Search

Diffuse midline gliomas are uniformly fatal pediatric central nervous system cancers that are refractory to standard-of-care therapeutic modalities. The primary genetic drivers are a set of recurrent amino acid substitutions in genes encoding histone H3 (H3K27M), which are currently undruggable. These H3K27M oncohistones perturb normal chromatin architecture, resulting in an aberrant epigenetic landscape. To interrogate for epigenetic dependencies, we performed a CRISPR screen and show that patient-derived H3K27M-glioma neurospheres are dependent on core components of the mammalian BAF (SWI/SNF) chromatin remodeling complex. The BAF complex maintains glioma stem cells in a cycling, oligodendrocyte precursor cell–like state, in which genetic perturbation of the BAF catalytic subunit SMARCA4 (BRG1), as well as pharmacologic suppression, opposes proliferation, promotes progression of differentiation along the astrocytic lineage, and improves overall survival of patient-derived xenograft models. In summary, we demonstrate that therapeutic inhibition of the BAF complex has translational potential for children with H3K27M gliomas. Significance: Epigenetic dysregulation is at the core of H3K27M-glioma tumorigenesis. Here, we identify the BRG1–BAF complex as a critical regulator of enhancer and transcription factor landscapes, which maintain H3K27M glioma in their progenitor state, precluding glial differentiation, and establish pharmacologic targeting of the BAF complex as a novel treatment strategy for pediatric H3K27M glioma.

Supplementary Table from BAF Complex Maintains Glioma Stem Cells in Pediatric H3K27M Glioma

Supplementary Figure from BAF Complex Maintains Glioma Stem Cells in Pediatric H3K27M Glioma

Diffuse midline gliomas are uniformly fatal pediatric central nervous system cancers that are refractory to standard-of-care therapeutic modalities. The primary genetic drivers are a set of recurrent amino acid substitutions in genes encoding histone H3 (H3K27M), which are currently undruggable. These H3K27M oncohistones perturb normal chromatin architecture, resulting in an aberrant epigenetic landscape. To interrogate for epigenetic dependencies, we performed a CRISPR screen and show that patient-derived H3K27M-glioma neurospheres are dependent on core components of the mammalian BAF (SWI/SNF) chromatin remodeling complex. The BAF complex maintains glioma stem cells in a cycling, oligodendrocyte precursor cell–like state, in which genetic perturbation of the BAF catalytic subunit SMARCA4 (BRG1), as well as pharmacologic suppression, opposes proliferation, promotes progression of differentiation along the astrocytic lineage, and improves overall survival of patient-derived xenograft models. In summary, we demonstrate that therapeutic inhibition of the BAF complex has translational potential for children with H3K27M gliomas.Significance:Epigenetic dysregulation is at the core of H3K27M-glioma tumorigenesis. Here, we identify the BRG1–BAF complex as a critical regulator of enhancer and transcription factor landscapes, which maintain H3K27M glioma in their progenitor state, precluding glial differentiation, and establish pharmacologic targeting of the BAF complex as a novel treatment strategy for pediatric H3K27M glioma.See related commentary by Beytagh and Weiss, p. 2730.See related article by Mo et al., p. 2906.This article is highlighted in the In This Issue feature, p. 2711

Supplementary figures 1-4 and supplementary table 3. Supplementary figure 1. Genomic DNA validation of probes, assessing sensitivity, and specificity of ddPCR platform. Supplementary figure 2. Novel detection of histone 3 mutation in fluid present in a brainstem tumor cyst found in a DIPG patient at postmortem. Supplementary figure 3. Longitudinal changes in plasma ctDNA in association with MR imaging findings and clinical assessments. Supplementary figure 4. Biofluid ctDNA and tumor spread as assessed by MRI, genomic, and/or histological studies. Supplementary table 3. Mutations analyzed in the study by ddPCR with corresponding design of primers and probes.

Clinical information for PNOC003 plasma ctDNA studies.

Purpose:Pediatric diffuse midline glioma (DMG) are highly malignant tumors with poor clinical outcomes. Over 70% of patients with DMG harbor the histone 3 p.K27M (H3K27M) mutation, which correlates with a poorer clinical outcome, and is also used as a criterion for enrollment in clinical trials. Because complete surgical resection of DMG is not an option, biopsy at presentation is feasible, but rebiopsy at time of progression is rare. While imaging and clinical-based disease monitoring is the standard of care, molecular-based longitudinal characterization of these tumors is almost nonexistent. To overcome these hurdles, we examined whether liquid biopsy allows measurement of disease response to precision therapy.Experimental Design:We established a sensitive and specific methodology that detects major driver mutations associated with pediatric DMGs using droplet digital PCR (n = 48 subjects, n = 110 specimens). Quantification of circulating tumor DNA (ctDNA) for H3K27M was used for longitudinal assessment of disease response compared with centrally reviewed MRI data.Results:H3K27M was identified in cerebrospinal fluid (CSF) and plasma in 88% of patients with DMG, with CSF being the most enriched for ctDNA. We demonstrated the feasibility of multiplexing for detection of H3K27M, and additional driver mutations in patient's tumor and matched CSF, maximizing the utility of a single source of liquid biome. A significant decrease in H3K27M plasma ctDNA agreed with MRI assessment of tumor response to radiotherapy in 83% (10/12) of patients.Conclusions:Our liquid biopsy approach provides a molecularly based tool for tumor characterization, and is the first to indicate clinical utility of ctDNA for longitudinal surveillance of DMGs.

Midline glioma patient demographic information, and samples analyzed for genomic and ctDNA studies.

Assessing specificity of ddPCR platform by testing non-CNS malignant pediatric CSF and plasma.

Children diagnosed with brain tumors have the lowest overall survival of all pediatric cancers. Recent molecular studies have resulted in the discovery of recurrent driver mutations in many pediatric brain tumors. However, despite these molecular advances, the clinical outcomes of high grade tumors, including H3K27M diffuse midline glioma (H3K27M DMG), remain poor. To address the paucity of tissue for biological studies, we have established a comprehensive protocol for the coordination and processing of donated specimens at postmortem. Since 2010, 60 postmortem pediatric brain tumor donations from 26 institutions were coordinated and collected. Patient derived xenograft models and cell cultures were successfully created (76% and 44% of attempts respectively), irrespective of postmortem processing time. Histological analysis of mid-sagittal whole brain sections revealed evidence of treatment response, immune cell infiltration and the migratory path of infiltrating H3K27M DMG cells into other midline structures and cerebral lobes. Sequencing of primary and disseminated tumors confirmed the presence of oncogenic driver mutations and their obligate partners. Our findings highlight the importance of postmortem tissue donations as an invaluable resource to accelerate research, potentially leading to improved outcomes for children with aggressive brain tumors.

