Medulloblastoma Patients Research Articles

Multiparametric MRI-based machine learning system of molecular subgroups and prognosis in medulloblastoma.

We aimed to use artificial intelligence to accurately identify molecular subgroups of medulloblastoma (MB), predict clinical outcomes, and incorporate deep learning-based imaging features into the risk stratification. The MRI features were extracted for molecular subgroups by a novel multi-parameter convolutional neural network (CNN) called Bi-ResNet-MB. Then, MR features were used to establish a prognosis model based on XGBoost. Finally, a novel risk stratification system to stratify the patients based on the M2R Score (Machine learning-based Medulloblastoma Risk Score) was proposed. A total of 139 MB patients (36 female, average age 7.27 ± 3.62 years) were treated at Beijing Tiantan Hospital. The Bi-ResNet-MB model excelled in molecular subgroup classification, achieving an average AUC of 0.946 (95% CI: 0.899-0.993). For prognostic prediction, our models achieved AUCs of 0.840 (95% CI: 0.792-0.888), 0.949 (95% CI: 0.899-0.999), and 0.960 (95% CI: 0.915-1.000) for OS, and 0.946 (95% CI: 0.905-0.987), 0.932 (95% CI: 0.875-0.989), and 0.964 (95% CI: 0.921-1.000) for PFS at 1, 3, and 5 years. In an independent validation dataset of 108 patients (33 female, average age 7.11 ± 2.92 years), the average AUC of molecular subgroup classification reached 0.894 (95% CI: 0.797-1.000). For PFS prediction at 1, 3, and 5 years, the AUCs were 0.832 (95% CI: 0.724-0.920), 0.875 (95% CI: 0.781-0.967), and 0.907 (95% CI: 0.760-1.000), respectively. Based on machine learning and MRI data, models for MB molecular subgroups and prognosis prediction and the novel risk stratification system may significantly benefit patients. Question Medulloblastoma exhibits significant heterogeneity, leading to considerable variations in patient prognosis and there is a lack of effective risk assessment strategies. Findings We have constructed a comprehensive machine learning system that excels in subgrouping diagnosis, prognosis assessment, and risk stratification for medulloblastoma patients preoperatively. Clinical relevance The utilization of non-invasive preoperative diagnosis and assessment is advantageous for clinicians in creating personalized treatment plans, particularly for high-risk patients. Additionally, it lays a foundation for the subsequent implementation of neoadjuvant therapy for medulloblastoma.

Demographic, medical, and neighborhood barriers to clinical neuropsychological services in pediatric medulloblastoma patients treated in the United States.

Medulloblastoma is the most commonly occurring malignant brain tumor of childhood. Treatment includes a combination of surgery, radiation therapy, and chemotherapy, all of which are associated with cognitive impairments. Despite appreciation of the value of neuropsychological evaluations to assess for cognitive impairments, there are barriers to these evaluations. This study aimed to identify potential demographic, medical, or neighborhood factors associated with barriers to neuropsychological evaluations. Retrospective chart review identified pediatric patients with a medulloblastoma diagnosed between 2005 and 2024. Demographic, medical, and neighborhood factors were gleaned from the medical records. Of the 67 identified patients, only 36 (53.73%) completed a clinical neuropsychological evaluation. Patients who obtained a neuropsychological evaluation were more likely to be non-Hispanic [X2 (1,62) = 9.20, p = <.01], diagnosed at younger ages [t(60) = - 8.40, p = < 0.001], treated with photon radiation therapy [X2 (1,62) = 15.24, p = < 0.001], live closer to the hospital [t(60) = 3.19, p = 0.001], and live in communities with higher health-related resources [t(60) = - 2.09, p = 0.02]. The two groups did not differ by any of the medical/clinical factors. Neuropsychological evaluations assess for cognitive changes following cancer directed treatment, clarify other diagnostic possibilities (e.g., learning or attention difficulties), and facilitate access to appropriate accommodations and school- and community- based services. Although neuropsychological evaluations are recommended as standard of care following treatment for medulloblastoma, we identified demographic, medical, and neighborhood-level factors that serve as barriers to clinical neuropsychological evaluations. Implications for clinical care and recommendations to address these barriers are discussed.

In-depth inference of transcriptional regulatory networks reveals NPM1 as a therapeutic ribosomal regulator in MYC-amplified medulloblastoma

Medulloblastoma (MB) is an aggressive pediatric brain tumor with distinct molecular heterogeneity. Identifying subtype-specific signatures within Group 3 and Group 4 remains challenging due to shared cytogenetic alterations and limitations of conventional differential gene expression analysis. To uncover the underlying molecular signatures and hidden regulators, we used the Cavalli transcriptomic profile of 470 Group 3 and Group 4 MB patients to reconstruct subtype-specific regulatory networks. A strong upregulation of the ribosomal pathway was linked to MYC amplification in Group 3, with Nucleophosmin 1 (NPM1) emerging as a key regulator. NPM1 upregulation defined a subset of Group3 and Group4 patients with poor prognosis. Inhibition of NPM1 led to apoptosis, reduced c-Myc stability, and impaired translation in MYC-amplified Group 3 MB cells. Together, our findings highlight NPM1 as a promising therapeutic target and provide new insights into the regulatory mechanisms in MB.

