Res (2013) 23(12):2126–35. doi: 10.1101/gr.161679.113

  • CrossRef
  • Google Scholar

• 159

  BanduraDRBaranovVIOrnatskyOIAntonovAKinachRLouXet al. Mass cytometry: Technique for real time single cell multitarget immunoassay based on inductively coupled plasma time-of-Flight mass spectrometry. Anal Chem (2009) 81(16):6813–22. doi: 10.1021/ac901049w

  • CrossRef
  • Google Scholar

• 160

  GadallaRNoamaniBMacLeodBLDicksonRJGuoMXuWet al. Validation of cytof against flow cytometry for immunological studies and monitoring of human cancer clinical trials. Front Oncol (2019) 9:415. doi: 10.3389/fonc.2019.00415

  • CrossRef
  • Google Scholar

• 161

  PaiJASatpathyAT. High-throughput and single-cell T cell receptor sequencing technologies. Nat Methods (2021) 18(8):881–92. doi: 10.1038/s41592-021-01201-8

  • CrossRef
  • Google Scholar

• 162

  BartosovicMKabbeMCastelo-BrancoG. Single-cell Cut&Tag profiles histone modifications and transcription factors in complex tissues. Nat Biotechnol (2021) 39(7):825–35. doi: 10.1038/s41587-021-00869-9

  • CrossRef
  • Google Scholar

• 163

  ZhangBSrivastavaAMimitouEStuartTRaimondiIHaoYet al. Characterizing cellular heterogeneity in chromatin state with Sccut&Tag-pro. Nat Biotechnol (2022) 40(8):1220–30. doi: 10.1038/s41587-022-01250-0

  • CrossRef
  • Google Scholar

• 164

  PetersonVMZhangKXKumarNWongJLiLWilsonDCet al. Multiplexed quantification of proteins and transcripts in single cells. Nat Biotechnol (2017) 35(10):936–9. doi: 10.1038/nbt.3973

  • CrossRef
  • Google Scholar

• 165

  StoeckiusMHafemeisterCStephensonWHouck-LoomisBChattopadhyayPKSwerdlowHet al. Simultaneous epitope and transcriptome measurement in single cells. Nat Methods (2017) 14(9):865–8. doi: 10.1038/nmeth.4380

  • CrossRef
  • Google Scholar

• 166

  CaoJCusanovichDARamaniVAghamirzaieDPlinerHAHillAJet al. Joint profiling of chromatin accessibility and gene expression in thousands of single cells. Science (2018) 361(6409):1380–5. doi: 10.1126/science.aau0730

  • CrossRef
  • Google Scholar

• 167

  BianSHouYZhouXLiXYongJWangYet al. Single-cell multiomics sequencing and analyses of human colorectal cancer. Science (2018) 362(6418):1060–3. doi: 10.1126/science.aao3791

  • CrossRef
  • Google Scholar

• 168

  HouYGuoHCaoCLiXHuBZhuPet al. Single-cell triple omics sequencing reveals genetic, epigenetic, and transcriptomic heterogeneity in hepatocellular carcinomas. Cell Res (2016) 26(3):304–19. doi: 10.1038/cr.2016.23

  • CrossRef
  • Google Scholar

• 169

  ClarkSJArgelaguetRKapouraniCAStubbsTMLeeHJAlda-CatalinasCet al. Scnmt-seq enables joint profiling of chromatin accessibility DNA methylation and transcription in single cells. Nat Commun (2018) 9(1):781. doi: 10.1038/s41467-018-03149-4

  • CrossRef
  • Google Scholar

• 170

  WangXAllenWEWrightMASylwestrakELSamusikNVesunaSet al. Three-dimensional intact-tissue sequencing of single-cell transcriptional states. Science (2018) 361(6400). doi: 10.1126/science.aat5691

  • CrossRef
  • Google Scholar

• 171

  LeeJHDaugharthyERScheimanJKalhorRYangJLFerranteTCet al. Highly multiplexed subcellular rna sequencing in situ. Science (2014) 343(6177):1360–3. doi: 10.1126/science.1250212

  • CrossRef
  • Google Scholar

• 172

  WangFFlanaganJSuNWangLCBuiSNielsonAet al. Rnascope: A novel in situ rna analysis platform for formalin-fixed, paraffin-embedded tissues. J Mol Diagn (2012) 14(1):22–9. doi: 10.1016/j.jmoldx.2011.08.002