Collateral lethality occurs when loss of one paralog renders cancer cells dependent on the remaining paralog. Combining genome scale CRISPR/Cas9 screens coupled with RNA-sequencing in over 900 cancer cell lines, we found that cancers of nervous system lineage, including adult and pediatric gliomas and neuroblastomas, required the nuclear kinase Vaccinia-Related Kinase 1 (VRK1) for their survival. VRK1 dependency was inversely correlated with expression of its paralog VRK2. VRK2 knockout (KO) sensitized cells to VRK1 suppression, and conversely, VRK2 overexpression increased cell fitness in the setting of VRK1 suppression. DNA methylation of the VRK2 promoter was associated with low VRK2 expression in human neuroblastomas, and adult and pediatric gliomas. Mechanistically, depletion of VRK1 reduced Barrier-to-Autointegration Factor (BAF) phosphorylation during mitosis, resulting in DNA damage and apoptosis. Together, these studies identify VRK1 as a synthetic lethal target in VRK2 promoter-methylated adult and pediatric gliomas and neuroblastomas.Statement of SignificanceWe credential VRK1 as a target in adult and pediatric gliomas, and neuroblastomas with VRK2 promoter methylation. This demonstrates the utility of paralog-driven synthetic lethal interactions for biomarker-linked, targeted therapeutics.

Histone 3 lysine27-to-methionine (H3-K27M) mutations most frequently occur in diffuse midline gliomas (DMGs) of the childhood pons but are also increasingly recognized in adults. Their potential heterogeneity at different ages and midline locations is vastly understudied. Here, through dissecting the single-cell transcriptomic, epigenomic and spatial architectures of a comprehensive cohort of patient H3-K27M DMGs, we delineate how age and anatomical location shape glioma cell-intrinsic and -extrinsic features in light of the shared driver mutation. We show that stem-like oligodendroglial precursor-like cells, present across all clinico-anatomical groups, display varying levels of maturation dependent on location. We reveal a previously underappreciated relationship between mesenchymal cancer cell states and age, linked to age-dependent differences in the immune microenvironment. Further, we resolve the spatial organization of H3-K27M DMG cell populations and identify a mitotic oligodendroglial-lineage niche. Collectively, our study provides a powerful framework for rational modeling and therapeutic interventions.

Pediatric high-grade glioma (pHGG) is an incurable disease with a median survival of less than 6 months post-progression and no effective targeted therapy. PDGFRA is commonly altered in pHGG, but targeting PDGFRA in this disease has been unsuccessful, likely due to poor central nervous system (CNS) penetrance. Avapritinib is a novel and CNS-penetrant PDGFRA/KIT inhibitor that is FDA-approved for adults with unresectable or metastatic PDGFRA exon 18-mutant gastrointestinal stromal tumor (GIST) and is being studied in CNS tumors. We performed a pre-clinical and clinical assessment to determine the potential suitability of avapritinib therapy in PDGFRA-driven glioma. A multi-institutional cohort genetic analysis revealed PDGFRA amplification and mutation in 10.2% and 6.1% of pHGG, respectively. Additionally, PDGFRA expression in the absence of genetic events was significantly increased in H3K27-altered diffuse midline glioma (DMG) compared to H3-wildtype pHGG. Avapritinib performed well in: (i) mutant PDGFRA enzyme inhibition and wildtype inhibition at high dose, (ii) minimal off-target kinase inhibition, (iii) brain penetration (peak 10 µM), and (iv) proliferation/pPDGFRA reduction in PDGFRA-amplified and mutant pHGG cell lines. Avapritinib treatment in an aggressive PDX model of pHGG resulted in significant survival benefit. We pursued treatment of eight pediatric and young adult HGG patients with avapritinib across seven institutions. Patients were a mixture of local (N = 4) and metastatic disease (N = 4); all patients were post-initial radiation, with 7/8 having progressed on prior treatment. 7/8 patients had PDGFRA amplifications or mutations, and 7/8 had H3K27M mutations. Therapy was generally well-tolerated. 4/8 patients showed radiographic response to avapritinib, with one patient demonstrating complete response of target lesion and remains on therapy. Avapritinib levels in patients’ CSF and brain tumor tissue reached micromolar levels. These results demonstrate that avapritinib is a potent, selective, and CNS-penetrant PDGFRA/KIT inhibitor that is promising for further study in pHGG with relevant alterations.

Histone 3 lysine27-to-methionine mutant diffuse midline gliomas (H3-K27M DMGs) are among the most lethal brain tumors. Their putative cellular hierarchy has been shown to be driven by self-renewing stem-like cells arrested in an oligodendrocyte precursor-like (OPC-like) state, of which few cells are able to differentiate towards more mature astrocyte (AC)-like and oligodendrocyte (OC)-like cells. However, the spatial organization underlying this tumor cell architecture and its microenvironmental interactions in intact H3-K27M DMG tissues remain unknown. Here, we profiled the single cell transcriptomes of 45 patient H3-K27M DMGs and derived cell population-specific marker gene combinations to characterize the single cell spatial organization of 16 tumors using targeted in situ sequencing. We thereby resolved different malignant and non-malignant cell populations including cycling, OPC-like, AC-like, OC-like, mesenchymal tumor cells, and non-malignant oligodendrocytes, astrocytes, neurons, myeloid cells, T cells, and vascular cells directly in situ. Global neighborhood analyses indicate a higher tendency of cycling OPC-like cells, vascular cells, and neurons to localize within a more restricted homogeneous compartment, whereas AC-like cells, non-malignant astrocytes and myeloid cells tend to intermingle with different cell populations in a more diffuse manner. Among malignant cells, we observed cycling OPC-like and OC-like cells to co-localize within a niche-like structure that is surrounded by more differentiated AC-like cells. We further validated this stem-like niche at the protein level using multiplexed immunofluorescence via the CODEX system. Finally, we characterized relationships between malignant and non-malignant cells, consistently identifying preferred neighborhoods of mesenchymal tumor cells with vascular and myeloid cells. Together, this study resolves the spatial architecture of H3-K27M DMG malignant and non-malignant cells at single cell resolution and identifies a local niche of the oligodendroglial lineage containing the OPC-like cancer stem-like cells, thus providing novel insights into the cancer stem-like compartment in H3-K27M DMGs and suggesting potential avenues for its perturbation.