DHODH Inhibition Suppresses MYC and Inhibits the Growth of Medulloblastoma in a Novel In Vivo Zebrafish Model.

Medulloblastoma (MB) is the most common high-grade paediatric brain tumour, with group 3 MB patients having the worst prognosis. A high prevalence of group 3 tumours shows overexpression of the MYC oncogene, making it a potential therapeutic target. However, attempts to directly inhibit MYC have so far demonstrated limited success. Dihydroorotate dehydrogenase (DHODH), a crucial enzyme of the pyrimidine biosynthesis process, has emerged as an up-and-coming target in oncology, as its inhibition has shown promise in several cancers. In this study, we investigated the efficacy of brequinar, a DHODH inhibitor, in MB, with a focus on group 3. In vitro, BRQ's effects on cell viability and MYC expression were tested in seven MB cell lines. In vivo, a novel zebrafish xenograft model was used to evaluate BRQ's impact on tumour growth and toxicity. High DHODH expression was identified in group 3 and shh MB subgroups, correlating with poor survival and MYC expression. BRQ demonstrated nanomolar efficacy in inducing apoptosis and reducing MYC expression in group 3 MB cell lines. Finally, we established a novel zebrafish xenograft model and demonstrated that BRQ significantly inhibited tumour growth at non-toxic concentrations in vivo, particularly in the D458 metastatic MB cell line. Our findings indicate that DHODH is a promising therapeutic target in group 3 MBs. Furthermore, BRQ shows potential for clinical application, effectively reducing tumour growth and MYC expression in vitro and in vivo. Moreover, our newly established zebrafish xenograft model offers a promising avenue for rapid in vivo drug testing for use in MB.

Genomic landscape of medulloblastoma subtypes in an Asian cohort.

Medulloblastoma (MB) is a highly malignant childhood brain tumor. Previous research on the genetic underpinnings of MB subtypes has predominantly focused on European and American cohorts. Given the notable genetic differences between Asian and other populations, a subtype-specific study on an Asian cohort is essential to provide comprehensive insights into MB within this demographic. The aim of this study is to investigate the genomic landscape of MB subtypes in an Asian cohort to better understand the genetic variations and potential implications for clinical practice. We conducted a study on an Asian cohort comprising 113 MB patients. Genomic sequencing was performed using MGISEQ-2000 platform. We analyzed the participants' characteristics and compared them with previous studies. All germline variants of the ten susceptibility genes of interest (APC, BRCA2, PTCH1, PTCH2, ELP1, SUFU, CTNNB1, SMARCA4, GPR161, and TP53) were annotated and validated. Our study identified 14 valid germline variants that met our criteria, with these variants being detected in the genes APC, BRCA2, PTCH1, PTCH2, ELP1, and SUFU. Of these, six variants were classified as pathogenic in ClinVar: two in PTCH2 (c.C1573T), one in ELP1 (c.C583T), and three in PTCH1 (c.G1370T, c.C2066T, c.C529T). The remaining eight variants were of uncertain significance, including those in SUFU (c.T833C), ELP1 (c.T2A), BRCA2 (c.G7488C), and APC (c.C3247A, c.A1G, c.A8042G, c.A3056G, c.G822C). Our findings highlight a subtype-based germline variant landscape specific to the Asian cohort and reinforce the connection between SUFU, PTCH1, and the SHH subtype of MB. Additionally, the identification of ELP1-related cases supports the newest findings in this area and provides typical copy number variation (CNV) results for future investigation. This study provides valuable insights into the genetic landscape of MB in an Asian cohort, emphasizing the importance of population-specific research. The subtype-specific germline variant landscape identified in this study contributes to the understanding of MB and its genetic underpinnings in Asian populations, potentially guiding future research and therapeutic strategies.

Inter-Fraction Motion and Dosimetric Analysis of Volumetric Modulated Arc Therapy for Craniospinal Irradiation in Adult Medulloblastoma Patients.

