Search

10% of all pediatric brain tumors arise in the brainstem. Amongst these gliomas are the most common while other entities are rare and infrequently described in the literature. In this study we investigated the prevalence of non-gliomatous tumors in the brainstem. Amongst the 1323 embryonal tumours received at the RBTC, we identified 17 cases of ETMRs (17/165) that presented as brainstem primaries. Previously grouped within CNS-PNETs, ETMR, is a new WHO diagnostic entity, characterized by C19MC alterations. ETMR is a disease of infancy, the clinical spectrum of which is poorly understood. ETMRs arise at multiple CNS locations including cerebrum being most common (60%,) followed by cerebellum (18%) and midline structures (6%); notably 10% were brainstem primaries, mimicking DIPG radiologically. All patients presented with a short history of progressive neurological symptoms, with most common signs and symptoms of cranial neuropathies, long tract signs and gait disturbance. Median age at diagnosis was 27 months (range 16-75months) with a male to female ratio of 0.9:1. Predominantly localized (M0-94%, M2-3 -6%) majority of patients underwent upfront biopsy or partial resection (15/17:88%), while complete tumor resection was achieved in 2 cases. All patients received heterogenous combination of chemotherapy with and without radiotherapy. Majority of patients progressed rapidly with median time to progression of 4 months and overall survival of < 13 months. The only long-term surviving patient had complete resection dose intensified chemotherapy and radiation (OS 202months). Primary ETMRs in the brainstem are under recognised entities and carry a dismal prognosis. Although rapidly progressive, prompt recognition, maximal resection and management with multimodal adjuvant therapy should be considered in cases with brainstem disease.

Background Group 3 medulloblastoma (MB) is often accompanied by MYC amplification. PLK1 is an oncogenic kinase that controls cell cycle and proliferation and has been preclinically validated as a cancer therapeutic target. Onvansertib (PCM-075) is a novel, orally available PLK1 inhibitor, which shows tumor growth inhibition in various types of cancer. We aim to explore the effect of onvansertib on MYC-driven medulloblastoma as a monotherapy or in combination with radiation. Methods Crisper-Cas9 screen was used to discover essential genes for MB tumor growth. Microarray and immunohistochemistry on pediatric patient samples were performed to examine the expression of PLK1. The effect of onvansertib in vitro was measure by cell viability, colony-forming assays, extreme limiting dilution assay, and RNA-Seq. ALDH activity, cell-cycle distribution, and apoptosis were analyzed by flow cytometry. DNA damage was assessed by immunofluorescence staining. Medulloblastoma xenografts were generated to explore the monotherapy or radio-sensitizing effect. Results PLK1 is overexpressed in Group 3 MB. The IC50 concentrations of onvansertib in Group 3 MB cell lines were in a low nanomolar range. Onvansertib reduced colony formation, cell proliferation, stem cell renewal and induced G2/M arrest in vitro. Moreover, onvansertib in combination with radiation increased DNA damage and apoptosis compared with radiation treatment alone. The combination radiotherapy resulted in marked tumor regression in xenografts. Conclusions These findings demonstrate the efficacy of a novel PLK1 inhibitor onvansertib in vitro and in xenografts of Group 3 MB, which suggests onvansertib is an effective strategy as monotherapy or in combination with radiotherapy in MB.

Background Medulloblastoma (MB) is a heterogeneous disease in which neoplastic cells and associated immune cells contribute to disease progression. We aimed to determine the influence of neoplastic and immune cell diversity on MB biology in patient samples and animal models. Methods To better characterize cellular heterogeneity in MB we used single-cell RNA sequencing, immunohistochemistry, and deconvolution of transcriptomic data to profile neoplastic and immune populations in patient samples and animal models across childhood MB subgroups. Results Neoplastic cells cluster primarily according to individual sample of origin which is influenced by chromosomal copy number variance. Harmony alignment reveals novel MB subgroup/subtype-associated subpopulations that recapitulate neurodevelopmental processes, including photoreceptor and glutamatergic neuron-like cells in molecular subgroups GP3 and GP4, and a specific nodule-associated neuronally differentiated subpopulation in the sonic hedgehog subgroup. We definitively chart the spectrum of MB immune cell infiltrates, which include subpopulations that recapitulate developmentally related neuron-pruning and antigen-presenting myeloid cells. MB cellular diversity matching human samples is mirrored in subgroup-specific mouse models of MB. Conclusions These findings provide a clearer understanding of the diverse neoplastic and immune cell subpopulations that constitute the MB microenvironment.

Medulloblastoma (MB) is a heterogeneous disease in which neoplastic cells and associated immune cells contribute to disease progression. To better understand cellular heterogeneity in MB we used single-cell RNA sequencing, immunohistochemistry and deconvolution of transcriptomic data to profile neoplastic and immune populations in childhood MB samples and MB genetically engineered mouse models (GEMM). Neoplastic cells clustered primarily according to individual sample of origin which is in part due to the effect of chromosomal copy number gains and losses. Harmony alignment of single cell transcriptomic data revealed novel MB subgroup/subtype-associated subpopulations that recapitulate neurodevelopmental processes and are associated with clinical outcomes. This includes photoreceptor-like cells and glutamatergic lineage unipolar brush cells in both GP3 and GP4 subgroups of MB, and a SHH subgroup nodule-associated neuronally-differentiated cell subpopulation. We definitively chart the spectrum of MB immune cell infiltrates, which reveals unexpected degree of myeloid cell diversity. Myeloid subpopulations include subgroup/subtype-associated developmentally-related neuron-pruning as well as antigen presenting myeloid cells. Human MB cellular diversity is recapitulated in subgroup-specific MB GEMM, supporting the fidelity of these models. These findings provide a clearer understanding of both the neoplastic and immune cell heterogeneity in MB and how these impact subgroup/subtype classification and clinical outcome.

Atypical teratoid/rhabdoid tumors (AT/RT) are characterized by loss-of-function mutations in the SMARCB1 component (and less commonly SMARCA4) of the SWI/SNF chromatin-remodeling complex. AT/RT demonstrate an overall silent genomic landscape with epigenetic dysregulation of the genome. CDK7 is a key transcriptional regulator that preferentially phosphorylates the Ser5 and Ser7 positions on RNA Polymerase C terminal domain and is involved early in transcription. In tumor cells, CDK7 is enriched at super enhancers which preferentially regulate genes involved in cell transformation, and expressed at significantly higher levels in transformed tissues than the surrounding normal brain. Our preliminary data shows that CDK7 is expressed in a number of AT/RT tumor cell lines and patient-derived tumor cultures, and that loss of CDK7 function though exposure to the novel CDK7 inhibitor THZ2 results in lack of proliferation at lower doses, and caspase-mediated apoptosis at higher concentrations. shRNA-based inhibition confirms that this effect is due specifically to loss of CDK7. RNA sequencing of cells treated with lower doses of THZ2 show significant alterations in transcript expression consistent with altered balance between antagonistic SWI/SNF and PRC2 chromatin-modeling complex activities, as well as alterations in DNA damage response pathways, cell cycle checkpoints, miRNA transcription, and numerous proliferative factors. THZ2 penetrates the blood brain barrier (BBB), is well tolerated, and results in prolonged survival in murine xenograft models of AT/RT. CDK7 inhibition also synergizes with a number of currently-approved oncology drugs, as well as with ionizing radiation, in order to inhibit AT/RT growth and viability.