  • CrossRef
  • Google Scholar

• 173

  ChenKHBoettigerANMoffittJRWangSZhuangX. Rna imaging. spatially resolved, highly multiplexed rna profiling in single cells. Science (2015) 348(6233):aaa6090. doi: 10.1126/science.aaa6090

  • CrossRef
  • Google Scholar

• 174

  XiaCFanJEmanuelGHaoJZhuangX. Spatial transcriptome profiling by merfish reveals subcellular rna compartmentalization and cell cycle-dependent gene expression. Proc Natl Acad Sci U S A (2019) 116(39):19490–9. doi: 10.1073/pnas.1912459116

  • CrossRef
  • Google Scholar

• 175

  EngCLLawsonMZhuQDriesRKoulenaNTakeiYet al. Transcriptome-scale super-resolved imaging in tissues by rna seqfish. Nature (2019) 568(7751):235–9. doi: 10.1038/s41586-019-1049-y

  • CrossRef
  • Google Scholar

• 176

  LubeckECoskunAFZhiyentayevTAhmadMCaiL. Single-cell in situ rna profiling by sequential hybridization. Nat Methods (2014) 11(4):360–1. doi: 10.1038/nmeth.2892

  • CrossRef
  • Google Scholar

• 177

  ShahSLubeckEZhouWCaiL. In situ transcription profiling of single cells reveals spatial organization of cells in the mouse hippocampus. Neuron (2016) 92(2):342–57. doi: 10.1016/j.neuron.2016.10.001

  • CrossRef
  • Google Scholar

• 178

  RodriquesSGStickelsRRGoevaAMartinCAMurrayEVanderburgCRet al. Slide-seq: A scalable technology for measuring genome-wide expression at high spatial resolution. Science (2019) 363(6434):1463–7. doi: 10.1126/science.aaw1219

  • CrossRef
  • Google Scholar

• 179

  StickelsRRMurrayEKumarPLiJMarshallJLDi BellaDJet al. Highly sensitive spatial transcriptomics at near-cellular resolution with slide-Seqv2. Nat Biotechnol (2021) 39(3):313–9. doi: 10.1038/s41587-020-0739-1

  • CrossRef
  • Google Scholar

• 180

  VickovicSEraslanGSalmenFKlughammerJStenbeckLSchapiroDet al. High-definition spatial transcriptomics for in situ tissue profiling. Nat Methods (2019) 16(10):987–90. doi: 10.1038/s41592-019-0548-y

  • CrossRef
  • Google Scholar

• 181

  GularyanSKGulinAAAnufrievaKSShenderVOShakhparonovMIBastolaSet al. Investigation of inter- and intratumoral heterogeneity of glioblastoma using tof-sims. Mol Cell Proteomics (2020) 19(6):960–70. doi: 10.1074/mcp.RA120.001986

  • CrossRef
  • Google Scholar

• 182

  JungnickelHLauxPLuchA. Time-of-Flight secondary ion mass spectrometry (Tof-sims): A new tool for the analysis of toxicological effects on single cell level. Toxics (2016) 4(1). doi: 10.3390/toxics4010005

  • CrossRef
  • Google Scholar

• 183

  SpragginsJMRizzoDGMooreJLNotoMJSkaarEPCaprioliRM. Next-generation technologies for spatial proteomics: Integrating ultra-high speed maldi-tof and high mass resolution maldi fticr imaging mass spectrometry for protein analysis. Proteomics (2016) 16(11-12):1678–89. doi: 10.1002/pmic.201600003

  • CrossRef
  • Google Scholar

• 184

  GiesenCWangHASchapiroDZivanovicNJacobsAHattendorfBet al. Highly multiplexed imaging of tumor tissues with subcellular resolution by mass cytometry. Nat Methods (2014) 11(4):417–22. doi: 10.1038/nmeth.2869

  • CrossRef
  • Google Scholar

• 185

  MaRReiMWoodhouseIFerrisKKirschnerSChandranAet al. Decitabine increases neoantigen and cancer testis antigen expression to enhance T cell-mediated toxicity against glioblastoma. Neuro Oncol (2022). doi: 10.1093/neuonc/noac107

  • CrossRef
  • Google Scholar

• 186

  GalanisEAndersonSKMillerCRSarkariaJNJaeckleKBucknerJCet al. Phase I/Ii trial of vorinostat combined with temozolomide and radiation therapy for newly diagnosed glioblastoma: Results of alliance N0874/Abtc 02. Neuro Oncol (2018) 20(4):546–56. doi: 10.1093/neuonc/nox161