Background/Objectives. Adult medulloblastoma (AMB) patients should receive postoperative craniospinal irradiation (CSI) as a standard treatment. Volumetric intensity-modulated arc therapy (VMAT) is a promising method for CSI. This report summarizes the repositioning and dosimetric data outcomes for six AMB patients. Methods. Complete CSI and posterior cranial fossa irradiation, or tumor bed boost irradiation with Linac-based VMAT, was performed and evaluated. Patients were immobilized in the supine position with two thermoplastic masks (head-neck and abdomen). To ensure inter-fraction reproducibility during radiotherapy (RT), a single cone-beam CT (CBCT) scan for each isocenter and real-time surface-guided RT using AlignRT® were performed daily before and during the RT session. Match values of all three translational axes (x = lateral, y = longitudinal, z = vertical) were recorded. Results. From August 2022 to September 2023, six AMB patients were treated with CSI: three women and three men with a median age of 32 (22-42). All cases were classical MB, four were low risk, and two were defined as high risk due to the metastatic disease. All patients underwent surgery; two received a gross total resection. Low-risk patients received 36 Gy for CSI and a 54 Gy boost, while high-risk patients received 39 Gy for CSI. No significant toxicities greater than G2 were observed during RT, and only two cases reported decreased platelet counts. The dose to the organs at risk was low and acceptable. The mean dose to the heart, lungs, eyes, stomach, and thyroid were 4.4 Gy, 8.5 Gy, 12 Gy, 8.7 Gy, and 11 Gy, respectively. In terms of repositioning data, 124 CBCT scans were analyzed. Inter-fraction CBCT mean values for the study population in all translational directions were inferior to 2 mm in more than 90% of cases. Conclusions. VMAT is a convenient and effective treatment for AMB. Positioning and immobilization with masks (head and neck plus abdomen) reduce inter-fraction motion.

Dissecting the Impact of Genetic Background on Oncogenic Response to Radiation Exposure in the Ptch1+/- Mouse Model.

Medulloblastoma (MB) is a common primary brain cancer in children. The sonic hedgehog (SHH) pathway is indispensable for the normal development of the cerebellum, and MB is often caused by persistent SHH activation owing to mutations in pathway components. Patched1 (PTCH1) is the primary receptor for the SHH ligand and a negative regulator of the SHH signal transduction pathway. Mice heterozygous for the Ptch1 gene (Ptch1+/-) are predisposed to MB development. Irradiation of newborn Ptch1+/- mice dramatically increases MB occurrence. A genetic background carrying the Ptch1 mutation significantly influences the risk of developing MB. This study aims to investigate the genetic background-related mechanisms that regulate radiation-induced cellular response and oncogenesis in the cerebellum. We employed multiple approaches, including: (a) analysis of cellular radiosensitivity in granule cell precursors (GCPs), the MB cells of origin, derived from Ptch1 mice with a genetic background that is sensitive (CD1) or resistant (C57Bl/6) to the induction of radiogenic MB; (b) identification of genes differentially expressed in spontaneous and radiation-induced MBs from these two mouse strains; (c) bioinformatic analysis to correlate the expression of radiation-induced genes with survival in MB patients; and (d) examining the expression of these genes in ex vivo MBs induced by single or repeated radiation doses. We have identified a potential gene expression signature-Trp53bp1, Bax, Cyclin D1, p21, and Nanog-that influences tumor response. In ex vivo cultured spontaneous MBs, the expression levels of these genes increase after irradiation in CD1 mice, but not in mice with a C57Bl/6 genetic background, suggesting that this signature could predict tumor response to radiation therapy and help develop strategies for targeting DNA damage repair in tumors. A detailed understanding of the mechanisms behind genetic background-related susceptibility to radiation-induced oncogenic responses is crucial for translational research.

High Expression of GABAA Receptor β Subunit Genes Is Associated with Longer Overall Survival in Medulloblastoma.

Background/Objectives: Most of the rapid inhibitory neurotransmission in the brain is mediated through activation of the γ-aminobutyric acid (GABA) type A (GABAA) receptor, which is a ligand-gated ion channel. GABAA receptor activation via GABA binding allows for an intracellular influx of Cl- ions, thus inducing cellular hyperpolarization. Each GABAA receptor consists of a combination of five subunits, and several subunits have been proposed as biomarkers and therapeutic targets in cancer. Here, we show the expression of genes encoding β subunits of the GABAA receptor, namely GABRB1, GABRB2, and GABRB3, across the four different molecular subgroups of medulloblastoma (MB), which is the most common malignant pediatric brain tumor. We also show the associations of GABAA receptor β subunits with MB patients' overall survival (OS). Methods: The expression of genes encoding GABAA receptor β subunits was analyzed using a previously described dataset comprising 763 MB tumor samples. Patients were classified into high- and low-gene-expression groups, and the Kaplan-Meier estimate was used to examine the relationship between gene expression levels and patient OS. Results: High GABRB1 expression was associated with better OS within each of the four molecular subgroups. The GABRB2 gene showed higher transcript levels in Group 3 MB compared to all other subgroups, and high expression was associated with better prognosis in Group 3 tumors. GABRB3 expression was significantly higher in Group 3 and Group 4 MB, and high expression of GABRB3 genes was associated with longer OS in the sonic hedgehog (SHH) subgroup. The high expression of GABRB1, GABRB2, and GABRB3 is associated with longer patient OS in a subgroup-specific manner. Conclusions: These results indicate a role for GABAA receptors containing β subunits in influencing MB progression.