Background Pediatric high-grade gliomas (pHGG) are aggressive tumors that together constitute the most common cause of childhood cancer mortality. Tumor stem cells that drive proliferation of pHGG resist chemotherapy and radiation, complicating treatment. The arginine methyltransferase PRMT5 maintains self-renewal in neural stem cells through epigenetic modifications. We hypothesized that PRMT5, which we identified as a potential driver of diffuse midline glioma (DMG) through an shRNA screen, plays a similar role in pHGG. Methods Using lentiviral delivery of shRNA, we knocked down (KD) PRMT5 in cortical pHGG and DMG cell lines and performed phenotypic, mechanistic and self-renewal assays. We irradiated PRMT5 KD and control cells to study sensitization. We orthotopically injected mice with PRMT5 KD pHGG cells, and with DMG cells in which PRMT5 was knocked out (KO) using CRISPR-Cas. Results In cellular models of cortical pHGG and DMG, PRMT5 KD significantly reduced proliferation, inhibited cell cycle progression, increased apoptosis resistance, and decreased self-renewing cell frequency. A relative shift of PRMT5 from the cytoplasm to the nucleus accompanied differentiation induced by PRMT5 KD. Epigenetic changes accompanying PRMT5 KD included increased H3K27me3, a global transcription inhibitor, and decreased H3K27M expression in DMG. PRMT5 KD sensitized pHGG cells to radiation, increasing cell death 17–30%. PRMT5 KD/KO significantly increased survival in mice and decreased tumor aggressiveness and proliferation, but mice still died of tumor-related effects. Conclusions PRMT5 maintains self-renewal and drives proliferation in preclinical pHGG models. In cellular and in vivo models, PRMT5 KD/KO produces epigenetic changes, including increased H3K27me3 levels and diminished H3K27M, that may reduce proliferation and self-renewal. Future work includes elucidation of the mechanisms by which PRMT5 produces the observed changes. Because PRMT5 KD/KO does not eliminate tumor growth, we plan to further study combining PRMT5 KD/KO and clinical-grade small molecule PRMT5 inhibitors with radiation and chemotherapeutic agents.

Our previous studies have shown that the epigenetic developmental regulator EZH2 is overexpressed in medulloblastoma (MB). EZH2 is known to repress developmental genes in embryonic stem cells to maintain their stem cell characteristics. To identify the role of EZH2 in medulloblastoma, we overexpressed EZH2 (EZH2(Ox)) in the C17.2 Myc immortalized cerebellar neural stem cell line. EZH2 overexpression in combination with Myc in the neural progenitor C17.2 cells generated orthotopic tumors that resembled clinical features of MYC-amplified group3 MB. To examine the mechanism of EZH2 mediated tumorigenesis we performed RNA-seq followed by Gene Set Enrichment Analysis (GSEA) of these EZH2 overexpressing neural stem cells. Transcriptomic analysis reveals that the gene sets associated with apoptosis were suppressed by EZH2 overexpression. Specifically, the p53 regulator and pro-apoptotic gene, HIPK2 was significantly suppressed by EZH2 overexpression. To examine the impact of EZH2 on apoptosis in Myc driven medulloblastoma cells we performed a CRISPR knock out of EZH2 in the D458 group 3 MB cell line. EZH2 deletion resulted in increased apoptosis leading to sensitization of cells to chemotherapy and radiation treatments. Further analysis of these CRISPR-edited EZH2 knockout cells showed a predominant upregulation in the HIPK2 expression compared to control D458 cells. While, overexpressing HIPK2 in D458 cells resulted in increased apoptosis after cisplatin or ionizing radiation, depleting HIPK2 in EZH2 CRISPR-KO cells rescued the proapoptotic phenotype. Treating mice with cisplatin resulted in significantly smaller orthotopic xenograft tumors of EZH2 knockout medulloblastoma cells compared to EZH2 sufficient medulloblastoma cells. Together, our results suggest that EZH2 contributes to treatment-resistance by inhibiting apoptosis via transcriptionally modulating HIPK2 expression in group 3 medulloblastoma. Therefore, targeting EZH2 will effectively sensitize tumors to standard therapies.

Myc-driven Medulloblastoma remains a major therapeutic challenge due to frequent metastasis and a poor 5-year survival rate. Myc overexpression results in transcriptional dysregulation, proliferation, and survival of malignant cells. However direct inhibition of Myc remains an obstacle to therapeutic interventions. To identify therapeutic targets in Myc amplified medulloblastoma we performed a CRISPR-Cas9 essentiality screen targeting 1140 genes annotated as the druggable genome. A strand guide library (4–10 Sg/gene) was transduced along with Cas9 into 3 different Myc amplified medulloblastoma cell lines in triplicate and DNA extracted and sequenced 18 days after transduction. Analysis of the resultant data revealed multiple gene networks as essential for Myc-medulloblastoma. The cyclin-dependent kinase, CDK7, was identified as a top candidate. CDK7 phosphorylates the c-terminal domain of RNA Pol II facilitating transcriptional initiation and elongation. We found that CDK7 expression is elevated in Myc-medulloblastoma patient samples and cell lines compared to other subgroups and normal pediatric cerebellum. Genetic depletion of CDK7 suppressed medulloblastoma cell clonogenicity in vitro. Recently a covalent inhibitor, THZ1, with high potency for CDK7 was identified and shown to efficiently reduce tumor growth in several pediatric cancers. To evaluate whether CDK7 is a therapeutic target in Myc-driven medulloblastoma we treated Myc-amplified medulloblastoma cells with THZ1. We observed a significant inhibition of clonogenic potential and induction of apoptosis in Myc-medulloblastoma cells in vitro. Using RNA-seq-based gene set enrichment analysis we show that THZ1 decreases c-Myc driven transcriptomic programs and enhanced inflammatory and apoptotic responses. Further studies established that chemical inhibition of CDK7 enhanced radiation sensitivity of Myc MB cells by potentiating DNA damage. Furthermore, CDK7 inhibition with THZ2 suppresses medulloblastoma tumor formation in vivo. Our studies are the first demonstration of CDK7 inhibition as a viable therapeutic strategy alone or in combination with radiation for Myc-amplified medulloblastoma.