Diffuse intrinsic pontine glioma (DIPG) remains one of the most lethal brain tumors in all of childhood with no effective treatments besides radiation, which only extends survival a few months. Against this backdrop, our lab recently executed a focused CRISPR negative selection screen in DIPG cell lines after treatment with the histone deacetylase (HDAC) inhibitor panobinostat and discovered a strong co-dependence with the histone demethylase LSD1. To further explore the therapeutic potential of this synergistic interaction, we tested a drug library of HDAC- and LSD1- targeting drugs with the goal of identifying a combination with optimal synergy and blood brain barrier (BBB) penetration suitable for clinical translation. We were surprised to find that traditional catalytic LSD1 inhibitors had minimal effect in isolation and did not seem to synergize with HDAC inhibitors, while a recently described CoREST/LSD1 degrader named UM171 phenocopied the effects seen in our CRISPR screen. Degraders are a class of compounds that recruit an E3 ubiquitin ligase to a protein-of-interest and cause target ubiquitination and proteasomal degradation. Given our unexpected finding, we hypothesized that UM171 induces synergy with HDAC inhibitors through elimination of a scaffolding function of LSD1. To prove this, we knocked out LSD1 using CRISPR/Cas9 and subsequently treated with a panel of HDAC inhibitors, which showed a signification sensitization of DIPG cells to HDAC inhibitors compared to standard controls. We also confirmed that UM171 interacts with the CoREST complex (members include LSD1, RCOR1, HDAC1/2) by performing streptavidin bead pull down with a newly synthesized biotin-conjugated UM171 probe. In summary, our results show that targeting LSD1 for degradation in combination with HDAC inhibition is a synergistic strategy in DIPG worthy of further translational study.

One of the biggest obstacles in developing effective therapies for CNS tumors is drug delivery due to the enigmatic challenge of penetrating the blood-brain barrier. 5-azacytidine is a DNA methyltransferase inhibitor which was the first epigenetic targeting chemotherapeutic approved by the FDA. Altered DNA methylation is a hallmark of many cancers, including diffuse intrinsic pontine glioma (DIPG). 5-azacytidine has been shown to be active in DIPG cell lines, with only modest in-vivo activity. We have previously shown in a non-human primate model that intravenous (IV) administration of 5-azacitidine does not result in measurable CSF penetration, while intrathecal (IT) administration does and is well tolerated. To follow up these studies in a tumor-bearing mouse model (HSJD DIPG007), we performed pharmacokinetic (PK) and pharmacodynamic (PD) analysis following IV vs. IT administration. Forty mice were randomized to receive four weekly doses of IV 5-azacytidine (25 mg/kg), IT 5-azacytidine (40 μg), or corresponding vehicle controls. Four mice from each arm were sacrificed 30 minutes after the last dose and brain tissue was collected for PK/PD analysis. Drug concentration was quantified using ultra-high-performance liquid chromatography with tandem mass spectrometric detection, while pharmacodynamic methylation profiling was performed using the Infinium MethylationEPIC BeadChip (850K). Brain tissue concentrations were comparable between IV (7.6-58.0 pg/mg) and IT (6.9-63.9 pg/mg) dosing. Methylation profiling unexpectedly showed a significantly more pronounced pharmacodynamic effect with IV dosing vs. IT, with a mean decrease of 13.6% vs. 2.6% in global DNA methylation score (GDMS = percentage of highly methylated (beta ≥ 0.7) genomic loci) compared to vehicle controls. For the remaining mice, there was no significant difference in survival. Our results are encouraging that phenotypically relevant demethylating effects can be achieved in the CNS with IV 5-azacytidine administration; however, further research is needed to develop promising combination strategies in DIPG.

Diffuse midline glioma (DMG) is a highly morbid pediatric brain tumor. Up to 80% of DMGs harbor mutations in histone H3-encoding genes, associated with poor prognosis. We previously showed the feasibility of detecting H3 mutations in circulating tumor DNA (ctDNA) in the liquid biome of children diagnosed with DMG. However, detection of low levels of ctDNA is highly dependent on platform sensitivity and sample type. To address this, we optimized ctDNA detection sensitivity and specificity across two commonly used digital droplet PCR (ddPCR) platforms (RainDance and BioRad), and validated methods for detecting H3F3A c.83A > T (H3.3K27M) mutations in DMG CSF, plasma, and primary tumor specimens across three different institutions. DNA was extracted from H3.3K27M mutant and H3 wildtype (H3WT) specimens, including H3.3K27M tumor tissue (n = 4), CSF (n = 6), plasma (n = 4), and human primary pediatric glioma cells (H3.3K27M, n = 2; H3WT, n = 1). ctDNA detection was enhanced via PCR pre-amplification and use of distinct custom primers and fluorescent LNA probes for c.83 A > T H3F3A mutation detection. Mutation allelic frequency (MAF) was determined and validated through parallel analysis of matched H3.3K27M tissue specimens (n = 3). We determined technical nuances between ddPCR instruments, and optimized sample preparation and sequencing protocols for H3.3K27M mutation detection and quantification. We observed 100% sensitivity and specificity for mutation detection in matched DMG tissue and CSF across assays, platforms and institutions. ctDNA is reliably and reproducibly detected in the liquid biome using ddPCR, representing a clinically feasible, reproducible, and minimally invasive approach for DMG diagnosis, molecular subtyping and therapeutic monitoring.

Background Diffuse midline glioma (DMG) is a highly morbid pediatric brain tumor. Up to 80% of DMGs harbor mutations in histone H3-encoding genes, associated with poorer prognosis. We previously showed the feasibility of detecting H3 mutations in circulating tumor DNA (ctDNA) in the liquid biome of children diagnosed with DMG. However, detection of low ctDNA concentrations is highly dependent on platform sensitivity and sample type. To address this, we optimized ctDNA detection sensitivity and specificity across two commonly used digital droplet PCR (ddPCR) platforms (RainDance and BioRad), and validated methods for detecting H3F3A mutations in DMG CSF, plasma, and primary tumor specimens across three different institutions. Methods DNA was extracted from H3.3K27M mutant and H3 wildtype (H3WT) specimens, including H3.3K27M tumor tissue (n=4), CSF (n=6), plasma (n=4), and human primary pediatric glioma cells (H3.3K27M, n=2; H3WT, n=1). ctDNA detection was enhanced via PCR pre-amplification and use of distinct custom primers and fluorescent LNA probes for c.83 AàT H3F3A mutation detection. Mutation allelic frequency (MAF) was determined and validated through parallel analysis of matched H3.3K27M tissue specimens (n=3). Results We determined technical nuances between ddPCR instruments, and optimized sample preparation and sequencing protocols for H3F3A mutation detection and quantification. We observed 100% sensitivity and specificity for mutation detection in matched DMG tissue and CSF across assays, platforms and institutionsConclusion Our study demonstrates that ctDNA is reliably and reproducibly detected in ctDNA using ddPCR, representing a clinically feasible and reproducible minimally invasive approach for DMG diagnosis, molecular subtyping and therapeutic monitoring.