Abstract B016: RNA demethylase ALKBH5 drives medulloblastoma growth and progression

Abstract Introduction: Medulloblastoma (MB) is the most common malignant childhood brain tumor. It accounts for 20% of all childhood cancer deaths. MB are very fast-growing tumors, which spreads to central nervous system through cerebrospinal fluid, leading to leptomeningeal metastases that occurs up to 66% in brain cancer patients. Despite the progress in treating MB, the 5-year survival rate for high-risk MB remains poor with high recurrence. Moreover, the quality of life for those kids who do survive is substantially reduced due to the high toxicity associated with the high radiation exposure and multiple drug chemotherapy they must endure at such an early age. Therefore, it is critical to identify novel factors and understand previously undefined mechanisms that drive MB growth and progression, so that safe and viable therapeutics can be developed for treating MB. In this study, we tested the hypothesis that epigenetic changes, specifically mRNA demethylation mediated by the RNA demethylase ALKBH5, play an important role in MB growth and tumor progression, and hence approaches aimed at inhibiting ALKBH5 activity could prove clinical utility for MB patients. Methods: Medulloblastoma cell lines (HD-MB03, D556, D425 and DAOY) were purchased from the ATCC. MB cancer stem cells (MB-CSC) were stained Aldefluor sorted by Flow cytometry and cultured in stem cell medium. MB cells were transfected either with ALKBH5 overexpression (OE) plasmid or siRNA (KD). These OE/KD MB cells were analyzed for cell viability, migration, invasion, colony formation, cell cycle, apoptosis assays, RNA sequencing, RT-qPCR, western blotting, RNA immunoprecipitation, and in vivo tumor xenograft study. Results: Our results revealed that MB cells are highly dependent on ALKBH5 for their survival. Using multiple MB cell lines with or without ALKBH5 depletion, we discovered that of ALKBH5 inhibited both short and long-term growth of MB cells. In addition, knockdown of ALKBH5 inhibited the migration of MB cells. ALKBH5 silenced MB cells showed increased apoptosis. Importantly, we show that ALKBH5 silencing suppressed the self-renewal/proliferation of MB stem cells (including medullosphere formation). ALKBH5 knockdown HDMB03-GFP-luciferase cells injected into mouse showed reduced tumor growth compared to the control. RNA sequencing analysis of ALKBH5 depleted MB cells showed alterations in several metabolism related pathways. Further, ALKBH5-depletion led to significantly decreased levels of cancer stem cell marker proteins including Nanog, OCT4 and SOX9. These are significant findings as MB stem cells are considered to be the major source of MB initiation, maintenance, relapse and render MB cells resistant to radiation. Conclusion: Collectively, our study results suggest that RNA demethylase ALKBH5 may play an important role in MB growth and progression by supporting MB cancer (tumor initiating) stem cells (MB-CSC) and ALKBH5 serves as a novel therapeutic target for MB. Citation Format: Panneerdoss Subbarayalu, Daisy Medina, Pitta Prabhakar, Shahad Abdulsahib, Saif Nirzhor, Desiree Denman, Santosh Timilsina, Deepika Singh, Phat Do, Krishna Evani, Dhiya Billa, Esha Reddy, Peter Houghton, Yogesh Gupta, Yidong Chen, Suryavathi Viswanadhapalli, Ratna Vadlamudi, Andrew Brenner, Manjeet Rao. RNA demethylase ALKBH5 drives medulloblastoma growth and progression [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: RNAs as Drivers, Targets, and Therapeutics in Cancer; 2024 Nov 14-17; Bellevue, Washington. Philadelphia (PA): AACR; Mol Cancer Ther 2024;23(11_Suppl):Abstract nr B016.

Single-Molecule Detection of Serum MicroRNAs for Medulloblastoma with Biphasic Sandwich Hybridization-Assisted Plasmonic Resonant Scattering Imaging.