Mutations in the histone 3 (H3) gene (H3K27M) are the eponymous driver in diffuse intrinsic pontine gliomas (DIPGs), aggressive pediatric brain tumors for which no curative therapy currently exists. In order to identify specific epigenetic dependencies which arise as a consequence of the H3K27M mutation, we performed an shRNA screen targeting 408 genes classified as epigenetic/chromatin-associated molecules in patient-derived DIPG cell lines. This identified AFF4, a component of the super elongation complex (SEC), as both highly expressed and necessary for DIPG cells to maintain growth and self-renewal. We hypothesized that SEC overexpression occurs as a direct consequence of the H3K27M mutation and that this relative abundance overcomes repressive transcriptional regulation in order to suppresses differentiation and promote self-renewal of DIPG tumor stem cells. We found that AFF4 expression is consistently elevated in both DIPG patient samples and established cell lines relative to normal pediatric pons. We interrogated the role of AFF4 in H3K27M-mutant DIPG using an shRNA lentiviral approach. Using live cell imaging, we demonstrate a significant decrease in in vitro clonogenicity and stem cell maintenance following AFF4 depletion. We employed RNA-seq-based gene set enrichment analysis to delineate differentiation programs under AFF4 regulatory control. Finally, we sought to determine whether CDK9, the catalytic subunit of the SEC, represents a therapeutic vulnerability in DIPG. Using ChIP-seq following CDK9 pharmacologic inhibition, we demonstrate that the loss of regulatory transcriptional input in DIPG may be reversed via CDK9 inhibition and that this acts as a potent inhibitor of DIPG cell growth both in vitro and in vivo orthotopic xenograft models. These studies represent the first pre-clinical validation of SEC inhibition as a novel therapeutic approach in DIPG.

Adamantinomatous craniopharyngioma (ACP) makes up between 6–8% of pediatric brain tumors and is the most common pediatric tumor arising in the sellar/suprasellar region of the brain. The 10-year survival for patients diagnosed with craniopharyngioma ranges between 64–92%, but given complicating factors such as its location, common cyst formation, and potential hypothalamic infiltration, there is significant morbidity in this population. There are a number of therapy options for children with craniopharyngioma including surgery, radiation, and cyst-directed therapies such as Interferon and Bleomycin. Research has raised concerns regarding efficacy and morbidity induced side effects with these conventional therapies and with difficulty in treating recurrent primary cystic ACP. Evidence from our group and others have shown that the cystic and solid tumor components of craniopharyngioma have high levels of IL-6R and IL-6, providing a potential target for therapy. Tocilizumab acts against soluble and membrane bound IL-6R, leading to the possibility that it may offer benefit for some patients with ACP. Two patients with recurrent primary cystic craniopharyngiomas treated at Children’s Hospital Colorado were offered systemic administration of the IL-6R antagonist, tocilizumab or a combination of tocilizumab and bevacizumab on a compassionate use basis. Both patients’ tumors had a significant response, with decreased cystic burden, supporting the assertion that tocilizumab and bevacizumab may be an option for patients suffering from cystic craniopharyngioma.

BACKGROUND Diffuse intrinsic pontine gliomas (DIPGs) driven by mutations in the histone 3 (H3) gene (H3K27M) are aggressive pediatric brain tumors for which there is no curative therapy. METHODS To identify novel therapeutic targets we performed a high throughput drug screen combined with an epigenetically targeted RNAi screen using H3K27M and H3.3 WT DIPG cells. RESULTS Chemical and genetic depletion of BMI1 in vitro resulted in inhibition of clonogenicity and cell self-renewal consistent with previous studies. We show for the first time that clinically relevant BMI1 inhibitors attenuates growth of orthotopic DIPG xenografts as measured by MRI and prolong survival in vivo. We found that BMI1 inhibition drives phenotypic cellular senescence and that the senescent cells were able reactivate to form new neurospheres in vitro and tumor growth in vivo. RNA-seq, ChIP-Seq and immuno-proteomic analysis revealed that the senescent cells induced the expression of the Senescence Associated Secretory Phenotype (SASP) cytokines by increasing occupancy of activated histone marks at SASP factor promoters. The SASP results in increased expression of anti-apoptotic BH3 proteins including BCLxl, and BCL2. Treatment of the PTC028 treated senescent DIPG cells with BH3 mimetics induces apoptosis and clears the senescent cells. Combining BH3 mimetics with BMI1 inhibition attenuates tumor growth in vivo synergistically and significantly prolongs survival of DIPG bearing mice compared to BMI1 inhibition alone. CONCLUSION These data inform the current trial of BMI1 inhibition as a monotherapy and predict the need for adding BH3 mimetics to achieve efficacy.

BACKGROUND Previously, we have reported on the development of advanced magnetic resonance imaging (MRI) protocols for mouse brain tumors. The goal of this follow-up pre-clinical study was to develop a machine-learning MRI classifier (radiomics) for four subtypes of childhood brain tumor in patient-derived xenograft (PDX) mice. METHODS MRI scans on orthotopic medulloblastoma, ependymoma, ATRT and DIPG PDX (each n=12 animals) were performed on the animal 9.4 Tesla scanner with an in-plane resolution of 47 microns. Image segmentation, as well as shape and texture based radiomics descriptors were modeled using a modified COLIAGE software for tumor classification and to characterize tumor habitat of each tumor subtype. RESULTS The mean tumor volumes were 11.2 mm3. Each MRI scan was segmented into three regions: (i) well defined tumor (including distant metastases); (ii) peritumoral edema; (iii) tumor necrosis. 360 radiomics features (capturing co-occurrence, grey-level dependence and directional gradients) were obtained for each region. The model classified four subtypes with high accuracy while achieving sufficient segmentation accuracy despite the small lesion size. A subset of fourteen tumoral, six peritumoral and five distant MRI radiomics features were found to be predictive of the tumor sub-type (p=0.0017) independently of tumor anatomical location. CONCLUSIONS MRI protocols followed by radiomics feature analysis discriminated among specific radiological features for four distinct orthotopic PDX models: medulloblastomas exhibit low ADC values, high angiogenesis and cortical metastases as compared to ependymomas (high levels of edema and olfactory bulb metastases), ATRT (the highest level of necrosis) and DIPG (highest T2 signal intensities and spinal metastases).

BACKGROUND Diffuse intrinsic pontine glioma (DIPG) and diffuse midline glioma (DMG) are metastatic diseases, as demonstrated by early convection-enhanced delivery (CED) clinical trials in which prolonged local tumor control can sometimes be achieved, but fatal disseminated disease then develops. We hypothesize that improvements in treatment of both focal disease and the entire neuraxis are necessary for long-term survival, and patient-derived xenograft (PDX) models can help advance these efforts. METHODS We used a BT245 murine orthotopic DIPG PDX model for this work. We developed a protocol and specialized platform to deliver craniospinal irradiation (CSI) with a pontine boost. We separately compared intratumoral drug concentration by CED and intraperitoneal delivery. In our CED model, mice receive gemcitabine 60 ug x1 in 15 ul at 0.5 ul/minute through a stepped catheter design with silica tubing extending 2mm beyond a 27G needle. RESULTS Mice receiving CSI (4 Gy x2d) plus boost (4 Gy x2d) showed minimal spinal and brain leptomeningeal metastatic disease by bioluminescence, MRI, and pathology compared to mice receiving radiation to the pons only (4 Gy x4d) or no radiation. CED achieved an intratumoral gemcitabine concentration 50-fold greater than intraperitoneal dosing when controlled for dose. CONCLUSIONS In a DIPG PDX model, CSI+boost minimizes tumor dissemination compared to focal radiation, and CED achieves clinically significant improvements in intratumoral chemotherapy concentration compared to systemic delivery. Adding these modalities to current treatment could improve both focal and metastatic tumor control, leading to meaningful improvements in survival.

