Search

MYC-driven medulloblastomas are a particularly devastating group of pediatric brain tumors that exhibit resistance and continued progression despite standard of care treatments. Our preclinical work identified BET-bromodomain inhibitors as a potentially promising new class of drugs for children with medulloblastoma and other MYC-driven cancers, providing rationale to evaluate these agents in clinical trials. However, treatment with BET inhibitor (BETi) alone is unlikely to be sufficient for a cure, with most tumors evolving to acquire resistance to single-agent targeted therapies. We applied an integrative genomics approach to identify genes and pathways mediating BETi response in medulloblastoma. These studies revealed that MYC-driven medulloblastoma cells with acquired resistance to BETi reinstate transcription of essential genes suppressed by drug, and exhibit changes in cell state and new vulnerabilities not present in drug-sensitive cells. We now have a growing body of evidence showing that BET inhibition downregulates the expression of key lipid metabolism genes and metabolism-related signaling pathways, and that medulloblastoma cells with adaptive resistance to drug differentially express and exhibit preferential dependency on specific lipid metabolic genes and transcriptional regulators. Our studies explore the possibility of exploiting these novel metabolic vulnerabilities in order to overcome BETi resistance and provide a more efficacious upfront therapy.

MYC-driven medulloblastomas are a particularly devastating group of pediatric brain tumors that exhibit resistance and continued progression despite standard of care treatments. Our preclinical work identified BET-bromodomain inhibitors as a potentially promising new class of drugs for children with medulloblastoma and other MYC-driven cancers, providing rationale to evaluate these agents in clinical trials. However, treatment with BET inhibitor (BETi) alone is unlikely to be sufficient to cure, with most tumors evolving to acquire resistance to single-agent targeted therapies. We applied an integrative genomics approach to identify genes and pathways mediating BETi response in medulloblastoma. These studies revealed that MYC-driven medulloblastoma cells with acquired resistance to BETi reinstate transcription of essential genes suppressed by drug and exhibit changes in cell state and new vulnerabilities not present in drug-sensitive cells. We now have a growing body of evidence showing that BET inhibition downregulates the expression of key lipid metabolism genes and metabolism-related signaling pathways, and that medulloblastoma cells with adaptive resistance to drug differentially express and exhibit preferential dependency on specific lipid metabolic genes and transcriptional regulators. Our studies explore the possibility of exploiting these metabolic vulnerabilities to overcome BETi resistance and provide a more efficacious upfront therapy.

BET-bromodomain inhibition (BETi) is a promising therapeutic strategy to target MYC-driven cancers, including Group 3 medulloblastoma, a deadly childhood brain tumor. We have shown that BET inhibitors exhibit preclinical efficacy against MYC¬-amplified medulloblastoma, providing motivation to evaluate this drug class in early phase clinical trials. However, our work has also found that MYC-amplified medulloblastoma cells can acquire resistance to BETi, suggesting that curative responses for this disease will require combination therapy. To guide the development of such combination therapies, we have focused our efforts on elucidating the mechanisms through which medulloblastoma cells acquire resistance to BETi. We found that medulloblastoma cells can develop tolerance to BETi by reinstating the expression of cell-essential “rescue genes,” which include bHLH transcription factors, cell-cycle regulators, and anti-apoptosis genes. This transition to the resistant cell state is mediated through changes in chromatin structure including the upregulation of H3K4me3 promoters. Our preliminary results suggest that BETi-resistant cells maintain mRNA transcription and protein translation of important mediators of resistance. Importantly, we observe that BETi-resistant medulloblastoma cells are more dependent on specific protein complexes involved in transcriptional regulation. This project explores the mechanisms through which these transcriptional regulators help maintain transcription of rescue genes that drive BETi resistance and evaluates the potential of targeting these drivers of BETi resistance. These results will help guide the development of combination approaches to improve the efficacy of BETi for the treatment of MYC-driven medulloblastoma.