MicroRNA (miRNA) dysregulation is closely related to the occurrence and progression of medulloblastoma (MB). However, the full potential of serum circulating miRNAs in MB diagnosis is restricted by their ultralow abundance in peripheral blood due to blood-brain barrier. Here, we report the direct preamplification-free detection of aberrant expression of oncogenic miRNAs in serum from MB patients by proposing a simple yet robust single-molecule assay that combines biphasic sandwich hybridization in nucleic acids and the dark-field single-particle plasmonic imaging (B2S2PI). In this strategy, signal DNA was prehybridized with target miRNA in homogeneous solution to form sDNA-RNA complexes. Then the captured DNA strands with rationally adjusted surface densities could efficiently capture the sDNA-RNA complexes to generate a well-separated DNA-RNA sandwich structure. The combination of homogeneous and heterogeneous reactions enabled interface-mediated hybridization reactions to maintain molecular stability with fewer bases, making it suitable for the direct amplification-free assays of short miRNA targets. Labeling the DNA-RNA hybrids with plasmatic gold nanotags allowed nondestructive recognition and imaging of individual miRNA targets under mild conditions with high signal-to-noise ratio. By digitally counting and analyzing the bright plasmonic resonant scattering spots, B2S2PI enabled both the measurement of a low femtomolar concentration of circulating miRNA-21 in 5 μL sample volume within a turnaround of 2 h and the discrimination of single base mismatches. Moreover, B2S2PI was universal for detecting miRNAs with different sequences and secondary structures. Further analysis of clinical serum samples revealed that B2S2PI was capable of accurately distinguishing MB patients from noncancer controls with an area under the curve (AUC) of 0.99, which was superior to that of qRT-PCR. B2S2PI holds promise as a novel alternative means for single-molecule miRNA assay and sheds light on the circulating nucleic acid-based liquid biopsy of intracranial malignant tumors.

Machine learning-based models for prediction of survival in medulloblastoma: a systematic review and meta-analysis.

Medulloblastoma (MB) is the pediatric population's most frequent malignant intracranial lesions. Prognostication plays a crucial role in optimizing treatment strategy in the MB setting. Several studies have developed ML-based models to predict survival outcomes in individuals with MB. In this systematic review and meta-analysis study, we aimed to evaluate the role of ML-based models in predicting survival in MB patients. Literature records were retrieved on May 14th, 2024, using the relevant key terms without filters in PubMed, Embase, Scopus, and Web of Science. Records were screened according to the eligibility criteria, and the data from the included studies were extracted. The quality assessment was performed using the QUADAS-2 tool. The meta-analysis and sensitivity analysis were conducted using R software. Six studies were included, with 2771 patients ranging from 46 to 1759 individuals. A total of 23 ML and DL models were developed, 20 of which were ML and three DL. Random forest (RF) was the most frequent classifier, as it was utilized in nine models, followed by support vector machine (SVM). Eight models were included in the meta-analysis. Our meta-analysis revealed a pooled AUC of 0.77 (95% CI: 0.75-0.80). In addition, the radionics-based and genomics-based models had a pooled AUC of 0.77 (95% CI: 0.76-079) and 0.76 (0.63-0.88), respectively (P = 0.77). Our results suggested that ML-based models, especially ML algorithms, could play a vital and efficient role in the prediction of survival of patients based on radiomics and genomics.

CTIM-20. IDENTIFYING CANDIDATE ANTIGEN TARGETS FOR PEDIATRIC BRAIN TUMOR IMMUNOTHERAPIES THROUGH COMPREHENSIVE IMMUNOGENOMIC ANALYSIS

Abstract BACKGROUND Identifying tumor rejection antigens continues to be a significant challenge in the development of effective immunotherapies and their application to pediatric brain tumors. METHODS To identify candidate antigen targets for Medulloblastoma (primary MB, n=170; recurrent MB, n=21), Diffuse Intrinsic Pontine Glioma (DIPG, n=10), and Pediatric High-Grade Astrocytoma (HGA, n=12) for adoptive cellular therapy, we conducted a comprehensive immunogenomic analysis of the transcription profiles of the most aggressive pediatric brain tumors and performed in silico antigen prediction across a wide range of antigen classes. IMPORTANCE OF THE STUDY Pediatric brain tumors are heterogeneous diseases, and current standard treatments do not adequately address this variability. Newer and safer patient-specific treatment approaches are needed for high-risk patients who do not respond to standard therapies. Cellular immunotherapy could be crucial for improving survival and reducing morbidity. For an effective immune response against these brain tumors, appropriate tumor antigens must be targeted. While subgroup-specific genetic mutations in medulloblastoma patients have been reported, the immunogenicity of these genetic alterations remains unknown. Using a custom antigen prediction pipeline, we identified potential tumor rejection antigens with significant implications for the development of cellular therapies for MB, DIPG, and HGA. RESULTS Antigen prediction analysis revealed that most patients express few candidate neoantigen targets that pass all filtering criteria. Overall, the proportion of predicted tumor-associated antigens (TAAs) was higher than that of neoantigens and gene fusions for all molecular subtypes, except for SHH, which exhibited a higher neoantigen burden. Notably, cancer-testis antigens and previously unrecognized neurodevelopmental antigens were found to be expressed in most patients across all medulloblastoma subgroups. Although the absolute immune cell content is predicted to be low, immune gene-signature analysis revealed subgroup-specific differences.

