The heterogeneous clinical behavior and biology of neuroblastoma as well as the relative paucity of somatic mutations render the development of effective therapies a challenge. As the genetic landscape of neuroblastoma is unraveled, certain aberrations have emerged as potent targets in the clinic. However, the challenge lies in the ability to determine the functional relevance of such mutations and to determine which ones to target and with what agents. The identification of treatment-sensitive mutations in the ALK tyrosine kinase receptor that opened neuroblastoma to targeted therapy offers a hallmark example of a therapeutically actionable genetic aberration. Attempts at targeting the effects of the oncogenic driver, MYCN, are also emerging—in terms of strategies to block its transcriptional program and its protein stabilization. However, the advent of targeted therapy has also ushered in the concept of resistance that needs to be addressed preemptively. This chapter summarizes the current status of therapies under investigation and in clinical use to target genetic aberrations in neuroblastoma.

10023 Background: Rapid and accurate identification of morphologic features of neuroblastic tumors (NTs) is critical for risk stratification and therapeutic decision making. The prognostic value of features like neuroblast differentiation, mitosis-karyorrhexis index (MKI), and Schwannian stromal presence is well established. Deep learning permits objective histopathological analysis, streamlining workflows for pathologists, notably in rare cancers. In rare cancers, our method minimizes bias and optimizes limited data using transfer and self-supervised learning (SSL) for feature extraction, with improved explainability. Here, we used an artificial intelligence-based model to morphologically classify NT tumors and MYCN-amplification. Methods: Annotated H&E-stained slides of diagnostic NT tumor biopsies from the University of Chicago and the Children’s Oncology Group were digitalized. Pathologists defined three binarized measures including diagnostic category (ganglioneuroblastoma/neuroblastoma), grade (differentiating/poorly differentiating), and MKI (low and intermediate/high). MYCN status was abstracted from patient records (amplified/non-amplified). Using Slideflow, our open-source pipeline, we developed an attention-based multiple instance learning model with features extracted by CTransPath, a SSL model pretrained on pan-cancer images from The Cancer Genome Atlas. For each measure, model performance was evaluated using 5-fold cross validation by aggregating k-fold model predictions across multiple metrics. Patients were excluded from a model if the measure of interest was unknown. Feature significance was assessed visually using Class Activation Mapping (Grad-CAM). Results: The mean age of the study cohort (n = 172) was 3.66 years. Of patients with clinical information, 84 of 138 (60.2%) had metastatic disease and 94 of 133 (70.7%) were high-risk. Of the 148 tumors with a diagnostic category of neuroblastoma, 93.2% were poorly differentiated and 25% had high MKI. Of the 135 tumors with known MYCN status, 40 were amplified (29.6%). The final models excelled across all outcomes, performing best for diagnostic category, grade, and MYCN status (Table 1). Physician review of the attention-based heatmaps for all measures highlighted biologically relevant regions such as neuropil. Conclusions: We created a deep learning pipeline for auto-characterization of digitized H&E-stained NT pathology slides. Our approach may also aid in identifying molecular features including MYCN-amplification. Review of heatmaps showed pertinent biological tissue, boosting model reliability.[Table: see text]

Supplementary Figures S1-S15 and Legends

AbstractPurpose: The importance of cellular context to the synergy of DNA damage response (DDR)-targeted agents is important for tumors with mutations in DDR pathways, but less well-established for tumors driven by oncogenic transcription factors. In this study, we exploit the widespread transcriptional dysregulation of the EWS-FLI1 transcription factor to identify an effective DDR-targeted combination therapy for Ewing sarcoma. Experimental Design: We used matrix drug screening to evaluate synergy between a DNA-PK inhibitor (M9831) or an ATR inhibitor (berzosertib) and chemotherapy. The combination of berzosertib and cisplatin was selected for broad synergy, mechanistically evaluated for Ewing sarcoma selectivity, and optimized for in vivo schedule. Results: Berzosertib combined with cisplatin demonstrates profound synergy in multiple Ewing sarcoma cell lines at clinically achievable concentrations. The synergy is due to loss of expression of the ATR downstream target CHEK1, loss of cell-cycle check-points, and mitotic catastrophe. Consistent with the goals of the project, EWS-FLI1 drives the expression of CHEK1 and five other ATR pathway members. The loss of CHEK1 expression is not due to transcriptional repression and instead caused by degradation coupled with suppression of protein translation. The profound synergy is realized in vivo with a novel optimized schedule of this combination in subsets of Ewing sarcoma models, leading to durable complete responses in 50% of animals bearing two different Ewing sarcoma xenografts. Conclusions: These data exploit EWS-FLI1 driven alterations in cell context to broaden the therapeutic window of berzosertib and cisplatin to establish a promising combination therapy and a novel in vivo schedule. See related commentary by Ohmura and Grünewald, p. 3358