Purpose of review Altered epigenetics is central to oncogenesis in many pediatric cancers. Aberrant epigenetic states are induced by mutations in histones or epigenetic regulatory genes, aberrant expression of genes regulating chromatin complexes, altered DNA methylation patterns, or dysregulated expression of noncoding RNAs. Developmental contexts of dysregulated epigenetic states are equally important for initiation and progression of many childhood cancers. As an improved understanding of disease-specific roles and molecular consequences of epigenetic alterations in oncogenesis is emerging, targeting these mechanisms of disease in childhood cancers is increasingly becoming important. Recent findings In addition to disease-causing epigenetic events, DNA methylation patterns and specific oncohistone mutations are being utilized for the diagnosis of pediatric central nervous system (CNS) and solid tumors. These discoveries have improved the classification of poorly differentiated tumors and laid the foundation for future improved clinical management. On the therapeutic side, the first therapies targeting epigenetic alterations have recently entered clinical trials. Current clinical trials include pharmacological inhibition of histone and DNA modifiers in aggressive types of pediatric cancer. Summary Targeting novel epigenetic vulnerabilities, either by themselves, or coupled with targeting altered transcriptional states, developmental cell states or immunomodulation will result in innovative approaches for treating deadly pediatric cancers.

High-grade gliomas harboring H3 G34R/V mutations exclusively occur in the cerebral hemispheres of adolescents and young adults, suggesting a distinct neurodevelopmental origin. Combining multimodal bulk and single-cell genomics with unbiased genome-scale CRISPR/Cas9 approaches, we here describe a GABAergic interneuron progenitor lineage as the most likely context from which these H3 G34R/V mutations drive gliomagenesis, conferring unique and tumor-selective gene targets essential for glioma cell survival, as validated genetically and pharmacologically. Phenotypically, we demonstrate that while H3 G34R/V glioma cells harbor the neurotransmitter GABA, they are developmentally stalled, and do not induce the neuronal hyperexcitability described in other glioma subtypes. These findings offer a striking counter-example to the prevailing view of glioma origins in glial precursor cells, resulting in distinct cellular, microenvironmental, and therapeutic consequences.

This clinical trial evaluated whether whole exome sequencing (WES) and RNA sequencing (RNAseq) of paired normal and tumor tissues could be incorporated into a personalized treatment plan for newly diagnosed patients (

Complications associated with upfront and repeat surgical tissue sampling present the need for minimally invasive platforms capable of molecular sub-classification and temporal monitoring of tumor response to therapy. Here, we describe our dPCR-based method for the detection of tumor somatic mutations in cell free DNA (cfDNA), readily available in patient biofluids. Although limited in the number of mutations that can be tested for in each assay, this method provides a high level of sensitivity and specificity. Monitoring of mutation abundance, as calculated by MAF, allows for the evaluation of tumor response to therapy, thereby providing a much-needed supplement to radiographic imaging.

INTRODUCTION: Molecular profiling of the tumor genome is critical for diagnosis and clinical management of CNS tumors. In the case of pediatric diffuse midline gliomas (DMGs), the neuroanatomical location of the tumor precludes surgical resection. Moreover, when tumor resection or biopsy is permitted, serial monitoring of the tumor is limited to MR imaging. We have recently reported on detection of tumor associated somatic mutations in patient liquid biopsy specimens. While powerful, our current approach is limited to multiplexed detection of up to two somatic mutations, limiting comprehensive profiling. To address this, we developed a sensitive approach for profiling tumor genomic aberrations in circulating tumor DNA (ctDNA). RESULTS: We show successful detection of tumor-specific mutations in plasma, serum, and cerebrospinal fluid (CSF) of pediatric DMGs, using a targeted panel to sequence exons of 170 candidate genes. Our analysis indicates overlap of somatic tumor mutations in CSF and plasma with matched tissue; a higher number of mutations were detected in CSF compared to plasma. Patterns of tumor evolution were revealed in CSF analyzed at end-stage disease. For example, loss of a TP53 mutation at progression compared to initial diagnosis was confirmed in CSF collected at postmortem, indicating utility of our platform for identification of tumor evolution. Additionally, subclonal heterogeneity of DMGs was captured in CSF compared to tumor specimens. Low allelic frequency mutations in genes involved in DMG tumorigenesis, such as PI3K signaling and chromatin regulators, were exclusively detected in CSF, but not in matched tumor tissue. IMPACT: Our novel approach enables tumor exome monitoring of low frequency variants in liquid biopsy specimens. Our sensitive targeted sequencing panel provides opportunity for profiling somatic tumor mutations at initial diagnosis and at progression, for mapping tumor evolution in the context of clinical trials.

RATIONALE: A major factor contributing to the expanding knowledge of CNS tumors, specifically DMG biology is the selfless gift from patients and their families of postmortem tumor tissue. To facilitate this, we developed a postmortem donation program for collection of tissue and biospecimens for accelerating biology studies of end stage disease for pediatric CNS tumors. METHODS: Our postmortem program entails coordinating tissue donations, compiling clinical data, processing of whole brain, spinal cord, biofluids, and germline controls. Through our protocol, primary and disseminated tumor specimens are designated for molecular characterization and for generation of a matched preclinical model. RESULTS: During the past eight years, we coordinated and accrued over 50 autopsy cases including diffuse midline gliomas, ATRT, ETANTR, CPC from 20 institutions across North America and the United Kingdom. We report that approaching a family at the time of progression has provided the most successful outcome for consent to our program. Patient age range was from 11months to 20 years old. The median time from postmortem donation to tissue processing range was 13 hours with the earliest time point being 4 hours and the longest being 96 hours. All collected specimens were successfully processed and assessed for DNA/RNA integrity using bioanalyzer, attempted primary neurosphere cultures and cryopreserved for future molecular studies. PDX generation from fresh tissue proved more successful compared to primary neurospheres and post mortem time did not seem to have an effect. We have successfully generated models that represent molecular subgroups of DMGs such as H3.3/ H3.1- K27M and H3 WT. All generated preclinical models were validated by STR analysis and immunohistochemical assays using human mitochondrial antibodies. IMPACT: Postmortem tissue donations serve as an invaluable source for access to end-stage disease biology, generation of PDX models, studying patterns of tumor migration, testing combination therapy and preclinical drug testing.

RATIONALE: One of the major hurdles in developing effective treatment for children with DIPG includes the lack of extensive combinatorial studies targeting major driver oncogenic pathways. Combination of H3K27M and TP53 mutations are found in over 50% of DIPG tumors, however, there is a lack of effective combinatorial precision therapy. METHODS: Tissue specimens obtained from subjects with DIPG were used to generate preclinical models. Subjects were enrolled in a prospective molecular diagnostic clinical trial and specimens were obtained by surgical biopsy both at initial diagnosis, and at the time of progression. RESULTS: Fourteen primary neurosphere cell lines and nine xenograft mouse models were successfully generated. The success rate of generating primary neurospheres and xenograft models was 45% and 75%, respectively. Among the 14 cell lines, one was derived from a biopsy performed after tumor progression, and one from washing a biopsy needle. Whole genome analysis of three biopsy-derived primary neurospheres revealed that these samples shared genomic alterations when compared to the primary tumors. All cell lines retained major oncogenic driver mutations in genes such as H3F3A, HIST1H3B, TP53, PPM1D, PIK3R1, and ACVR1. The needle-wash derived cell line was immortalized with hTERT, and exhibited the most deviation in mutation profiles and copy number alterations compared to primary tumor. Global DNA methylation profiling of seven primary tumor biopsies and matched primary neurospheres revealed similar epigenetic profiles. Primary neurospheres derived from biopsies were treated with the same specialized panel of agents recommended by a personalized medicine tumor board, which was based on the mutational profiles of the original tumor. SIGNIFICANCE: All cell lines exhibited varying levels of sensitivity to single agent indicating patterns of intertumor heterogeneity. The use of these model systems allows testing of combinatorial precision therapy for patients with DIPG, where there is a need for rapid translation into clinical trials.