Medulloblastoma (MB), a central nervous system tumor that predominantly affects children, requires aggressive therapy. Recent advances in the noncoding RNA genome could contribute to the sub-classification of medulloblastoma. The focus of this study is to identify novel long noncoding RNAs (lncRNAs) as molecular markers and potential therapeutic targets within each subgroup of MBs, in particular within Group 3. We analyzed publicly available 175 RNA-seq datasets to identify a group of putative lncRNA signatures that may be able to differentiate medulloblastoma subgroups accurately. Among those, lncRNA lnc-HLX-2–7 was highly upregulated in Group 3 MB cell lines, patient-derived xenografts, FFPE samples compared to other groups. CRISPR/Cas9 deletion of the lnc-HLX-2–7 followed by the fluorescence-activated sorting and generating monoclonal Group 3 MB cells significantly reduced the cell growth and 3-D colony formation together with the induction of apoptosis. Intracranial injection to mouse cerebellum using lnc-HLX-2–7 deleted cells resulted in reduced tumor growth compared to parental cells, and tumors were further characterized by single-cell sequencing. We identified that oncogene MYC regulates lnc-HLX-2–7 and its expression can be controlled by the small molecule JQ1, a BET-bromodomain (BRD4) inhibitor that disrupts interactions with MYC. RNA-FISH analysis using FFPE, PDX, and tissue microarrays revealed that lnc-HLX-2–7 expression is specific to Group 3 MB compared to other groups. We present supporting evidence that lnc-HLX-2–7 is a novel molecular marker and a potential therapeutic target for Group 3 MBs in children.

Mutations in the histone 3 gene (H3K27M) are the eponymous driver in diffuse intrinsic pontine gliomas (DIPGs) and other diffuse midline gliomas (DMGs), aggressive pediatric brain tumors for which no curative therapy currently exists. To identify specific epigenetic dependencies within the context of the H3K27M mutation, we performed an shRNA screen targeting 408 genes classified as epigenetic/chromatin-associated molecules in patient-derived DMG cultures. This identified AFF4, a component of the super elongation complex (SEC), as necessary for DMG cells to maintain growth and self-renewal. We hypothesized that aberrant SEC expression occurs as a consequence of the H3K27M mutation and that this dysregulated SEC signaling overcomes repressive transcriptional regulation in order to suppresses differentiation and promote self-renewal of DMG tumor stem cells. We interrogated the role of AFF4 in DMG using an shRNA lentiviral approach. We demonstrate a significant decrease in in vitro clonogenicity and stem cell maintenance following AFF4 depletion. We employed RNA-seq-based gene set enrichment analysis to delineate differentiation programs under SEC regulatory control. Finally, we sought to determine whether CDK9, the catalytic subunit of the SEC, represents a therapeutic vulnerability in DMG. Using RNA polymerase II ChIP-seq, we demonstrate that CDK9 pharmacologic inhibition restores regulatory Pol II pausing, promotes cellular differentiation, and leads to potent anti-tumor effect both in vitro and in patient-derived xenograft models. These studies present a biologic rationale for the translational exploration of CDK9 inhibition as a promising therapeutic approach.

Group 3 medulloblastoma (MB) is often accompanied by MYC amplification and has a higher rate of metastatic disease. So, it is critical to have more effective therapies for high MYC expressing sub-groups. Here we report that FBXW7, a substrate recognition component of the SKP1-CUL1-Fbox (SCF) E3 ligase, interacts with and targets c-MYC for polyubiquitination and proteasomal degradation. FBXW7 shows lower expression level in MYC-driven MB compared with other MB subgroups suggesting activity as a tumor suppressor. Genomic deletion or mutation of Fbxw7 has frequently been identified in many human cancers but not in MB. We demonstrate that overexpression of Fbxw7 in MB cells induces apoptosis and suppresses proliferation in vitro and in vivo. Both phospho-deficient (T205A) and phosphomimetic aspartic acid (T205D) mutants deactivate its tumor suppressor function suggesting a conformational change of its protein structure. Mechanistically, PLK1 kinase specifically phosphorylates FBXW7 and promotes its auto-polyubiquitination and proteasomal degradation, counteracting FBXW7-mediated degradation of oncogene substrates, including c-MYC and PLK1. Chip-Seq results show stabilized c-MYC in turn directly activates PLK1 and FBXW7 transcription, constituting a feedforward regulatory loop. Co-immunoprecipitation demonstrates that FBXW7 directly binds to PLK1 and c-MYC, facilitating their protein degradation by promoting the ubiquitination of both proteins. Furthermore, we show that FBXW7 protein can be stabilized by various kinase inhibitors, proposing a mechanism of kinase-targeted agents to treat MYC-driven MB. These results collectively demonstrate how kinase inhibition stabilizes the tumor suppressor FBXW7 in MYC-driven MB, thus revealing an important function of FBXW7 in suppressing MB progression.

BACKGROUND Cannabinoids, including cannabidiol (CBD) and tetrahydrocannabinol (THC), are a class of compounds found in marijuana. Numerous studies in adults have examined cannabinoid use in management of cancer-related symptoms such as nausea, anorexia, and pain. Less is known about the use in the pediatric oncology population. METHODS A prospective observational study has been ongoing since 2016 at Children’s Hospital Colorado to evaluate cannabinoids’ impact using PedsQL™ modules on quality of life of pediatric patients with central nervous system (CNS) tumors who are 2–18 years old. Laboratory assessments of T-cell activity and pharmacokinetics of CBD, THC and associated metabolites are in process. Diaries with exploratory information on cannabinoid use patterns are being collected. RESULTS Thirty-three patients (14:19; male:female) have been enrolled with a median age of 6.4 years (range, 2.9–17.7 years). The most common tumor type in enrolled patients is embryonal tumors (13/33; 39%). Nine (27%) patients have low-grade glial/glioneuronal tumors, and eight (24%) had high-grade/diffuse midline gliomas. The remaining patients had ependymoma or craniopharyngioma. The median time on cannabinoids is 9 months. Most (n=20) patients have used oral products with CBD and THC. One patient continues on cannabinoid therapy in follow up. Preliminary immune function analyses identified impaired neutrophil superoxide anion production and chemotaxis in patients taking cannabinoids at early time points on therapy. CONCLUSIONS Families of children with various CNS tumors are pursuing cannabinoid therapy for both antitumor and supportive care purposes. Analysis of the impact of cannabinoids on patients’ quality of life is ongoing.