  • CrossRef
  • Google Scholar

• 187

  GalanisEJaeckleKAMaurerMJReidJMAmesMMHardwickJSet al. Phase ii trial of vorinostat in recurrent glioblastoma multiforme: A north central cancer treatment group study. J Clin Oncol (2009) 27(12):2052–8. doi: 10.1200/JCO.2008.19.0694

  • CrossRef
  • Google Scholar

• 188

  PetersKBLippESMillerEHerndonJE2ndMcSherryFDesjardinsAet al. Phase I/Ii trial of vorinostat, bevacizumab, and daily temozolomide for recurrent malignant gliomas. J Neurooncol (2018) 137(2):349–56. doi: 10.1007/s11060-017-2724-1

  • CrossRef
  • Google Scholar

• 189

  PuduvalliVKWuJYuanYArmstrongTSVeraEWuJet al. A Bayesian adaptive randomized phase ii multicenter trial of bevacizumab with or without vorinostat in adults with recurrent glioblastoma. Neuro Oncol (2020) 22(10):1505–15. doi: 10.1093/neuonc/noaa062

  • CrossRef
  • Google Scholar

• 190

  IwamotoFMLambornKRKuhnJGWenPYYungWKGilbertMRet al. A phase I/Ii trial of the histone deacetylase inhibitor romidepsin for adults with recurrent malignant glioma: North American brain tumor consortium study 03-03. Neuro Oncol (2011) 13(5):509–16. doi: 10.1093/neuonc/nor017

  • CrossRef
  • Google Scholar

• 191

  LeeEQReardonDASchiffDDrappatzJMuzikanskyAGrimmSAet al. Phase ii study of panobinostat in combination with bevacizumab for recurrent glioblastoma and anaplastic glioma. Neuro Oncol (2015) 17(6):862–7. doi: 10.1093/neuonc/nou350

  • CrossRef
  • Google Scholar

• 192

  XiaLLiuJYZhengZZChenYJDingJCHuYHet al. Brd4 inhibition boosts the therapeutic effects of epidermal growth factor receptor-targeted chimeric antigen receptor T cells in glioblastoma. Mol Ther (2021) 29(10):3011–26. doi: 10.1016/j.ymthe.2021.05.019

  • CrossRef
  • Google Scholar

• 193

  JiangZShiYZhaoWZhangYXieYZhangBet al. Development of an immune-related prognostic index associated with glioblastoma. Front Neurol (2021) 12:610797. doi: 10.3389/fneur.2021.610797

  • CrossRef
  • Google Scholar

• 194

  QiuHLiYChengSLiJHeCLiJ. A prognostic microenvironment-related immune signature Via estimate (Promise model) predicts overall survival of patients with glioma. Front Oncol (2020) 10:580263. doi: 10.3389/fonc.2020.580263

  • CrossRef
  • Google Scholar

• 195

  Consortium G. Glioma through the looking glass: Molecular evolution of diffuse gliomas and the glioma longitudinal analysis consortium. Neuro Oncol (2018) 20(7):873–84. doi: 10.1093/neuonc/noy020

  • CrossRef
  • Google Scholar

• 196

  PuchalskiRBShahNMillerJDalleyRNomuraSRYoonJGet al. An anatomic transcriptional atlas of human glioblastoma. Science (2018) 360(6389):660–3. doi: 10.1126/science.aaf2666

  • CrossRef
  • Google Scholar

• 197

  LouisDNPerryAWesselingPBratDJCreeIAFigarella-BrangerDet al. The 2021 who classification of tumors of the central nervous system: A summary. Neuro Oncol (2021) 23(8):1231–51. doi: 10.1093/neuonc/noab106

  • CrossRef
  • Google Scholar

• 198

  StuartTSatijaR. Integrative single-cell analysis. Nat Rev Genet (2019) 20(5):257–72. doi: 10.1038/s41576-019-0093-7

  • CrossRef
  • Google Scholar

• 199

  NamASChaligneRLandauDA. Integrating genetic and non-genetic determinants of cancer evolution by single-cell multi-omics. Nat Rev Genet (2021) 22(1):3–18. doi: 10.1038/s41576-020-0265-5