TMIC-17. TARGETING THE FLUIDIC FORCE-SENSING MECHANISM TO TREAT BRAIN TUMOR METASTASIS

Abstract Biofluid flow generates fluid shear stress (FSS), a mechanical force widely present in tissue microenvironment. How brain tumor growth alters the conduit of biofluid, thereby impacting FSS-regulated cancer progression is unknown. Medulloblastoma (MB) is the most common malignant brain tumor in children. Dissemination of MB cells into the cerebrospinal fluid (CSF) initiates metastasis within the central nervous system. By simulating CSF dynamics based on magnetic resonance imaging of MB patients, we discovered that FSS is elevated at the cervicomedullary junction. Strikingly, we found that MB-relevant FSS promotes metastasis along mouse spinal cords. Mechanistically, FSS induces metastatic cell behaviours, including weakened cell-substrate adhesion, increased motility, cell clustering, and plasma membrane localization of glucose transporter 1 (GLUT1) to enhance glucose uptake. FSS is perceived by mechanosensitive ion channel PIEZO2, which drives actomyosin contractility-dependent GLUT1 recruitment at the plasma membrane. Genetic targeting of PIEZO2 or pharmacologic inhibition of GLUT1 mitigates metastasis. Collectively, these findings define a targetable FSS-activated mechano-metastatic cascade for the treatment of MB metastasis.

STEM-08. LIPID SATURATION HOMEOSTASIS PRESENTS A TARGETABLE VULNERABILITY IN MYC-AMPLIFIED GROUP 3 MEDULLOBLASTOMA

Abstract Medulloblastoma (MB) is the most common malignant pediatric brain cancer, originating from the cerebellum. Despite advancements in standard of care (SoC), MB remains fatal for 30% of patients. Tumor relapse, spinal metastasis and treatment resistance are the most prevalent in MYC-driven Group 3 MB (G3-MB). Patients surviving SoC are faced with life-long neurocognitive and neurodevelopmental deficits. These issues highlight the urgent need for improved treatment modalities. Reprogramming of cellular lipid metabolism is an emerging hallmark of cancer and may yield novel cancer-specific therapeutic options. Here we explore the lipidome of both G3-MB and its proposed cell of origin, human neural stem cells (hNSCs) by comparing untargeted lipidomics using Liquid-Chromatography-Mass Spectrometry (LC-MS). Comparative analyses revealed a differential abundance of distinct lipid species in G3-MB, with an overall reduced saturation level of fatty acids (FAs). These findings implicate the de novo lipid synthesis (DNL) pathway. We identified the enzymes involved in DNL to be essential for MB survival in our genome-wide CRISPR KO screen. Additionally, mRNA expression of the DNL enzymes increases at relapse in our SoC-adapted murine patient-derived xenograft (PDX) model. Pharmacological and genetic targeting of the DNL enzyme Stearoyl-CoA Desaturase 1 (SCD1) selectively targets G3-MB. Furthermore, small molecule treatment of SCD1 demonstrates efficacy against G3-MB PDX models in vivo. We identified SCD expression as a prognostic marker in Group 3 and 4 MB patients and identified possible pathways for MB treatment. Overall, these findings indicate that SCD1 is a potent target for the treatment of G3-MB.

EXTH-24. THERAPEUTIC TARGETING OF MISLOCALIZED NUCLEAR ENVELOPE PROTEINS INMYC-DRIVEN GROUP 3 MEDULLOBLASTOMA

Abstract Medulloblastoma (MB) is the most common, malignant pediatric brain tumour comprised of four distinct molecular subgroups (WNT, SHH, Group 3, and Group 4). A subset of Group 3 MB tumors harbors focal amplifications of the MYC oncogene (MYC-G3MB) and are particularly prone to tumour recurrence due to their highly proliferative nature linked to its embryonic stem cell-of-origin and programming. We discovered that LBR, an inner nuclear transmembrane protein, while present in all cells at the nuclear envelope (NE), is aberrantly presented to the cell surface in MYC-G3MBs. Transcriptomic profiling of publicly available datasets identified significant NE protein enrichment in MYC-G3MB compared to normal tissue and high expression during fetal development that dampens postnatally and in adult samples. Immunohistochemistry analysis of LBR on patient tumors was able to significantly prognosticate overall and progression free survival. Flow cytometry analysis identified LBR cell surface expression in MYC-G3MBs with low to no expression in human neural stem cells and other MB subgroups. High resolution microscopy of endogenously tagged LBR revealed cell surface presentation to be linked to ER-like vesicles that directly trafficked to the cell surface in MYC-G3MBs. Transcriptomic analysis of LBR cell surface positive and negative cells identified LBR to be linked to mitotic processes and mislocalization was significantly enriched following exposure to chemotherapy and radiation in vitro and in vivo. Therefore, we demonstrate that LBR is an ideal therapeutic target for patients facing treatment-refractory MB and have generated single domain antibodies (sdAb) to develop LBR-CAR T cells. Validation of LBR sdAbs demonstrate on-target binding and comparable affinity and avidity to commercial LBR antibodies. There is an undeniable paucity of targets for MB patients stemming from the lack of neoantigens and genomic aberrations in pediatric tumors. The preclinical development and validation of novel immunotherapeutic targets such as LBR provides alternative therapies for patients otherwise facing palliation.