Chemotherapy has yet to show reproducible therapeutic benefit in ependymoma, and this is partly due to a lack of models for preclinical testing. Treatment of ependymoma has changed little over the last 40 years and is essentially limited to surgery and radiation. Our group has published establishing 2 posterior fossa 1q+ ependymoma cell lines which provides a model for high throughput testing of therapeutic treatments. With the goal of immediate clinical translation, we screened these cell lines for ependymoma-selective drug sensitivity across a panel of 124 FDA-approved oncology drugs. Treatment effect was measured using a tritiated-thymidine incorporation proliferation assay and ependymoma-selective sensitivity was determined by comparison to a panel of other pediatric brain tumor types (medulloblastoma, glioblastoma, DIPG and AT/RT). Upon screening we identified retinoids bexarotene, tretinoin and isotretinoin as effective and ependymoma-selective. Retinoid Acid Receptors alpha and beta (RARɑ and RARβ) and Retinoid X Receptor alpha (RXRɑ) were overexpressed in cell lines and primary samples versus other pediatric brain tumors and normal brain. We examined the effect of retinoids on ependymoma using transcriptomic analysis which revealed Retinoic Acid Receptor Responder 1 (RARRES1) as the most upregulated gene in bexarotene-treated EPN cell lines. RARRES1 is a known tumor suppressor gene described in a variety of human cancers and has been shown to repress mitogen-activated protein kinase (MAPK) activation. We aim to further investigate the mechanism by which retinoids inhibit ependymoma cell growth. These findings may support further clinical evaluation of retinoids as effective chemotherapeutics for ependymoma.

Atypical Teratoid Rhabdoid Tumor (ATRT) is a highly aggressive and malignant pediatric brain tumor with poor outcomes. The central mechanism of ATRT tumorigenesis is SMARCB1 deficiency that leads to epigenetic dysregulation. To identify key epigenetic regulators that co-operate with SMARB1 deficiency and contribute to the tumorigenesis of ATRT, we performed an RNAi-based functional genomic screen. We targeted 406 genes associated with epigenetic regulation in ATRT cell lines. This screening revealed Sirtuin2 (SIRT2) as one of the top targets required for continued cell growth. SIRT2 is an NAD(+) (nicotinamide adenine dinucleotide)-dependent deacetylase and involved in the direct deacetylation of α-tubulin and histone H4. Inhibition of SIRT2 promotes c-Myc ubiquitination and degradation. C-Myc is an oncogene and highly expressed in ATRT, so SIRT2 is a promising target to treat c-Myc driven cancers. Genetic inactivation of SIRT2 with shRNA significantly suppressed ATRT cell growth, decreased c-Myc expression and activated p53. Recent studies suggest that targeting SIRT2 by small molecule inhibitors is a promising approach to tumor therapy. We found that chemical inhibition SIRT2 with AGK2,Tenovin6 and Thiomyristoyl (TM) significantly decreased cells growth, inhibited clonogenic potential, suppressed c-Myc level, and decreased S-phase length. In vivo, Tenovin6 and TM limited ATRT tumor growth both in flank and in intracranial mouse model due to increased apoptosis as measured by cleaved caspase 3 and decreased ki67 proliferative index. Moreover, TM treatment increased mice survival. In summary, we demonstrated that ATRT can be targeted by chemical or genetic SIRT2 inhibition.

INTRODUCTION: Atypical teratoid/rhabdoid tumor (ATRT) comprises 20% of all brain tumors in children less than 3 years old and is incurable at recurrence. The aim of our study is to determine transcriptional differences that may predict recurrence and raise more effective therapeutic strategies at recurrence. METHODS: We performed microarray RNA expression profiling on three primary ATRT samples from long-term recurrence-free survivors (two from the SHH and one from the MYC epigenetic subgroup) and three sets of matched primary and recurrent patient ATRT samples (two MYC and one SHH). RNA-Seq was also performed on the primary/recurrent sets. We conducted clustering and geneset enrichment analysis (GSEA) on the microarray data. RESULTS: Clustering analysis identified two subgroups, with the first group containing the three non-recurrent tumors and the six samples from the primary/recurrent pairs in the second (coph=1.0). GSEA demonstrated upregulation of the Myc oncogenic pathway (Hallmark Myc geneset, NES=1.53, FDR=0) and upregulated tumor invasion pathways (Hallmark EMT geneset, NES=1.0, FDR=0.09) in primary tumors that recurred compared to those that did not. In recurrent tumors versus their primary pairs, GSEA showed upregulation of Myc pathway genes (Hallmark Myc geneset, NES=3.68, FDR=0), epigenetic regulators (BMI1 geneset, NES=3.15, FDR=0), and neuronal genes (NES=2.93, FDR=0.0), and downregulation of PTEN (NES=2.36, FDR=0.002). CONCLUSIONS/CURRENT WORK: Primary ATRTs that will eventually recur show upregulation of genes associated with oncogenesis and tumor invasion as compared to non-recurrent tumors. At recurrence, these tumors show further upregulation of Myc pathway genes. RNA-Seq analysis is currently in process.

KDM6A and EZH2 are epigenetic regulators that play a key role in embryogenesis and development. Presumably, EZH2 and KDM6A have opposite impact on H3K27me3. KDM6A functions as a demethylase of Histone3 lysine 27 trimethylation and is closely associated with the mixed lineage leukemia 2/3 in creating a transcriptionally active chromatin state through removal of methyl groups. EZH2, on the other hand, is a methyl transferase which catalyzes the trimethylation of Histone3 lysine 27. Several studies including ours have shown that KDM6A is genomically deleted in Group 4 medulloblastoma suggesting it may have a role as a tumor suppressor. The biological role of KDM6A in medulloblastoma is not clear. To better understand the role of KDM6A in MB, in our current study we used the Crispr/Cas9 system to knock out KDM6A in MB cell lines and to determine the effect of KDM6A on MB tumor cell stemness and proliferation. Increased expression of stemcell-associated genes including Nanog, Nestin and Oct4 were observed in cells depleted of KDM6A. Intriguingly, Crispr/Cas9-mediated deletion of KDM6A significantly abolished MB cell neurosphere formation and cell proliferation compared to control cells. As expected, in the knockdown cells there was a concomitant increase in H3K27me3 and decrease in H3K4me3 expression but the overall biological impact of KDM6A deletion was unexpected. This puzzling effect on cells with the knockdown of KDM6A opens a door to investigate the target genes that are affected by the depletion of KDM6A. This study will shine light on new epigenetic therapies in medulloblastoma.

Advances in ependymoma (EPN) treatment have been hampered due to lack of relevant in vitro and in vivo models. Our lab successfully generated patient derived posterior fossa group A (PFA) EPN cell lines. Our present goal is to establish patient derived xenograft (PDX) mouse models. Disaggregated cells from a primary PFA tumor (811) were injected subcutaneously into NSG mice flanks forming tumors 8 months post injection. Tumors were confirmed as EPN by typical histologic features by a neuropathologist. This flank tumor was successfully used to establish monolayer cell lines. 850K methylation array on flank tumor showed the same chromosome abnormalities as the primary tumor, including gain of 1q. Transcriptomic clustering analysis showed that flank tumor grouped with PFA EPN primary tumors. Flank tumor cells were disaggregated into single cell suspension and serially injected both subcutaneous and intracranial. Serial flank tumors formed within 2 months of injection. MRI imaging of one intracranial injected mouse showed a large mass in the cerebellum. Additionally, before MRI scans could be performed, a second symptomatic mouse with intra-ventricle injection was euthanized. Necropsy revealed an encapsulated mass between the midbrain and cerebellum and histology confirmed classic EPN features. This mass was serially transplanted into lateral ventricle of several additional mice. In addition to MAF811_5, we have preliminary evidence, that needs to be confirmed, that 2 further EPN tumors have established flank xenografts and cell lines. These PDXs are promising in showing the feasibility of establishing PFA xenograft models together with matching cell lines.