  • CrossRef
  • Google Scholar

• 200

  SuvaMLTiroshI. Single-cell rna sequencing in cancer: Lessons learned and emerging challenges. Mol Cell (2019) 75(1):7–12. doi: 10.1016/j.molcel.2019.05.003

  • CrossRef
  • Google Scholar

• 201

  GuilhamonPChesnelongCKushidaMMNikolicASinghalDMacLeodGet al. Single-cell chromatin accessibility profiling of glioblastoma identifies an invasive cancer stem cell population associated with lower survival. Elife (2021) 10. doi: 10.7554/eLife.64090

  • CrossRef
  • Google Scholar

• 202

  SimondsEFLuEDBadilloOKarimiSLiuEVTamakiWet al. Deep immune profiling reveals targetable mechanisms of immune evasion in immune checkpoint inhibitor-refractory glioblastoma. J Immunother Cancer (2021) 9(6). doi: 10.1136/jitc-2020-002181

  • CrossRef
  • Google Scholar

• 203

  Pombo AntunesARScheyltjensILodiFMessiaenJAntoranzADuerinckJet al. Single-cell profiling of myeloid cells in glioblastoma across species and disease stage reveals macrophage competition and specialization. Nat Neurosci (2021) 24(4):595–610. doi: 10.1038/s41593-020-00789-y

  • CrossRef
  • Google Scholar

• 204

  YeoATRawalSDelcuzeBChristofidesAAtaydeAStraussLet al. Single-cell rna sequencing reveals evolution of immune landscape during glioblastoma progression. Nat Immunol (2022) 23(6):971–84. doi: 10.1038/s41590-022-01215-0

  • CrossRef
  • Google Scholar

• 205

  SettyMKiseliovasVLevineJGayosoAMazutisLPe'erD. Characterization of cell fate probabilities in single-cell data with palantir. Nat Biotechnol (2019) 37(4):451–60. doi: 10.1038/s41587-019-0068-4

  • CrossRef
  • Google Scholar

• 206

  StreetKRissoDFletcherRBDasDNgaiJYosefNet al. Slingshot: Cell lineage and pseudotime inference for single-cell transcriptomics. BMC Genomics (2018) 19(1):477. doi: 10.1186/s12864-018-4772-0

  • CrossRef
  • Google Scholar

• 207

  TrapnellCCacchiarelliDGrimsbyJPokharelPLiSMorseMet al. The dynamics and regulators of cell fate decisions are revealed by pseudotemporal ordering of single cells. Nat Biotechnol (2014) 32(4):381–6. doi: 10.1038/nbt.2859

  • CrossRef
  • Google Scholar

• 208

  WagnerAWangCFesslerJDeTomasoDAvila-PachecoJKaminskiJet al. Metabolic modeling of single Th17 cells reveals regulators of autoimmunity. Cell (2021) 184(16):4168–85.e21. doi: 10.1016/j.cell.2021.05.045

  • CrossRef
  • Google Scholar

• 209

  BrowaeysRSaelensWSaeysY. Nichenet: Modeling intercellular communication by linking ligands to target genes. Nat Methods (2020) 17(2):159–62. doi: 10.1038/s41592-019-0667-5

  • CrossRef
  • Google Scholar

• 210

  KumarMPDuJLagoudasGJiaoYSawyerADrummondDCet al. Analysis of single-cell rna-seq identifies cell-cell communication associated with tumor characteristics. Cell Rep (2018) 25(6):1458–68.e4. doi: 10.1016/j.celrep.2018.10.047

  • CrossRef
  • Google Scholar

• 211

  SmillieCSBitonMOrdovas-MontanesJSullivanKMBurginGGrahamDBet al. Intra- and inter-cellular rewiring of the human colon during ulcerative colitis. Cell (2019) 178(3):714–30.e22. doi: 10.1016/j.cell.2019.06.029

  • CrossRef
  • Google Scholar

• 212

  Vento-TormoREfremovaMBottingRATurcoMYVento-TormoMMeyerKBet al. Single-cell reconstruction of the early maternal-fetal interface in humans. Nature (2018) 563(7731):347–53. doi: 10.1038/s41586-018-0698-6

  • CrossRef
  • Google Scholar

• 213

  LongoSKGuoMGJiALKhavariPA. Integrating single-cell and spatial transcriptomics to elucidate intercellular tissue dynamics. Nat Rev Genet (2021) 22(10):627–44. doi: 10.1038/s41576-021-00370-8