Anaplastic astrocytomas are aggressive glial cancers that present poor prognosis and high recurrence. Heterozygous IDH1 R132H mutations are common in adolescent and young adult anaplastic astrocytomas. In a majority of cases, the IDH1 R132H mutation is unique to the tumor, although rare cases of anaplastic astrocytoma have been described in patients with mosaic IDH1 mutations (Ollier disease or Maffucci syndrome). Here, we present two siblings with IDH1 R132H mutant high grade astrocytomas diagnosed at 14 and 26 years of age. Analysis of IDHR132H mutations in the siblings' tumors and non-neoplastic tissues, including healthy regions of the brain, cheek cells, and primary teeth indicate mosaicism of IDHR132H. Whole exome sequencing of the tumor tissue did not reveal any other common mutations between the two siblings. This study demonstrates the first example of IDH1 R132H mosaicism, acquired during early development, that provides an alternative mechanism of cancer predisposition.

There is no curative treatment for patients with diffuse intrinsic pontine glioma (DIPG). However, with the recent availability of biopsy and autopsy tissue, new data regarding the biologic behavior of this tumor have emerged, allowing greater molecular characterization and leading to investigations which may result in improved therapeutic options. Treatment strategies must address both primary disease sites as well as any metastatic deposits, which may be variably sensitive to a particular approach. In this case report, we present a patient with DIPG treated with irradiation and serial investigational agents. The clinical, pathological and molecular phenotypes of both the progressive primary tumor as well as concomitant metastatic deposits obtained at autopsy are discussed. While some mRNA differences were demonstrated, all analyzed sites of disease shared similar mutational arrangements, suggesting that targeting the mutations of the primary tumor may be effective for all sites of disease.

Children diagnosed with diffuse intrinsic pontine glioma (DIPG), a type of high grade glioma in the brainstem, currently have a dismal 5-year overall survival of only 2%. The majority of DIPG patients harbor a K27M mutation in histone 3.3 encoding genes (H3.3K27M). To understand if the aberrant epigenetic landscape induced by H3.3K27M provides an opportunity for novel targeted therapies, we conducted the first CRISPR/Cas9 screen using a focused library of 1,350 epigenetic regulatory and cancer related genes in six H3.3K27M DIPG patient-derived primary neurosphere cell lines. We identified gene dependencies in chromatin regulators, polycomb repressive complexes 1 and 2 (PRC1 and PRC2), histone demethylases, acetyltransferases and deacetylators as novel tumor cell dependencies in DIPG. We hypothesized that targeting dysregulated functions of chromatin regulators by genetically deleting and chemically targeting these epigenetically induced vulnerabilities, we could ameliorate, or even reverse the downstream oncogenic effects of the aberrant epigenetic landscape of DIPG. In our secondary CRISPR nanoscreen, we first used six single guide RNAs (sgRNA) to knockout each gene using CRISPR/Cas9 ribonucleoprotein nucleofections, followed by use of three best sgRNAs combined with homology directed repair templates. Compared to lentiviral delivery, nucleofection is a rapid method, with reduced off-target toxicity, suitable for single gene knockouts in DIPG neurospheres. Secondary CRISPR validations confirmed dependencies in BMI1, CBX4, KDM1A, EZH2, EED, SUZ12, HDAC2, and EP300. Next, we conducted a chemical screen using 20 inhibitors and degraders to target the aberrant activity of HDAC, KDM1A, P300/CBP, PRC1 and PRC2. We identified eight chemical compounds that were effective in H3.3K27M DIPG neurosphere cell lines at low drug concentrations. Among these, an inhibitor and degrader targeting P300/CBP activity indicates a novel strategy of epigenetic therapy in DIPG. Through our combinatorial testing, we will identify a synergistic combination of epigenetic therapy for treating children diagnosed with H3.3K27M DIPG.

Purpose: Pediatric diffuse midline glioma (DMG) are highly malignant tumors with poor clinical outcomes. Over 70% of patients with DMG harbor the histone 3 p.K27M (H3K27M) mutation, which correlates with a poorer clinical outcome, and is also used as a criterion for enrollment in clinical trials. Because complete surgical resection of DMG is not an option, biopsy at presentation is feasible, but rebiopsy at time of progression is rare. While imaging and clinical-based disease monitoring is the standard of care, molecular-based longitudinal characterization of these tumors is almost nonexistent. To overcome these hurdles, we examined whether liquid biopsy allows measurement of disease response to precision therapy. Experimental Design: We established a sensitive and specific methodology that detects major driver mutations associated with pediatric DMGs using droplet digital PCR (n = 48 subjects, n = 110 specimens). Quantification of circulating tumor DNA (ctDNA) for H3K27M was used for longitudinal assessment of disease response compared with centrally reviewed MRI data. Results: H3K27M was identified in cerebrospinal fluid (CSF) and plasma in 88% of patients with DMG, with CSF being the most enriched for ctDNA. We demonstrated the feasibility of multiplexing for detection of H3K27M, and additional driver mutations in patient's tumor and matched CSF, maximizing the utility of a single source of liquid biome. A significant decrease in H3K27M plasma ctDNA agreed with MRI assessment of tumor response to radiotherapy in 83% (10/12) of patients. Conclusions: Our liquid biopsy approach provides a molecularly based tool for tumor characterization, and is the first to indicate clinical utility of ctDNA for longitudinal surveillance of DMGs.

Anaplastic astrocytomas are aggressive glial cancers that present poor prognosis and high recurrence. Heterozygous mutations in the IDH1 (R132H) gene are common in adult anaplastic astrocytomas, but are uncommon in the pediatric population. In a majority of cases, the IDH1 R132H mutation is somatic, although rare cases of anaplastic astrocytoma have been described in patients with mosaic germline mutations of IDH1 (Ollier disease or Maffucci syndrome). Here, we present two siblings who developed IDH1 R132H mutant high grade astrocytomas at 15 and 26 years of age. Our analysis of IDH1 R132H mutations in the siblings’ tumors and non-neoplastic tissues, including healthy regions of the brain (post-mortem samples from one sibling), cheek cells and primary teeth identified IDH1 R132H mosaicism, using digital PCR system. While both tumors had IDH1 R132H mutant allelic frequencies around 50%, mutant allelic frequencies ranging from 0.01% to 0.64% were detected in teeth, cheek cells, blood and non-tumorous brain tissue. Whole exome sequencing of the tumor tissue did not reveal any other common mutations between the two siblings. Our study demonstrates an example of IDH1 R132H mosaicism in two siblings with anaplastic astrocytoma, which could have gone unnoticed by traditional sequencing technologies. This finding supports a trend, seen in other cancers, of mosaic-like mutant allelic gene frequencies associated with cancer development. Further large-scale studies are needed to better understand the prevalence of mosaic IDH1, or other mutations, in patients with brain tumors, as this could impact the diagnosis and management of these patients.