Ependymoma (EPN) is the third most common pediatric central nervous system tumor, associated with poor survival and long-term morbidity. Posterior fossa Group A EPN (PFA) tumors are chemotherapy insensitive and often fatally recur. There is a critical need for new treatments and an untested avenue is targeting of epigenetic modifiers. Our lab has previously shown that the epigenetic suppression of a protein, LDOC1, in PFA drives the NF-kB pathway and constitutive secretion of IL-6. We believe that by targeting specific epigenetic modifiers we can reverse LDOC1 silencing and disrupt this oncogenic pathway in EPN PFA. Using two EPN PFA cell lines, we performed shRNA epigenetic screen identifying key epigenetic regulators including NAD-dependent deacetylase sirtuin 2 (SIRT2). SIRT2 actively inhibits transcription and may silence key tumor suppressing genes. Next, using an epigenetic drug screen we showed that SIRT2 inhibitor AGK2 (2-cyano-3-[5-(2,5-dichlorophenyl)-2-furanyl]-N-5-quinolinyl-2-propenamide) reduced proliferation in cell lines. Inhibition of SIRT2 by AGK2 recovered LDOC1 and down regulated many NF-kB genes. In addition, treatment of an EPN PFA organotype culture with AGK2 re-established LDOC1 and down-regulated IL-6 gene expression. We are now targeting SIRT2 in-vivo in murine sub-cutaneous flank and orthotopic intracranial models of EPN using the compound thiomyristoyl lysine (TM), a SIRT2 inhibitor that is available for in-vivo treatments. These studies suggest that targeting epigenetic modifiers, like SIRT2, can be used as a therapy to target EPN PFA.

Medulloblastoma (MB) is the most common malignant brain tumor in children. Despite treatment with surgery, radiation, and chemotherapy, outcomes are unsatisfactory. MB is a heterogenous disease with four major and distinct molecular subgroups (WNT, SHH, Group 3 and Group 4) characterized by unique genomic and clinical features. Group 3 tumors have the worst outcomes and are frequently characterized by MYC amplification. Among the features of many of the group 3 tumors are the overexpression of EZH2, gains of the chromosome 7 and increased global H3K27me3 chromatin-repressive marks. EZH2, the histone methyltransferase responsible for H3K27me3 and located on 7q36.1 is critical during embryonic development to establish cell identities. In many cancers, both oncogenic and tumor suppressor functions have been observed for EZH2. More studies are required to better understand the role of EZH2 overexpression in medulloblastomas. Using RNAi, chromatin immunoprecipitation and pharmacological approaches, we were able to show that EZH2 inhibition impairs the pluripotency and self-renewal properties of human medulloblastoma cells and reactivates a differentiation program associated with differentiated cerebellar cells. Combinatorial drug screenings have been developed to identify drugs synergizing with EZH2 inhibition in human MB cell lines. Using human cerebellar stem cells and medulloblastoma cell lines, we also evaluated the effects of both EZH2 overexpression and pharmacological inhibition in the context of MYC overexpression in immunocompromised mice. The results of our in vitro and in vivo experiments will be presented.

INTRODUCTION: Hundreds of chemotherapy trials have been conducted in diffuse intrinsic pontine glioma (DIPG) without survival improvement. One potential reason is a lack of intratumoral penetration, but this has not previously been assessed in humans. We used gemcitabine as a model agent to assess DIPG intratumoral pharmacokinetics (PK). We chose gemcitabine for its known blood-brain barrier penetration and established pediatric safety data. METHODS: We first assessed intratumoral PK in a mouse orthotopic xenograft model of DIPG. We subsequently opened a phase 0 trial in which newly-diagnosed DIPG patients receive one intravenous gemcitabine dose immediately before biopsy. We measured gemcitabine concentration by liquid chromatography-tandem mass spectrometry. RESULTS: In the xenograft model, intraperitoneal gemcitabine showed a significantly greater tumor/normal brain penetration ratio in cortical compared to pontine tumors (p=0.032). Based on these results, we began our phase 0 trial. In the first patient enrolled, concentrations from four spatially heterogeneous tumor biopsies ranged from 4.8–9.7 µM. These compare favorably to intratumoral gemcitabine levels published from a similar study in adult glioblastoma (0.06–3.6 µM) and to our measured in vitro IC(50) levels in three DIPG cell lines (44–162 nM). Levels were similar to those found in our xenograft pontine tumors (3.4–7.3 µM, p=0.24). CONCLUSIONS/CURRENT WORK: We demonstrated adequate intratumoral chemotherapy penetration for therapeutic effect in DIPG, suggesting that this is not the reason for chemotherapy failure. We are now analyzing samples from a second trial subject, and from a xenograft study comparing penetration and clearance rate based on H3K27M mutant/wild-type status and tumor location.

Mutations in the histone 3 (H3) gene (H3K27M) are the eponymous driver in diffuse midline gliomas (DMGs). Emerging understanding of H3K27M biology suggests that secondary cooperating factors are required to promote oncogenesis. We have identified AFF4, a component of the super elongation complex (SEC), as critical for DMG viability. We hypothesized that AFF4 promotes DMG tumorigenesis by co-operating with the H3K27M mutation to drive a gene expression program which suppresses scheduled transcription of pro-differentiation pathways and promotes self-renewal of tumor stem cells. We initially performed an shRNA screen targeting 408 genes classified as epigenetic/chromatin-associated molecules in a patient-derived DMG cell line. We validated initial hits in secondary screens using multiple shRNA constructs. Live cell imaging systems were utilized to assay growth and neurosphere formation. Finally, we employed RNA-seq-based gene expression analysis to delineate downstream transcriptional changes under AFF4 regulatory control. We found that AFF4 expression is consistently elevated in DMG lines relative to normal pons, and AFF4 inhibition results in decreased rates of growth and stem cell maintenance uniquely within the context of the H3K27M mutation. Further mechanistic studies will lay the groundwork for pre-clinical validation of SEC inhibition as a novel therapeutic approach in DMG.

Ionizing Radiation (IR) is a key treatment modality for DIPG, but it provides only temporary relief as the tumor cells develop resistance to radiation. Recently, we and others have shown that inhibition of BMI1 either alone or in combination with radiation attenuates DIPG cell proliferation in vitro. While we are demonstrating the in vivo efficacy of pharmacological inhibition of BMI1 and understanding the mechanism of anti-tumor effect of BMI1 inhibition in DIPG, the existence of treatment-resistant cells remains a major obstacle for a prolonged cure. Both IR and genetic or pharmacological inhibition of BMI1 induces cellular senescence as a mechanism to suppress tumor cell proliferation, implying that senescence can be considered as tumor suppressor. Paradoxically, recent studies have shown that accelerated senescence can mediate tumor recurrence due to the development of pro-oncogenic environment. In line with this, we are investigating whether clearance of treatment-induced senescent cells enhances treatment outcomes. DIPG cells exposed to different doses of radiation followed by treatment with ABT-263 (Navitoclax), a drug which selectively clears the senescent cells, resulted in increased radiosensitization. Treatment of pre-radiated DIPG cells with ABT-263 decreased the activity of senescence-associated beta-galactosidase and anti-apoptotic protein expression. Similarly, chemical inhibition of BMI1 in combination with ABT-263 showed synergistic killing of DIPG cells. The synergy was most pronounced in DIPG cells harboring wildtype p53. Our study highlights the importance of eliminating treatment-induced senescent cells while inhibiting proliferation of DIPG tumors, a combination which can immensely improve therapeutic efficacy in DIPG patients.

