The goal of the Reitman Lab is to make discoveries that guide the design of improved treatment strategies for children and adults with brain tumors. We leverage molecular biology techniques, genetically engineered mouse models and cancer genomic approaches to carry out our work. The Reitman Lab is based in the Department of Radiation Oncology and the team works closely with the faculty and staff in the Preston Robert Tisch Brain Tumor Center at Duke, the Department of Neurosurgery and the Duke Cancer Institute. We are interested in several broad research themes:
Enhancing the efficacy of radiation therapy
Radiation therapy plays a critical role in the treatment of many brain tumor patients. For many patients with brain tumors, radiation therapy is part of a curative treatment regimen but can be associated with long-term toxicity. For patients with other types of brain tumors, toxicity to normal tissues limits the ability to deliver a curative dose of radiation therapy to the brain tumor. The Reitman Lab is investigating new approaches to widen the therapeutic ratio of radiation therapy by making it more effective in killing brain tumor cells and helping to reduce toxicity to normal tissues. For example, one approach involves studying the DNA damage response molecular pathway that regulates the cellular response to radiation. Genes involved in the DNA damage response pathway are mutated in many brain tumors, especially in gliomas of the brainstem (Figure 1). Since these tumors have deregulated DNA damage response machinery, treatments that target key nodes in the DNA damage response network might be particularly effective in sensitizing these tumors to radiation therapy. We are using primary genetically engineered mouse models of diffuse midline gliomas to test if modulating key DNA damage response molecules can sensitize these tumors to radiation therapy.
Figure 1. Frequent DNA damage response pathway gene mutations in brainstem and thalamus gliomas.
Each column represents a patient with a brainstem or thalamus glioma. This includes predominantly diffuse midline gliomas with H3K27M mutations (H3F3A mutated) and diffuse intrinsic pontine gliomas (DIPGs). The top two rows delineate the location and histopathologic grade of the tumor. The bottom four rows indicate the presence of mutations in key cancer genes in each tumor specimen. The data show that the >50% of these brain tumors harbor inactivating mutations in the tumor suppressor TP53, which regulates apoptosis and cell cycle progression especially after DNA damage is caused by radiation therapy. Additional tumors that do not contain TP53 mutations instead contain mutations that activate PPM1D, which encodes a phosphatase that dephosphorylates and represses p53 function. Thus, the majority of these brain tumors contain mutations that deregulate the DNA damage response pathway by perturbing p53 function. We are testing if targeting the serine/threonine kinase ataxia telangiectasia mutated (ATM), which plays a critical role in the detection of DNA damage caused by radiation therapy, can specifically radiosensitize diffuse midline gliomas with TP53 or PPM1D mutations. This work is primarily being carried out in genetically engineered mouse models of diffuse midline gliomas that faithfully replicate the genetic mutations in these tumors.
Figure from: Zhang L… Reitman ZJ, Bigner DD, Yan H. Exome sequencing identifies somatic gain-of-function PPM1D mutations in brainstem gliomas. Nat Genet. 2014 PMID: 24880341
New approaches to target brain tumor mutations
Knowledge of the genomic landscape of brain tumors has exploded over the past decade. In many types of cancer outside the brain, this type of genomic knowledge has resulted in new therapies that improve patient survival by targeting mutations found in the genome of the cancer cells. However, advances in genomic characterization of brain tumors have not yet produced new survival-improving therapies for many types of brain tumors. To overcome this challenge, the Reitman Lab is investigating creative new approaches to target frequent mutations found in brain tumors. For instance, we are interested in finding novel actionable vulnerabilities associated with mutations in the promoter of the telomerase reverse transcriptase gene (TERT). TERT promoter mutations are found in more than 80% of glioblastomas, which are the most frequent and lethal primary brain tumor in adults, and the mutations confer a replicative immortality phenotype to brain tumor cells (Figure 2).
Figure 2. Identifying new approaches to target TERT promoter mutations in cancer.