Midline high-grade gliomas (HGGs) are amongst the most challenging childhood cancers to treat. Recent advances in understanding the molecular drivers of these tumors have resulted from donated autopy or biopsy specimens. We have initiated a midline glioma preclinical initiative for producing in vitro and in vivo models that have well defined genomic profiles. We obtained 22 autopsy, 11 upfront biopsy (PNOC NCT 2274987) and 2 surgical resection samples. Collectively, specimens represented DIPGs (n=30) and thalamic HGGs (n=5). Single cell suspensions were generated for neurosphere culture and intracranial injection into mice. Specimens were considered viable if they resulted in two consecutive generations of murine tumor models or five in vitro passages. A total of 14 cell lines representing DIPG and thalamic glioma were generated and validated by short tandem repeat (STR) analysis. Orthotopic injections into non scid gamma (NSG) mice resulted in tumor formation (n=14) with molecular characteristics similar to the corresponding patient tumor. Eleven mouse implantations are still being monitored. In vivo models were validated by immunohistochemical staining for human mitochondria and H3K27M. Methylation and whole genome analysis of preclinical models were generated to assess similarities/shared pathways with human disease. In vitro and in vivo response to therapeutics including panobinostat and TAK228 was also assessed. A microdialysis procedure was optimized to assess blood brain barrier penetration of systemically administered therapeutics. Direct drug delivery was also evaluated using subcutaneously placed osmotic pumps with optimization of delivery rates, cannula depth and placement.

Diffuse intrinsic pontine glioma (DIPG) is one of the most lethal pediatric central nervous system (CNS) cancers. Recently, a surge in molecular studies of DIPG has occurred, in large part due to the increased availability of tumor tissue through donation of post-mortem specimens. These new discoveries have established DIPGs as biologically distinct from adult gliomas, harboring unique genomic aberrations. Mutations in histone encoding genes are shown to be associated with >70% of DIPG cases. However, the exact molecular mechanisms of the tumorigenicity of these mutations remain elusive. Understanding the driving mutations and genomic landscape of DIPGs can now guide the development of targeted therapies for this incurable childhood cancer.

Children diagnosed with diffuse midline histone 3 K27M mutant (H3K27M) glioma continue to have a dismal prognosis with a median survival time of about nine months post-diagnosis. MRI, the standard for assessing tumor growth, is often not sensitive for detecting early tumor regression/progression. We evaluate the utility of biofluids from patients and their xenograft models, for detection of circulating tumor DNA (ctDNA). The goal is to develop a comprehensive panel for ctDNA detection and quantification, capable of assessing response to treatment. Twenty two CSF samples were collected at various stages of disease (upfront, recurrence, and postmortem) from pediatric high grade midline gliomas. Upfront and longitudinal plasma samples (n=68) were collected with each MRI, from 19 diffuse intrinsic pontine glioma (DIPG) patients enrolled in a Phase I clinical trial. For monitoring ctDNA, we developed a liquid biopsy assay using a sensitive and robust digital droplet PCR system on CSF and plasma samples. We successfully detected major DIPG driver mutations including H3F3A, HIST1H3B, and ACVR1. Specifically, we detected H3K27M in 80% of CSF and upfront plasma samples. Our data indicate a concordance of H3K27M in tumor and CSF for 70% of patients (n=13). In accordance with published studies in adult CNS cancers, we detected higher levels of ctDNA in CSF compared to plasma. Longitudinal analysis of plasma showed a drastic decrease in ctDNA following radiotherapy in 86% of patients (n=14), and patients were categorized as responders, where dynamic changes in ctDNA correlated with changes in MRI tumor measurements. Furthermore, we have detected H3K27M in ctDNA of plasma samples in xenograft models of DIPG. We conclude that biofluid is a suitable medium for detection and quantification of ctDNA. Liquid biopsy has the potential to complement, or in some cases eliminate the need for biopsies, and may inform recurrence and tumor response to treatment.

Diffuse intrinsic pontine glioma (DIPG) is a fatal pediatric brain cancer which presents an immense therapeutic challenge. Our previous work has shown that histone mutations (H3K27M) arise in the pontine tumor but in most cases necessitates an obligate partner mutation for tumorigenesis and spread. Our established standard operating procedure for performing autopsy has allowed us to collect 20 whole brains from DIPG patients. Sagittal sections from whole brains (n=8) were processed for histological analyses to identify the tumor migratory path (pontine to extra-pontine). Histological examination of all neuroanatomical locations indicated: i) Tumor cell infiltration to extra-pontine regions in 7/8 brains as indicated by histology, ii) In 4/7 brains, tumor cells migrated to extra-pontine regions but failed to form tumors. These regions were histologically “normal” according to the World Health Organization grading criteria, but harbored tumor cells as indicated by H3K27M stains and digital PCR. Given the age of disease onset, and neuroanatomical location, it is plausible that tumor microenvironment plays a critical role in tumor spread. To test the role of pontine microenvironment in tumorigenesis, we performed a pilot study by injecting human pontine tumor cells (n= 4 patients) into 3-4 week old murine cerebral cortex, which represents the age of DIPG disease onset. Necropsy specimens were examined for tumor presence by histological analyses for H&E and H3K27M stains. We identified tumor in both cortex (injection site) and brainstem. Tumor cells migrated to populate brainstem in all four, and cerebellum in 2/4 xenograft models. Furthermore these tumors were positive for human mitochondrial protein (MAB1273) and were authenticated by STR to demonstrate their human cell of origin. These observations strongly show the presence of: i) anatomically suitable migration paths (e.g. axonal guidance, leptomeningeal), and ii) tissue-specific microenvironment and molecular events, governing susceptibility and sensitivity of particular neuroanatomical locations to tumor formation.

