Understanding the molecular basis of malignant pediatric brain tumors and developing preclinical pipelines to evaluate therapeutic approaches.
Our lab studies some of the most aggressive, and difficult to treat, forms of pediatric brain tumors. We have developed a research pipeline that integrates molecular characterization, model development, and preclinical and clinical strategy. Our long-term goal is to find new, and improve existing, therapies for patients with malignant brain tumors.
Our laboratory uses an integrated and multidisciplinary state of the art approaches including extensive molecular and immunologic characterization of malignant pediatric brain tumors to understand the etiology of disease and identify drivers of tumor development. This, in turn, has allowed us to develop several accurate mouse models of medulloblastoma (MB) that recapitulate human disease and facilitate translational experiments. We have also generated a large cohort of orthotopic xenografts of primary tumor samples from patients treated at St. Jude that are currently used to validate the efficacy of known drugs and targeted therapies in pre-clinical trials. Our work continues to evolve as we partner with internal and external collaborators through interdisciplinary research consortia.
In order to understand the development and resilience of malignant pediatric brain tumors, we need access to model systems that recapitulate human disease. Built on our understanding of cell cycle regulation and MYC family signaling pathways, we have developed several models of MB representing the most aggressive forms of disease. Subsequently, we generated a mouse model with endogenous, CRISPR-driven, overexpression of MYC that has facilitated additional questions around transcriptional-targeting therapies and epigenetic regulation. Our next step is to generate these models in human-based cell systems for even greater translational relevance. Critically, my laboratory has generated a cohort of molecularly conserved and fully characterized patient-derived orthotopic xenograft (PDOX) models of primary tumor samples acquired from St. Jude patients. These models are essential to understand the heterogeneity of pediatric brain tumors and to develop effective preclinical strategies to inform clinical trial design.
Transcription factors have been considered “undrugable” but recent development of molecules including PROTACs and Molecular Glues provide a unique opportunity to directly degrade these proteins that do not have pockets targetable by drugs. In collaboration with experts in chemical biology, structure biology and molecular biology, we are targeting MYCN, GFI1 and GFI1B, transcription factors; all factors required for medulloblastoma development. Such degraders will be validated in mouse and human medulloblastoma models using our pre-clinical pipeline.
In addition to amplification of oncogenes and mutation of tumor suppressors, genetic alterations in epigenetic regulators account for the majority of genetic perturbations in Group 3 and Group 4 MB. An altered spectrum of epigenetic regulators and chromatin modifiers, resulting in aberrant patterns of histone lysine methylation, further distinguishes subgroups of MB. Using unbiased, genetic-based screening approaches, we have identified putative epigenetic regulators of tumorigenesis in MB and are currently working to define their molecular mechanisms and functional roles. Our laboratory collaborates with experts across St. Jude, leveraging informatics analysis, structural modeling, and functional genomics to uncover drivers and vulnerabilities in MB.
Ultimately, our laboratory is interested in developing new approaches and optimizing current strategies to more effectively treat pediatric brain tumor patients. We have established collaborative research teams with clinicians, pharmacokineticists, chemical biologists and statistical experts to create an integrated pipeline from molecular analysis to clinical trial design and launch. In previous work, high throughput screening of a library of FDA-approved drugs identified several chemotherapeutic compounds that have been advanced into ongoing clinical trials for the treatment of MYC-driven MB. Utilizing a fully characterized cohort of PDOX models, targeted therapies and craniospinal radiation protocols are also being optimized and explored for MB and ATRT.
Dr. Martine F. Roussel is a pioneer in the field of molecular oncology and translation cancer research, having discovered some of the earliest recognized retroviral oncogenes and defined central elements in cell cycle regulation. Born and educated in France, with research experiences spanning the globe, Dr. Roussel carved an impressive path that has shattered glass ceilings and earned renowned honors. Elected to the American Academy of Arts and Sciences in 2011, National Academy of Science in 2019 and a Fellow of the AACR in 2021, Dr. Roussel continues her long-standing and illustrious career at St. Jude driving discovery, creating clinical impact, and inspiring the next generation of scientists. Dr. Roussel is a transformative force in the pediatric brain tumor community thanks not only to her robust research portfolio, but her commitment to mentorship and training.
Dr. Charles J. Sherr has made seminal contributions to the understanding of oncogenes and tumor suppressors, the mechanics of cell division cycle control, and how key cell cycle regulators are perturbed in cancer. His accomplishments have justly earned his election to the AAAS, to the National Academies of Science and Medicine, and as an Inaugural Fellow of AACR, among notable other awards and appointments. Dr. Sherr was the recipient of continuous, unfettered support from the Howard Hughes Medical Institute for 30 years until he transitioned to emeritus status in 2019. In partnership with longtime collaborator and spouse, Dr. Martine Roussel, Dr. Sherr transformed the basic and translational pediatric cancer research landscape of St. Jude.
Dynamic team of scientists and technicians with expertise in molecular biology, genetics, informatics, and preclinical modeling.