  • CrossRef
  • Google Scholar

• 214

  YuanQCaiHQZhongYZhangMJChengZJHaoJJet al. Overexpression of Igfbp2 mrna predicts poor survival in patients with glioblastoma. Biosci Rep (2019) 39(6). doi: 10.1042/BSR20190045

  • CrossRef
  • Google Scholar

• 215

  YuanQWangSQZhangGTHeJLiuZDWangMRet al. Highly expressed of Serpina3 indicated poor prognosis and involved in immune suppression in glioma. Immun Inflammation Dis (2021) 9(4):1618–30. doi: 10.1002/iid3.515

  • CrossRef
  • Google Scholar

• 216

  CornettDSFrappierSLCaprioliRM. Maldi-fticr imaging mass spectrometry of drugs and metabolites in tissue. Anal Chem (2008) 80(14):5648–53. doi: 10.1021/ac800617s

  • CrossRef
  • Google Scholar

• 217

  DililloMAit-BelkacemREsteveCPellegriniDNicolardiSCostaMet al. Ultra-high mass resolution maldi imaging mass spectrometry of proteins and metabolites in a mouse model of glioblastoma. Sci Rep (2017) 7(1):603. doi: 10.1038/s41598-017-00703-w

  • CrossRef
  • Google Scholar

• 218

  Gomez-CaudilloLOrtega-LozanoAJMartinez-BatallarAGRosas-VargasHMinauro-SanmiguelFEncarnacion-GuevaraS. Principal component analysis on Lcms/Ms and 2demalditof in glioblastoma cell lines reveals that mitochondria act as organelle sensors of the metabolic state in glioblastoma. Oncol Rep (2020) 44(2):661–73. doi: 10.3892/or.2020.7625

  • CrossRef
  • Google Scholar

• 219

  ParkCKJungJHParkSHJungHWChoBK. Multifarious proteomic signatures and regional heterogeneity in glioblastomas. J Neurooncol (2009) 94(1):31–9. doi: 10.1007/s11060-009-9805-8

  • CrossRef
  • Google Scholar

• 220

  PetreGDurandHPelletierLPoulenardMNugueGRayPFet al. Rapid proteomic profiling by maldi-tof mass spectrometry for better brain tumor classification. Proteomics Clin Appl (2020) 14(5):e1900116. doi: 10.1002/prca.201900116

  • CrossRef
  • Google Scholar

• 221

  GomezNCHepperlaAJDumitruRSimonJMFangFDavisIJ. Widespread chromatin accessibility at repetitive elements links stem cells with human cancer. Cell Rep (2016) 17(6):1607–20. doi: 10.1016/j.celrep.2016.10.011

  • CrossRef
  • Google Scholar

• 222

  StergachisABNephSReynoldsAHumbertRMillerBPaigeSLet al. Developmental fate and cellular maturity encoded in human regulatory DNA landscapes. Cell (2013) 154(4):888–903. doi: 10.1016/j.cell.2013.07.020

  • CrossRef
  • Google Scholar

• 223

  SuvaMLRheinbayEGillespieSMPatelAPWakimotoHRabkinSDet al. Reconstructing and reprogramming the tumor-propagating potential of glioblastoma stem-like cells. Cell (2014) 157(3):580–94. doi: 10.1016/j.cell.2014.02.030

  • CrossRef
  • Google Scholar

• 224

  BeverKMThomasDL2ndZhangJDiaz RiveraEARosnerGLZhuQet al. A feasibility study of combined epigenetic and vaccine therapy in advanced colorectal cancer with pharmacodynamic endpoint. Clin Epigenet (2021) 13(1):25. doi: 10.1186/s13148-021-01014-8

  • CrossRef
  • Google Scholar

• 225

  ChienKSKimKNogueras-GonzalezGMBorthakurGNaqviKDaverNGet al. Phase ii study of azacitidine with pembrolizumab in patients with intermediate-1 or higher-risk myelodysplastic syndrome. Br J Haematol (2021) 195(3):378–87. doi: 10.1111/bjh.17689

  • CrossRef
  • Google Scholar

• 226

  DaverNGarcia-ManeroGBasuSBodduPCAlfayezMCortesJEet al. Efficacy, safety, and biomarkers of response to azacitidine and nivolumab in Relapsed/Refractory acute myeloid leukemia: A nonrandomized, open-label, phase ii study. Cancer Discovery (2019) 9(3):370–83. doi: 10.1158/2159-8290.CD-18-0774