Children with high-grade glioma, including diffuse intrinsic pontine glioma (DIPG), have a poor prognosis despite multimodal therapy. Identifying novel therapeutic targets is critical to improve their outcome. We evaluated prognostic roles of telomere maintenance mechanisms in children with HGG, including DIPG. A multi-institutional retrospective study was conducted involving 50 flash-frozen HGG (35 non-brainstem; 15 DIPG) tumors from 45 children (30 non-brainstem; 15 DIPG). Telomerase activity, expression of hTERT mRNA (encoding telomerase catalytic component) and TERC (telomerase RNA template) and alternative lengthening of telomeres (ALT) mechanism were assayed. Cox Proportional Hazard regression analyses assessed association of clinical and pathological variables, TERC and hTERT levels, telomerase activity, and ALT use with progression-free or overall survival (OS). High TERC and hTERT expression was detected in 13/28 non-brainstem HGG samples as compared to non-neoplastic controls. High TERC and hTERT expression was identified in 13/15 and 11/15 DIPG samples, respectively, compared to controls. Evidence of ALT was noted in 3/11 DIPG and 10/19 non-brainstem HGG specimens. ALT and telomerase use were identified in 4/19 non-brainstem HGG and 2/11 DIPG specimens. In multivariable analyses, increased TERC and hTERT levels were associated with worse OS in patients with non-brainstem HGG, after controlling for tumor grade or resection extent. Children with HGG and DIPG, have increased hTERT and TERC expression. In children with non-brainstem HGG, increased TERC and hTERT expression levels are associated with a worse OS, making telomerase a promising potential therapeutic target in pediatric HGG.

INTRODUCTION Pediatric diffuse midline glioma (DMG) with histone H3 mutations (80%), are highly morbid tumors with poor response to therapy. We previously detected H3 mutations in circulating tumor DNA (ctDNA) from CSF derived from children with DMG and high grade glioma. Here, we describe a high-throughput, sensitive and specific approach for H3 mutation detection and quantification in plasma and CSF specimens, validated across multiple centers. METHODS DNA extracted from tissue-validated H3.3K27M specimens (4 tumor, 5 CSF, 4 plasma) and pediatric glioma cell lines (1 wild-type, 1 H3.3K27M) were used to standardize ddPCR workflows. ctDNA extracted from 500uL of CSF or plasma was pre-amplified using sequence-specific primers, then analyzed on RainDance and BioRad ddPCR systems using two sets of custom primers and fluorescent LNA probes (Assays A and B). H3 mutation allelic frequency (MAF) was determined across specimens, and validated through parallel analysis of additional matched H3.3K27M tumor tissue, CSF and plasma specimens (n=3). RESULTS ctDNA processing and detection was standardized across three institutions and two platforms with sensitive, specific and reproducible H3 mutation detection. H3K27M mutations were detected in as little as 0.3-2pg input DNA on both platforms, with comparable MAFs across instruments and assays. RainDance yielded greater overall positive droplet number detection (increased sensitivity). Differences in optimal sample input volume between platforms were circumvented by vacuum concentration or dilution to maintain equal input DNA. Assay A yielded superior specificity so was used for subsequent matched specimen analysis. Mutation detection in plasma, as in previous studies, remained challenging due to low [ctDNA], while CSF analysis yielded reliable results across assays and platforms. CONCLUSIONS We demonstrate the utility of liquid biopsy for identifying H3K27M mutations in plasma and CSF via ddPCR with low starting [DNA], representing a rapid, minimally invasive method for diagnosis and therapeutic monitoring of DMG.

BACKGROUND: Diffuse intrinsic pontine glioma (DIPG) is an infiltrative, inoperable midline tumor with devastating outcomes. We have recently completed a prospective clinical trial for children and young adults with newly diagnosed DIPG that included clinical whole exome sequencing (WES) and RNA sequencing (RNAseq) of tumors and development of a personalized treatment plan. In this study, we retrospectively expanded sequencing efforts to include whole genome sequencing (WGS) and methylation profiling of paired tumor/normal (T/N) samples as well as longitudinal tumor specimens. We sought to perform a pan-omics analysis to determine which sequencing platform combination could most benefit patient outcomes in future precision medicine DIPG trials. METHODS: Thirty-three DIPG patients underwent successful biopsy in PNOC003. Clinical WES/exome panel sequencing and RNAseq was generated for 30 primary tumors and 2 progression tumors. We performed research-based WGS and methylation profiling of all tumors to do a pan-omics analysis. Cross-platform analysis was performed to compare WGS and WES and define additional mutations. RESULTS: In addition to WES and RNAseq, we performed WGS on all 33 primary T/N pairs, 2 progression samples, and 3 autopsy samples. Methylation profiling was done on 30 primary, 1 recurrent and 6 autopsy tumor samples. WGS identified all the actionable coding alterations called by WES and further identified non-coding variants of potential clinical significance. A pan-omics analysis spanning WGS, WES, RNAseq and methylation data is being performed. Tumor evolution was demonstrated by longitudinal comparison of genetic alterations across primary, recurrent and autopsy tumor samples to highlight tumor-intrinsic versus treatment-driven alterations in DIPGs. DISCUSSION: The ongoing pan-omic analysis from PNOC003 has begun informing the next generation of molecularly-driven DIPG clinical trials by suggesting inclusion of WGS in clinical sequencing panels as well as offering additional insights into the diverse mutational landscape of DIPGs.

OBJECTIVE: We evaluated the impact on overall survival at 12 months (OS12) of a personalized therapy plan based on tumor/germline whole exome (WES) and tumor RNA sequencing (RNAseq) of diffuse intrinsic pontine glioma (DIPG). METHODS: We enrolled newly diagnosed DIPG patients ≤ 25 years of age. Tumors were sequenced and drug selection was performed using a custom drug-matching pipeline and pharmacopeia. A personalized treatment strategy using up to 4 FDA-approved drugs was determined by a specialized tumor board with target of 21 business days from biopsy. Patients were followed for adverse events (AEs) and OS12. Circulating tumor DNA (ctDNA) was collected at diagnosis and with surveillance MRIs. Xenograft development and cell culture expansion were attempted for each patient. RESULTS: Nineteen patients (6 females; median age 6 years; range 4–25 years) followed therapy recommendations. Average mapped coverages for WES were 490X (tumor) and 194X (germline). An average of 245,951,030 total mapped reads was achieved for tumor RNA. Panobinostat was the most commonly recommended drug (n=12; 63%). The multi-agent therapy plan was well tolerated with mainly grade 1/2 AEs. Grade 3/4 AEs were predominantly hematologic, including thrombocytopenia (n=13; 22%) and neutropenia (n=16; 27%). The OS12 did not significantly differ from historical controls (OS12 0.47; 95% CI 0.24, 0.70) or from a feasibility cohort that did not follow tumor board recommendations (OS12 0.66; 95% CI 0.36, 0.97; p=0.34). Fourteen DIPG cell lines were established. CtDNA analysis at diagnosis confirmed the H3K27M mutation in 16/20 patients, known to harbor H3K27M in tumor. CONCLUSION: A multi-agent therapy recommendation based on WES and RNAseq analysis of DIPG tumors is feasible; however, there is no demonstrated impact on clinical outcome. This is likely due to lack of efficacious FDA-approved drugs that target key mutations in DIPG, as well as molecular evolution of these tumors.