INTRODUCTION: Genomic assessment of BRAF V600E has become routine in specific pediatric glioma subtypes in recent years given availability of targeted therapies. Comparison of imaging characteristics of pediatric gliomas with or without the V600E mutation has not been previously reported. METHODS: Pathology reports of pediatric brain tumors diagnosed from 2010-2017 at a single institution were searched for keywords “V600E” or “VE1” using institutional databases (Nuance mPower, Nuance Communications). Cases where V600E testing was successfully performed and whose pathologic diagnosis was ganglioglioma, pleomorphic xanthoastrocytoma (PXA) or epithelioid glioblastoma were included. Preoperative MRI studies were reviewed by a single neuroradiologist (NVS) for tumor volume, T1 and T2 signal characteristics, cystic or solid components, enhancement, adjacent non-enhancing T2 hyperintensity, hemorrhage, DWI signal and ADC values in solid components, and location. Review of charts and follow-up imaging studies was conducted to evaluate tumor recurrence or progression. Fisher’s exact test was used for categorical data and a two-tailed unpaired t-test was used for continuous data. RESULTS: 30 tumors met pathologic inclusion criteria, including 18 gangliogliomas, 10 PXAs and 2 epithelioid glioblastomas. Average age was 10.6 years (range 0.64-24 years). IHC or molecular analysis identified BRAF V600E in 16 tumors. These tumors were significantly more likely to exhibit hyperintensity on DWI (p=0.009), lower ADC values (p=0.046), and non-enhancing T2 hyperintensity adjacent to solid enhancing components. CONCLUSIONS: BRAF V600E mutant tumors demonstrate restricted diffusion, which may indicate differences in cellularity or tumor microarchitecture. Further study with advanced imaging sequences, including MR spectroscopy and perfusion, is needed.

Reproducibly effective chemotherapy has yet to be identified for ependymoma (EPN), partly due to a lack of models for preclinical drug screening. Treatment of EPN has therefore changed little over the last 40 years and is essentially limited to surgery and radiation. Recurrence risk is high, particularly in those EPN harboring chromosome 1q gain. To address this, we have established 2 posterior fossa 1q+ EPN cell lines and we have utilized these in a high-throughput drug screen of FDA-approved oncology drugs. Treatment effect was measured using a tritiated-thymidine incorporation proliferation assay and EPN-selective sensitivity was determined by comparison to a panel of other pediatric brain tumor types (medulloblastoma, glioblastoma, DIPG and AT/RT). Using this approach, we identified a group of 3 small molecule tyrosine kinase inhibitors (TKIs) that showed greater potency in EPN than control cell lines. These TKIs – namely axitinib, imatinib and pazopanib – commonly target VEGFR, PDGFR and c-kit. Of note, axitinib has been proposed for an upcoming phase 1 trial for recurrent pediatric brain tumors including EPN. Studies of axitinib in other tumors demonstrated a beneficial immunomodulatory effect, of particular relevance in EPN given the key role of immunity in this tumor. Axitinib may therefore have a dual mechanism of tumor control in EPN. We selected this drug for further evaluation in our preclinical model of EPN. Dose curve analysis confirmed that axitinib more is potent in EPN than other cell types. Transcriptomic analysis of axitinib treated EPN revealed a decrease in mitosis-related gene expression and in increase in type-I interferon-associated gene expression. Further screening studies identified treatments that showed synergistic effects in EPN, most notably cisplatin. Collectively these data support clinical evaluation of axitinib for treatment EPN and provide information of potentially synergistic treatment combinations that might overcome single agent resistance.

BACKGROUND: Cranial radiation therapy (RT) for pediatric cancer is a major risk factor for glioblastoma. Pediatric radiation-induced glioblastoma (RIG) is poorly understood, incurable, and unresponsive to treatment. We previously found that RIG’s array gene expression profile differs from spontaneous pediatric glioblastoma, raising hope that RIG may respond to different treatments. METHODS: 11 RIG patients undergoing resection from 1995–2014 were identified; RNA was extracted from each tumor and seven matched blood samples. RNA-Seq was conducted, and mutation and geneset enrichment analysis (GSEA) performed. Two samples were grown as cell lines immediately following surgery and utilized for drug screening and validation using a panel of all FDA-approved chemotherapy agents. RESULTS: RIG samples formed two distinct clustering subgroups on GSEA. One group showed upregulated mesenchymal characteristics, compared to pro-neural and neural characteristics in the other (false discovery rate (FDR)

BACKGROUND: Ependymomas in infants less than 1 year of age (iEPN) have a poorer clinical outcome than EPNs in older children. While radiation therapy is the standard of care for the latter, it is withheld in iEPN to avoid neurotoxicity to immature brain. It is unknown whether the adverse outcome in iEPNs is due to treatment differences or aggressive biology. We examined this question using molecular profiling. METHODS: Six anaplastic posterior fossa (PF) iEPNs had transcriptomic analysis and fluorescence in situ hybridization for loss of p16 and gains of 1q sequences. Results were compared with anaplastic PF EPNs from older children and findings validated by immunohistochemistry. RESULTS: All 6 iEPNs grouped within EPN PF subgroup A (PFA). Further transcriptomic analyses identified E2F targets and G2M checkpoint as the most enriched gene sets in iEPN, which was validated in a larger independent cohort. Accordingly, MIB-1 immunohistochemistry demonstrated a higher mitotic rate in iEPNs than non-infant anaplastic EPN PFA. Genetic analyses demonstrated that loss of p16 was almost exclusive to iEPN, and that none harbored 1q gains. Kaplan-Maier analysis confirmed the poorer clinical outcome of the iEPN cohort. CONCLUSIONS: Biological differences, characterized by genetic loss of p16 without gains of 1q in iEPNs, as well as deregulated E2F target gene transcription, are indicative of deregulated p16-CDK4/6-pRB-E2F pathway activity. This may underlie the poor clinical outcome seen in this group of iEPNs, rather than the withholding of radiation therapy. Results suggest a potential actionable targeted therapy for iEPN, namely CDK4/6 inhibitors such as palbociclib, which has shown favorable safety profiles in children.