DNAR-15. GROUP 3 MEDULLOBLASTOMAS EXPLOIT A TRANSIENT MECHANISM OF TELOMERE REPAIR MEDIATED BY ZSCAN4 WHICH IS TARGETABLE FOR THERAPEUTIC BENEFIT WITH BRAIN-PENETRANT DRUGS

Abstract All cancers need some mechanism of telomere repair. However, it is unknown how Group 3 (G3) medulloblastomas (MB) repair their telomeres, even though MB is the most common malignant brain tumor in kids and G3 is its most aggressive subtype. With rare exceptions, G3 MBs lack telomerase gain-of-function alterations or evidence of constitutive alternative-lengthening-of-telomeres pathway activation. We profiled tumors from N=38 MB patients from UCSF via single-nucleus RNA sequencing; 13 cases were also subjected to single-cell assay for transposase-accessible chromatin by sequencing, and 4 were subjected to single-cell cleavage under targets and tagmentation for H3K27ac. These data identified a network of gene enhancers that were specifically active in rhombic-lip progenitor-like MB cells, the putative G3 MB cell of origin. This program was driven by zinc finger and SCAN domain containing 4 (ZSCAN4), a gene reported to drive transient telomere repair during zygotic genome activation in early embryos. CRISPR inhibition of ZSCAN4 led to significant telomere shortening, telomeric DNA damage, telomere fusions, decreased proliferation, and increased survival of orthotopic xenografts. Conversely, ZSCAN4 over-expression induced significant telomere elongation, increased proliferation, and decreased survival in vivo. Analysis of endogenous-reporter cells indicated that ZSCAN4 is transiently expressed in response to telomere shortening and increases global histone acetylation by inducing chromatin-remodeling factors. Rebastinib targeted ZSCAN4, histone acetylation, and telomere repair, was brain penetrant, and significantly enhanced orthotopic-xenograft survival with no toxicity. Acetyltransferase inhibitor CPI-1612 was brain penetrant and extended xenograft survival with no toxicity. Our data support the hypothesis that G3 MBs exploit a transient mechanism of telomere repair mediated by ZSCAN4. Targeting ZSCAN4 or downstream histone acetylation is therapeutically viable. Pan-cancer whole-genome sequencing data showed that MBs have lower telomeric content than most other cancers. Thus, MBs may be particularly vulnerable to therapeutic strategies that inhibit telomere repair.

Risk factors for treatment-related sensorineural hearing loss and hearing aid use in medulloblastoma patients: an observational cohort study.

This study aimed to analyze treatment-related risk factors for sensorineural hearing loss (SNHL) and an indication for hearing aids (IHA) in medulloblastoma patients after craniospinal radiotherapy (CSRT) and platin-based chemotherapy (PCth). Atotal of 58patients (116 ears) with medulloblastoma and clinically non-relevant pre-treatment hearing thresholds were included. Cranial radiotherapy and PCth were applied sequentially according to the HIT 2000 study protocol or post-study recommendations, the NOA-07 protocol, or the PNET (primitive neuroectodermal tumor) 5MB therapy protocol. Audiological outcomes up to amaximum post-therapeutic follow-up of 4years were assessed. The incidence, post-treatment progression, and time-to-onset of SNHL, defined as Muenster classification grade ≥MS2b, were evaluated. Risk factors for IHA were analyzed separately. While 39patients received conventionally fractionated RT (CFRT; group1), 19patients received hyperfractionated RT (HFRT; group2). Over amedian follow-up of 40months, 69.2% of ears in group1 experienced SNHL ≥MS2b compared to 89.5% in group2 (p = 0.017). In multivariable Cox regressions analysis, younger age and increased mean cochlear radiation dose calculated as the equivalent dose in 2‑Gy fractions (EQD2) were associated with time-to-onset of SNHL ≥MS2b (p = 0.019 and p = 0.023, respectively) and IHA (p < 0.001 and p = 0.016, respectively). Tomotherapy and supine positioning were associated with alower risk for IHA in univariable modelling only (p = 0.048 and p = 0.027, respectively). Young age and cochlear EQD2 Dmean ≥40 Gy are significant risk factors for the incidence, degree, and time-to-event of SNHL as well as for IHA in medulloblastoma patients.