A deep learning model using attention-based multiple instance learning (aMIL) and self-supervised learning (SSL) was developed to perform pathologic classification of neuroblastic tumors and assess MYCN-amplification status using H&E-stained whole slide images from the largest reported cohort to date. The model showed promising performance in identifying diagnostic category, grade, mitosis-karyorrhexis index (MKI), and MYCN-amplification with validation on an external test dataset, suggesting potential for AI-assisted neuroblastoma classification.

The N6-methyladenosine (m6A) RNA modification is an important regulator of gene expression. m6A is deposited by a methyltransferase complex that includes methyltransferase-like 3 (METTL3) and methyltransferase-like 14 (METTL14). High levels of METTL3/METTL14 drive the growth of many types of adult cancer, and METTL3/METTL14 inhibitors are emerging as new anticancer agents. However, little is known about the m6A epitranscriptome or the role of the METTL3/METTL14 complex in neuroblastoma, a common pediatric cancer. Here, we show that METTL3 knockdown or pharmacologic inhibition with the small molecule STM2457 leads to reduced neuroblastoma cell proliferation and increased differentiation. These changes in neuroblastoma phenotype are associated with decreased m6A deposition on transcripts involved in nervous system development and neuronal differentiation, with increased stability of target mRNAs. In preclinical studies, STM2457 treatment suppresses the growth of neuroblastoma tumors in vivo. Together, these results support the potential of METTL3/METTL14 complex inhibition as a therapeutic strategy against neuroblastoma.

Abstract BACKGROUND Diffuse Midline Glioma (DMG) is a highly aggressive (CNS WHO Grade IV) pediatric brain tumor characterized by the histone mutation H3K27M, resulting in global hypomethylation of histones. H3K27M mutant cells are highly dependent on methionine and methionine Adenosyltransferase 2A (MAT2A), a central regulator of the methionine cycle, representing a key vulnerability in H3K27M gliomas. MAT2A is upregulated by methyltransferase-like protein 16 (METTL16), which deposits N6-methyladenosine (m6A) on a subset of RNA residues to regulate gene expression. MAT2A is a sensor of methyl group donor S-adenosylmethionine (SAM) that, when placed, increases MAT2A intron splicing to compensate for a decrease in MAT2A function. METHODS To investigate our hypothesis, H3K27M histone mutation knockout cell lines were established using CRISPR Cas9 (Jabado lab) in DIPG4 cells. shRNA-mediated knockdown was used to attenuate METTL16 expression in DIPG4 H3K27M cells, and growth was assessed via sulforhodamine B (SRB) assay. RESULTS In silico analysis of normal brain tissue compared to low-grade glioma and high-grade glioma demonstrated increased METTL16 at both the RNA and protein levels (p=<0.0001). Therefore, we postulated that targeting METTL16 would affect H3K27M growth. Results showed increased cell death compared to control cells by day 4 (p=0.0034) and day 8 (p=<0.0001). Furthermore, knockdown of METTL16 in DMG cells with H3K27M decreased m6A levels on poly (A) RNA (p=0.049). CONCLUSIONS Our study highlights the vulnerability of H3K27M mutant cells to METTL16 modulation, elucidating underlying molecular mechanisms and paving the way for innovative therapeutic strategies targeting this axis.

Cyclin-dependent kinase 12 (CDK12) modulates transcription elongation by phosphorylating the carboxy-terminal domain of RNA polymerase II and selectively affects the expression of genes involved in the DNA damage response (DDR) and mRNA processing. Yet, the mechanisms underlying such selectivity remain unclear. Here we show that CDK12 inhibition in cancer cells lacking CDK12 mutations results in gene length-dependent elongation defects, inducing premature cleavage and polyadenylation (PCPA) and loss of expression of long (>45 kb) genes, a substantial proportion of which participate in the DDR. This early termination phenotype correlates with an increased number of intronic polyadenylation sites, a feature especially prominent among DDR genes. Phosphoproteomic analysis indicated that CDK12 directly phosphorylates pre-mRNA processing factors, including those regulating PCPA. These results support a model in which DDR genes are uniquely susceptible to CDK12 inhibition primarily due to their relatively longer lengths and lower ratios of U1 snRNP binding to intronic polyadenylation sites.