MB-20. WDR5 ORCHESTRATES CANCER CELL SELF RENEWAL IN Myc DRIVEN MEDULLOBLASTOMA Swati Ghosh1, Sujatha Venkataraman1, Diane Birks1, Nicholas Foreman1,2, and Rajeev Vibhakar1,2; University of Colorado, Aurora, CO, USA; Children Hospital Colorado, Aurora, CO, USA MYC oncogene driven medulloblastoma (MB) has a poor overall outcome and there is a need for more effective therapies. Wehypothesized that epigenetic mechanisms may mediate MYC-driven medulloblastoma. To address this, we performed a functional epigenome wide RNAi screen targeting 406 epigenetic molecules associated with chromatin remodeling and histone modification in Myc MB cells. Unbiased analysis identified several key regulators of long-term viability in these Myc amplified MB cells. Remarkably we identified multiple proteins that are part of the COMPASS like macromolecular complex as critical mediators of MB cell viability. These include ASH2L, MLL2 (KMT2D), WDR5, DPY30 and HFC1. This complex is critical for methylationof histone3 lysine4 (H3K4me3) andmediating transcriptional activation. We established that WDR5 is important for MYC driven medulloblastoma cell survival. Knock down of WDR5 attenuates clonogenic growth of tumor cells, suppresses stem cell associated gene expression programs and prevents formation of new neurospheres suggesting an inhibition of self-renewal of the medulloblastoma cells. Co-immunoprecipitation studies revealed an interaction between cMYC and WDR5. Confocal microscopy using high cMYC medulloblastoma cells also demonstrated that WDR5 co-localizes with cMYC in nucleus. C-MYC overexpressing neural stem cells differentiate down a neuronal lineage upon WDR5 knockdown. We are now performing more detailed studies on the activity of WDR5 in vitro and in vivo in MYC driven medulloblastoma. Further characterization of WDR5 will provide deeper insight into understanding of group3 medulloblastoma and its therapeutic intervention. Neuro-Oncology 18:iii97–iii122, 2016. doi:10.1093/neuonc/now076.18 #The Author(s) 2016. Published by Oxford University Press on behalf of the Society for Neuro-Oncology. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

Diffuse intrinsic pontine gliomas (DIPGs) are highly aggressive, universally fatal, childhood tumors that arise in the brainstem. DIPGs have no effective treatment, and their location and diffuse nature render them inoperable. Radiation therapy remains the standard of care for this devastating disease. Polo-like kinase 1 (PLK1) regulates critical steps in mitotic progression and has been shown to be highly expressed in many human cancers. We examined the expression of PLK1 mRNA in DIPG tumor samples through microarray analysis and found it to be upregulated versus normal pons. This finding highlighted PLK1 as a potential small-molecule target. Recent studies have shown that inhibition of PLK1 leads to a decrease in cell proliferation and radiosensitization in other forms of pediatric brain tumors. Treatment of DIPG cell lines with BI 6727, a new generation, highly selective inhibitor of PLK1, resulted in decreased cell proliferation and a marked increase in cellular apoptosis. Cell cycle analysis of treated cells showed a significant arrest in G2/M phase and a substantial increase in sub-G0/G1 cell population, a finding that correlates with the expected mechanism of action of BI 6727. Treatment also resulted in an increased γH2AX expression, indicating BI 6727 can induce DNA damage. Together, these findings suggest that targeting PLK1 with small-molecule inhibitors, in combination with radiation therapy, will hold a novel strategy in the treatment of DIPG that warrants further investigation.

BACKGROUND: Medulloblastoma is the most common childhood malignant intracranial tumor that carries significant therapy-related morbidity. Checkpoint kinase 1 (CHK1) is an integral component of the cell cycle as well as the DNA Damage Response (DDR) pathway, Previous work has demonstrated the effectiveness of inhibiting CHK1 with small-molecule inhibitors, such as AZD-7762, in epithelial tumors, but research is scarce in terms of targeting the DDR for treatment in CNS malignancies. METHODS: Low-dose administration of AZD-7762 and cisplatin, the standard chemotherapy agent for medulloblastoma, were used both separately and together to assess the effect of CHK1 inhibition on c-Myc amplified medulloblastoma cell lines. Clonogenic assays were performed to assess the effect of AZD-7762, with and without cisplatin, on cell survival. Flow cytometric cell cycle and γH2AX analyses were utilized to investigate the impact of AZD-7762 on cell status and active apoptosis, respectively. RESULTS: Genetic analysis revealed a significant increase in expression of CHK1 in medulloblastoma which correlated with previously established data sets. Increased CHK1 mRNA and protein expression in medulloblastoma cell lines versus normal cerebellum was confirmed via qPCR and western blot assays. Targeting CHK1 with low-dose AZD-7762 resulted in a decrease in cell viability and cell survival as well as an increase in actively apoptotic cells. Finally, combined treatment of low-dose AZD-7762 (20 nM) with low-dose cisplatin (175 nM) was synergistic towards reducing cell survival. CONCLUSION: Overall, results demonstrate that small-molecule inhibition of CHK1 alongside the ubiquitous chemotherapeutic, cisplatin, is more advantageous than either treatment alone, especially for Group 3 medulloblastoma, and therefore this combined therapeutic approach serves as an avenue for further investigation.

Medulloblastoma is the most common pediatric brain tumor. Four molecular subgroups have been identified. Group 3 medulloblastomas are characterized by high Myc amplification and have a particularly poor prognosis, further complicated by the fact that MYC has proven very difficult to target due to the lack of clear binding ligands. In this study we evaluated the effect of a new small-molecule drug, ON 013105, in Myc driven medulloblastoma cell growth and cell death. The drug, ON 013105, a novel benzyl styryl sulfone kinase inhibitor has been shown to inhibit Myc mediated translation, cyclin D and eventually decreases tumor cell proliferation in a mouse model of mantle cell lymphoma. In our studies with high Myc expressing medulloblastoma cells, ON 013105 demonstrates an ability to reduce cell proliferation, cell survival, colony formation, and cell self-renewal capacity. Interestingly, in high Myc medulloblastoma cells, we found that ON 013105 inhibits cMyc expression in addition to cyclin D1 inhibition. Combining ON 013105 with JQ1, another small-molecule drug which inhibits cMyc expression by inhibiting BRD4, showed synergistic decrease in tumor cell proliferation. Our study suggests that ON 013105 either alone or in combination with JQ1 can be potential therapeutic agents to target high MYC medulloblastoma.

Overexpression of the Polycomb repressive complex 2 (PRC2) subunit Enhancer of Zeste 2 (EZH2) occurs in several malignancies, including prostate cancer, breast cancer, medulloblastoma, and glioblastoma multiforme. Recent evidence suggests that EZH2 may also have a role in rhabdoid tumors. Atypical teratoid/rhabdoid tumor (ATRT) is a rare, high-grade embryonal brain tumor that occurs most commonly in young children and carries a very poor prognosis. ATRTs are characterized by absence of the chromatin remodeling protein SMARCB1. Given the role of EZH2 in regulating epigenetic changes, we investigated the role of EZH2 in ATRT.Microarray analysis was used to evaluate expression of EZH2 in ATRT tumor samples. We used shRNA and a chemical inhibitor of EZH2 to examine the impact of EZH2 inhibition on cell growth, proliferation, and tumor cell self-renewal.Here, we show that targeted disruption of EZH2 by RNAi or pharmacologic inhibition strongly impairs ATRT cell growth, suppresses tumor cell self-renewal, induces apoptosis, and potently sensitizes these cells to radiation. Using functional analysis of transcription factor activity, we found the cyclin D1-E2F axis to be repressed after EZH2 depletion in ATRT cells.Our observations provide evidence that EZH2 disruption alters cell cycle progression and may be an important new therapeutic target, particularly in combination with radiation, in ATRT.