Development of an orthotopic medulloblastoma zebrafish model for rapid drug testing.

Medulloblastoma (MB) is one of the most common malignant brain tumors in children. Current preclinical in vivo model systems for MB have increased our understanding of molecular mechanisms regulating MB development. However, they may not be suitable for large-scale studies. The aim of this study was to investigate if a zebrafish-based xenograft model can recapitulate MB growth and enable rapid drug testing. Nine different MB cell lines or patient-derived cells were transplanted into blastula-stage zebrafish embryos. Tumor development and migration were then monitored using live imaging. RNA sequencing was performed to investigate transcriptome changes after conditioning cells in neural stem cell-like medium. Furthermore, drug treatments were tested in a 96-well format. We demonstrate here that transplantation of MB cells into the blastula stage of zebrafish embryos leads to orthotopic tumor growth that can be observed within 24 hours after transplantation. Importantly, the homing of transplanted cells to the hindbrain region and the aggressiveness of tumor growth are enhanced by pre-culturing cells in a neural stem cell-like medium. The change in culture conditions rewires the transcriptome towards a more migratory and neuronal phenotype, including the expression of guidance molecules SEMA3A and EFNB1, both of which correlate with lower overall survival in MB patients. Furthermore, we highlight that the orthotopic zebrafish MB model has the potential to be used for rapid drug testing. Blastula-stage zebrafish MB xenografts present an alternative to current MB mouse xenograft models, enabling quick evaluation of tumor cell growth, neurotropism, and drug efficacy.

Proteomic profiling of cerebrospinal fluid reveals TKT as a potential biomarker for medulloblastoma

Cerebrospinal fluid (CSF) plays an important role in brain tumors, including medulloblastoma (MBL). Recent advancements in mass spectrometry systems and ‘Omics’ data analysis methods enable unbiased, high proteome depth research. We conducted proteomic profiling of the total CSF in MBL patients with the purpose of finding a potential diagnostic biomarker for MBL. We quantified 1112 proteins per CSF sample. Feature selection identified four elevated soluble proteins (SPTBN1, HSP90AA1, TKT, and NME1-NME2) in MBL CSF. Validation with ELISA confirmed that TKT was significantly elevated in MBL. Additionally, TKT-positive extracellular vesicles were significantly enriched in MBL CSF and correlated with the burden of leptomeningeal seeding. Our results provide insights into the proteomics data of the total CSF of MBL patients. Furthermore, we identified the significance of TKT within the total CSF and its presence within circulating EVs in the CSF. We suggest that TKT may serve as a biomarker for MBL.

Identification of tumor rejection antigens and the immunologic landscape of medulloblastoma

BackgroundThe current standard of care treatments for medulloblastoma are insufficient as these do not take tumor heterogeneity into account. Newer, safer, patient-specific treatment approaches are required to treat high-risk medulloblastoma patients who are not cured by the standard therapies. Immunotherapy is a promising treatment modality that could be key to improving survival and avoiding morbidity. For an effective immune response, appropriate tumor antigens must be targeted. While medulloblastoma patients with subgroup-specific genetic substitutions have been previously reported, the immunogenicity of these genetic alterations remains unknown. The aim of this study is to identify potential tumor rejection antigens for the development of antigen-directed cellular therapies for medulloblastoma.MethodsWe developed a cancer immunogenomics pipeline and performed a comprehensive analysis of medulloblastoma subgroup-specific transcription profiles (n = 170, 18 WNT, 46 SHH, 41 Group 3, and 65 Group 4 patient tumors) available through International Cancer Genome Consortium (ICGC) and European Genome-Phenome Archive (EGA). We performed in silico antigen prediction across a broad array of antigen classes including neoantigens, tumor-associated antigens (TAAs), and fusion proteins. Furthermore, we evaluated the antigen processing and presentation pathway in tumor cells and the immune infiltrating cell landscape using the latest computational deconvolution methods.ResultsMedulloblastoma patients were found to express multiple private and shared immunogenic antigens. The proportion of predicted TAAs was higher than neoantigens and gene fusions for all molecular subgroups, except for sonic hedgehog (SHH), which had a higher neoantigen burden. Importantly, cancer-testis antigens, as well as previously unappreciated neurodevelopmental antigens, were found to be expressed by most patients across all medulloblastoma subgroups. Despite being immunologically cold, medulloblastoma subgroups were found to have distinct immune cell gene signatures.ConclusionsUsing a custom antigen prediction pipeline, we identified potential tumor rejection antigens with important implications for the development of immunotherapy for medulloblastoma.