Purpose: Imetelstat is a covalently-lipidated 13-mer thiophosphoramidate oligonucleotide that acts as a potent specific inhibitor of telomerase. It binds with high affinity to the template region of the RNA component of human telomerase (hTERC) and is a competitive inhibitor of telomerase enzymatic activity. The purpose of this study was to determine the recommended phase II dose of imetelstat in children with recurrent or refractory solid tumors. Experimental Design: Imetelstat was administered intravenously more than two hours on days 1 and 8, every 21 days. Dose levels of 225, 285, and 360 mg/m2 were evaluated, using the rolling-six design. Imetelstat pharmacokinetic and correlative biology studies were also performed during the first cycle. Results: Twenty subjects were enrolled (median age, 14 years; range, 3–21). Seventeen were evaluable for toxicity. The most common toxicities were neutropenia, thrombocytopenia, and lymphopenia, with dose-limiting myelosuppression in 2 of 6 patients at 360 mg/m2. Pharmacokinetics is dose dependent with a lower clearance at the highest dose level. Telomerase inhibition was observed in peripheral blood mononuclear cells at 285 and 360 mg/m2. Two confirmed partial responses, osteosarcoma (n = 1) and Ewing sarcoma (n = 1), were observed. Conclusions: The recommended phase II dose of imetelstat given on days 1 and 8 of a 21-day cycle is 285 mg/m2. Clin Cancer Res; 19(23); 6578–84. ©2013 AACR.

Diffuse intrinsic pontine glioma (DIPG) is a tumor of the brainstem, specifically in the pons, accounting for 10–20% of all of central nervous system (CNS) tumors in children. Unfortunately, DIPG is the leading cause of death in children with CNS cancers. Clinical interventions trying to effectively treat children with DIPG have failed despite 40 years of clinical trials. The critical location of these tumors eliminates extensive surgical resection as an option. Radiation therapy (RT) is the standard of care, and although it improves the clinical symptoms of most patients, the results are temporary, with tumor progression typically occurring months post radiation. Given the dismal prognosis associated with this disease and the challenge to find chemotherapeutic agents, especially molecularly targeted drugs that improve the survival of the patients, there is a strong incentive to move new treatments forward into clinical trials. The more effective treatment would potentially involve combinatory therapeutic regimens with new epigenetic drugs that can offer synergistic benefits and potentially increase therapeu‐ tic efficacy. The increasing knowledge of genomic, epigenomic, and proteomic characteristics of DIPG is opening doors to new therapeutic avenues and provides hope and promise for this devastating childhood cancer.

OBJECTIVE: We evaluate the utility of biofluids (CSF, plasma, and urine) from patients and their xenograft models, for detection and quantification of circulating tumor DNA (ctDNA). The goal has been to develop a comprehensive panel for ctDNA detection capable of assessing response to treatment. METHODS: Biofluids (plasma, CSF, and urine) from patients diagnosed with low and high grade midline gliomas were collected for this study. Fifty seven specimens were collected. These include fifty plasma samples collected at diagnosis with each MRI, from 15 patients diagnosed with DIPG, five CSF samples from patients with high grade midline glioma, and two from patients with low grade midline glioma. Major mutations associated with DIPG were selected for monitoring ctDNA. These include H3F3A K27M, HIST1H3B K27M, ACVR1 G328V/R206H, and PPM1D E525X. Plasma and CSF samples collected at diagnosis from low grade glioma patients were analyzed for IDH1 R132H and BRAF V600E mutations. ctDNA was further assessed in preclinical models of DIPG. RESULTS: For monitoring ctDNA, we have developed a liquid biopsy technique using plasma and CSF. Using our sensitive and robust digital droplet PCR (ddPCR) system, we identified the presence of major histone 3 K27M partner mutations associated with DIPG in plasma and CSF ctDNA. We have also successfully conducted multiplexed detection of H3K27M and its partner mutations in CSF samples obtained from DIPG patients. In a high grade thalamic astrocytoma, where biopsy specimen was not available, we tested CSF and determined H3K27M mutation status of the tumor. Furthermore, we have detected H3K27M mutation in ctDNA of plasma samples in xenograft models of DIPG. CONCLUSIONS: Biofluid is a suitable medium for detection and quantification of circulating tumor DNA (ctDNA). Liquid biopsy has the potential to complement, or in some cases eliminate the need for biopsies, and may inform recurrence and tumor response to treatment.

Anaplastic astrocytomas are aggressive cancers of glial cells that present poor prognosis and high recurrence. Heterozygous mutations in IDH1 and IDH2 genes are common in the patients with anaplastic astrocytomas and glioblastomas. In the majority of cases, if not all, the IDH mutations seen in patients with brain tumors have been heterozygous somatic mutations. Patients with Ollier disease and Maffucci syndrome, which are rare enchondromatosis syndromes, have been shown to carry mosaic mutations of IDH1 or IDH2 in non-neoplastic tissues but no patients with anaplastic astrocytoma have been shown to carry such mosaicism. Here, we present two siblings with high grade astrocytoma with IDH1 R132H mutation. Our analysis of IDH1 R132H mutations in the siblings’ tumors and non-neoplastic tissues, including healthy regions of the brain, cheek cells and primary teeth has identified IDH1 R132H mosaicism, using digital PCR system. IDH1 R132H mutant allelic frequencies ranging from 0.01% to 0.64% were detected in teeth, cheek cells and non-tumorous brain tissue. The findings were confirmed by next-generation sequencing with similar single nucleotide variation frequencies. Our study demonstrated an example of IDH1 R132H mosaicism in anaplastic astrocytoma patients, which could have gone unnoticed by traditional sequencing technologies.

Pediatric diffuse midline glioma (MLG) constitutes the deadliest childhood CNS cancers. Due to the sensitive anatomical location, lack of surgical access to the tumor has hindered molecularly informed clinical interventions; thus, there is a dire need for the development of a sensitive and accurate approach for mapping tumor DNA fingerprints and monitoring tumor response to therapy. Liquid biopsy is emerging as a minimally invasive method for detecting circulating tumor DNA (ctDNA) in patients’ biofluid. To use liquid biopsy for mapping genomic mutations in pediatric brain tumors, we designed, optimized, and validated probes, and then showed their sensitivity and specificity for clinical assessment of children diagnosed with MLGs. Specifically, we demonstrated the feasibility of generating multiplex assays for detection of multiple driver mutations in patients’ biofluid. Tumor ctDNA was detected in cyst fluid (n=1), cerebrospinal fluid (CSF, n=30) and plasma (n=77), representing 48 subjects admitted or enrolled in a clinical trial. We show that while both CSF and plasma contain ctDNA, CSF tended to be a more enriched medium for tumor-associated ctDNA. Furthermore, we have detected H3K27M in ctDNA of plasma samples in xenograft models of DIPG. Finally, we show that ctDNA analysis closely agree with tumor response, often providing similar trends compared to MR imaging. Our data are the first to indicate the clinical utility of ctDNA for monitoring tumor response, and provide a molecularly-based rapid and sensitive approach for detection and longitudinal survey of childhood CNS tumors.