The two most frequent TERT promoter mutations found in glioblastoma are shown. The C228T mutation occurs 146 bp upstream of the start codon of TERT. The C250T mutation occurs 126 bp upstream of the start codon of TERT. Either mutation generates a de novo sequence which contains an ETS transcription factor binding motif. The new ETS binding motif aberrantly recruits transcription factors from the ETS family to the TERT promoter. This aberrantly activates TERT expression to enable replicative immortality in glioblastoma cells. Our long-term goal is to determine if targeting molecular pathways upstream of these transcription factors could be used as a cancer therapeutic approach for tumors with TERT promoter mutations. To identify such vulnerabilities, we are using genome-wide CRISPR-based screening approaches in patient-derived glioblastoma cell lines faithfully harboring TERT promoter mutations.
Figure from: Reitman ZJ, et al. Promoting a new brain tumor mutation: TERT promoter mutations in CNS tumors. Acta Neuropathol. 2013 PMID: 24217890
Examining tumor heterogeneity and treatment resistance
Another hurdle to the success of new brain tumor treatments is that the tissue of many brain tumors is heterogeneous and does not uniformly respond to treatment. Furthermore, brain tumors evolve when the patient undergoes treatment, ultimately leading to treatment resistance. We are leveraging emerging technologies such as single cell RNA-sequencing to profile brain tumors at single cell resolution to define tumor heterogeneity and treatment resistance mechanisms in brain tumors. We are particularly interested in exploring pediatric brain tumors, low grade gliomas, glioneuronal tumors and tumors with mutations in the BRAF oncogene in this manner (Figure 3).
Figure 3. Single cell genomics to reveal brain tumor developmental hierarchies.
Model for glioma differentiation hierarchies based on single cell RNA-sequencing analyses of primary human brain tumors. The schematic shows differences in abundance of cycling cells and cellular differentiation state for various subtypes of gliomas. The y-axis shows differentiation state of the tumor cells, ranging from undifferentiated neuronal progenitor-like cells at the top to more differentiated mature glia-like cells at the bottom. The x-axis shows three major glioma subytpes, including IDH-mutated oligodendrogliomas (IDH-O) and astrocytomas (IDH-A), diffuse midline gliomas with H3K27M mutation and pilocytic astrocytomas with alterations in the BRAF oncogene (BRAF-PA). The BRAF-PAs resemble a more differentiated lineage hierarchy compared to the IDH- and H3K27M-mutated tumors. These findings may underlie the differing clinical behavior and varying responses to treatment of these brain tumor subtypes.
Figure from: Reitman ZJ, Paolella BR, et al., Mitogenic and progenitor gene programmes in single pilocytic astrocytoma cells. Nat Commun. 2019 PMID: 31427603
We are also interested in innovatively harnessing gain-of-function cancer mutations to inform the design of valuable new enzymes, which could potentially be useful for drug and chemical production processes (Figure 4).
Figure 4. Enzyme redesign guided by cancer-derived mutations.
Cover artwork from Nature Chemical Biology showing (on the left) oligodendroglioma tumor cells, which contain an isocitrate dehydrogenase 1 (IDH1) R132H mutation. The active site of IDH1 is shown on the right. The R132H mutation affects a critical active site residue of IDH1, conferring a change-of-function to the IDH1 enzyme. We applied analogous mutations to the active sites of distantly related enzymes to redesign those enzymes into novel, useful enzymes with new functions. Future work seeks to identify additional cancer-derived mutations in enzyme active sites that could be used to guide the design of useful enzymes in a similar manner.
Figure from: Reitman ZJ, et al. Enzyme redesign guided by cancer-derived IDH1 mutations. Nat Chem Biol. 2012 PMID: 23001033
Additionally, we are interested in circulating tumor DNA (ctDNA) that is released from brain tumors into the circulation, especially as they undergo radiation therapy, to determine if testing for these byproducts could be used to improve the clinical management of brain tumor patients.
The Reitman Lab has benefited from support from the Developmental Research Program and the Career Enhancement Program within the Duke Brain Tumor SPORE grant from the National Cancer Institute. The Lab has also received support from research foundations including the Pediatric Brain Tumor Foundation, the St. Baldrick’s Foundation, the Michael Mosier Defeat DIPG Foundation, the ChadTough Foundation, the SoSo Strong Foundation and the Conquer Cancer Foundation of the American Society for Clinical Oncology. Finally, support from the Duke Cancer Institute, the Preston Robert Tisch Brain Tumor Center at Duke and the Departments of Radiation Oncology and Neurosurgery are critical to carry out our work. We are extremely grateful to these sponsors and for the advocacy of patients and their families who make this research possible.