TP53 is the most frequently mutated tumor suppressor with somatic alterations found in approximately 50% of all human cancers. In the remaining TP53 wild-type (WT) tumors, functional inactivation of the p53 pathway may be achieved through a variety of other mechanisms, including gene deletion, epigenetic silencing, and/or alterations in prominent negative regulators, including MDM2/MDM4 and PPM1D. These alterations block p53 activity and lead to uncontrolled cell proliferation and oncogenesis in a majority of cancers, including the highly aggressive, universally fatal glial cell tumors of childhood, known as Diffuse Intrinsic Pontine Gliomas (DIPGs). DIPGs are inoperable due to their location within the brainstem. Available treatment options, including radiotherapy, have had a palliative effect at best, with almost all children succumbing to the disease within 18 months of diagnosis. Recent advances have led to an improved understanding of the biological underpinnings of this disease and identification of recurrent genetic alterations that represent potential therapeutic targets for these patients. Prominent among these targets in DIPGs with WT p53 status (50%) are MDM2/4 and PPM1D, whose suppression lead to p53 reactivation specifically in the WT p53 context. We have undertaken a combination of approaches to better understand the therapeutic potential of MDM2 and PPM1D inhibition in DIPG, characterizing the genomic, transcriptomic, and cell-state changes that drive resistance, and identifying novel vulnerabilities that can be exploited with combination therapies towards a cure.

A majority of high grade gliomas retain a wild-type TP53 gene and are amenable to strategies for activation of the pathway to inhibit tumor growth. The interaction between p53 and MDM2 has served as target for such strategies currently in clinical trials for glioblastoma. As the effects and resistance mechanisms of MDM2 inhibition (MDM2i) remain poorly understood in glioma, we performed genomic and transcriptomic analyses in patient-derived models to better characterize sensitive tumors and identify putative biomarkers of drug response. Treatment with an MDM2 inhibitor (KRT232/AMG232) impaired the growth of cell lines with wild-type TP53 status, particularly in tumors with amplification of MDM2/4 or PPM1D activating mutations. Treatment with KRT232 upregulated both cell cycle arrest and apoptotic cellular responses, with unique temporal and transcriptional differences correlated with MDM2/4 or PPM1D status. In other tumors resistance to MDM2i is mainly mediated by TP53 mutations, but in a subset of chronic KRT232-treated glioma models we noted lack of TP53 mutations and identified cell state and transcriptional changes as potentially more treatable mediators of resistance.

Over 80% of high-grade gliomas have alterations in members of the p53 pathway, a central regulator of cell cycle progression and apoptosis that becomes activated in response to cellular stress and DNA damage. For tumors that retain wild-type p53, pathway deregulation frequently occurs through the amplification of negative regulators of p53, including the E3 ubiquitin ligase MDM2. The p53/MDM2 interaction axis has served as basis for the development of several classes of MDM2 inhibitors, with AMG232 being the most potent molecule currently undergoing clinical evaluation. As the effects of MDM2 inhibition (MDM2i) remain poorly understood in high-grade glioma, we performed genomic and transcriptomic analyses in patient-derived models to better characterize sensitive tumors and identify putative biomarkers of drug response. Treatment with AMG232 impaired the growth of cell lines with wild-type p53 status, particularly in tumors with additional amplification of MDM4 or PPM1D activating mutations. Treatment with AMG232 upregulated both cell cycle arrest and apoptotic cellular responses, as measured by annexin V/PI staining and immunoblotting. Interestingly, the dynamics of these two downstream p53 signaling axis were dependent on treatment duration across models. In addition to p53 pathway activation and apoptotic induction, RNA-sequencing revealed MDM2i to be associated with the activation of oncogenic MAPK and KRAS signaling as well as epithelial to mesenchymal transition markers. In most solid tumors, resistance to MDM2i is mainly mediated by acquisition of p53 inactivating mutations. We hypothesized that resistance mechanisms in glioma may be partially driven by transcriptional changes, as these tumors consist of subpopulations with diverse cell differentiation states. By chronic AMG232 treatment, we have developed in vitro and in vivo models of acquired MDM2i resistance that are not mediated by p53 inactivation. Ongoing work is focused on characterizing the transcriptional profile of these cells to identify transcriptional changes leading to decreased drug response.