  • CrossRef
  • Google Scholar

• 227

  Di GiacomoAMCovreAFinotelloFRiederDDanielliRSigalottiLet al. Guadecitabine plus ipilimumab in unresectable melanoma: The nibit-M4 clinical trial. Clin Cancer Res (2019) 25(24):7351–62. doi: 10.1158/1078-0432.CCR-19-1335

  • CrossRef
  • Google Scholar

• 228

  Holmberg-ThydenSDufvaIHGangAOBreinholtMFSchejbelLAndersenMKet al. Epigenetic therapy in combination with a multi-epitope cancer vaccine targeting shared tumor antigens for high-risk myelodysplastic syndrome - a phase I clinical trial. Cancer Immunol Immunother (2022) 71(2):433–44. doi: 10.1007/s00262-021-02993-6

  • CrossRef
  • Google Scholar

• 229

  KuangCParkYAugustinRCLinYHartmanDJSeighLet al. Pembrolizumab plus azacitidine in patients with chemotherapy refractory metastatic colorectal cancer: A single-arm phase 2 trial and correlative biomarker analysis. Clin Epigenet (2022) 14(1):3. doi: 10.1186/s13148-021-01226-y

  • CrossRef
  • Google Scholar

• 230

  LevyBPGiacconeGBesseBFelipEGarassinoMCDomine GomezMet al. Randomised phase 2 study of pembrolizumab plus cc-486 versus pembrolizumab plus placebo in patients with previously treated advanced non-small cell lung cancer. Eur J Cancer (2019) 108:120–8. doi: 10.1016/j.ejca.2018.11.028

  • CrossRef
  • Google Scholar

• 231

  LiuYWangCLiXDongLYangQChenMet al. Improved clinical outcome in a randomized phase ii study of anti-Pd-1 camrelizumab plus decitabine in Relapsed/Refractory Hodgkin lymphoma. J Immunother Cancer (2021) 9(4). doi: 10.1136/jitc-2021-002347

  • CrossRef
  • Google Scholar

• 232

  NieJWangCLiuYYangQMeiQDongLet al. Addition of low-dose decitabine to anti-Pd-1 antibody camrelizumab in Relapsed/Refractory classical Hodgkin lymphoma. J Clin Oncol (2019) 37(17):1479–89. doi: 10.1200/JCO.18.02151

  • CrossRef
  • Google Scholar

• 233

  NyLJespersenHKarlssonJAlsenSFilgesSAll-ErikssonCet al. The pemdac phase 2 study of pembrolizumab and entinostat in patients with metastatic uveal melanoma. Nat Commun (2021) 12(1):5155. doi: 10.1038/s41467-021-25332-w

  • CrossRef
  • Google Scholar

• 234

  PalombaMLCartronGPopplewellLRibragVWestinJHuwLYet al. Combination of atezolizumab and tazemetostat in patients with Relapsed/Refractory diffuse Large b-cell lymphoma: Results from a phase ib study. Clin Lymphoma Myeloma Leuk (2022) 22(7):504–12. doi: 10.1016/j.clml.2021.12.014

  • CrossRef
  • Google Scholar

• 235

  SaxenaKHerbrichSMPemmarajuNKadiaTMDiNardoCDBorthakurGet al. A phase 1b/2 study of azacitidine with pd-L1 antibody avelumab in Relapsed/Refractory acute myeloid leukemia. Cancer (2021) 127(20):3761–71. doi: 10.1002/cncr.33690

  • CrossRef
  • Google Scholar

• 236

  TaylorKLoo YauHChakravarthyAWangBShenSYEttayebiIet al. An open-label, phase ii multicohort study of an oral hypomethylating agent cc-486 and durvalumab in advanced solid tumors. J Immunother Cancer (2020) 8(2). doi: 10.1136/jitc-2020-000883

  • CrossRef
  • Google Scholar

• 237

  TopperMJVazMMarroneKABrahmerJRBaylinSB. The emerging role of epigenetic therapeutics in immuno-oncology. Nat Rev Clin Oncol (2020) 17(2):75–90. doi: 10.1038/s41571-019-0266-5

  • CrossRef
  • Google Scholar

• 238

  SuvaMLTiroshI. The glioma stem cell model in the era of single-cell genomics. Cancer Cell (2020) 37(5):630–6. doi: 10.1016/j.ccell.2020.04.001

  • CrossRef
  • Google Scholar