Medulloblastoma and the DNA Damage Response

Abstract

Medulloblastoma (MB) is the most common malignant brain tumor in children with standard of care consisting of surgery, radiation, and chemotherapy. Recent molecular profiling led to the identification of four molecularly distinct MB subgroups - Wingless (WNT), Sonic Hedgehog (SHH), Group 3, and Group 4. Despite genomic MB characterization and subsequent tumor stratification, clinical treatment paradigms are still largely driven by histology, degree of surgical resection, and presence or absence of metastasis rather than molecular profile. Patients usually undergo resection of their tumor followed by craniospinal radiation (CSI) and a 6 month to one-year multi-agent chemotherapeutic regimen. While there is clearly a need for development of targeted agents specific to the molecular alterations of each patient, targeting proteins responsible for DNA damage repair could have a broader impact regardless of molecular subgrouping. DNA damage response (DDR) protein inhibitors have recently emerged as targeted agents with potent activity as monotherapy or in combination in different cancers. Here we discuss the molecular underpinnings of genomic instability in MB and potential avenues for exploitation through DNA damage response inhibition.

Keywords: medulloblastoma; p53 status; pediatrics; radiation oncology; therapeutic targeting.

Figure 1

Medulloblastoma is a tumor of the cerebellum. Medulloblastoma constitutes the most common childhood malignant brain tumor, accounting for approximately 20% of all central nervous system tumors. Patient stratification consists of categorization based on age and presence of metastasis, histology, and genotyping. Following initial staging*, maximal surgical resection is typically attempted for all patients. Subsequent histology and molecular profiling introduces subgroup classification and potential alterations to the treatment protocol. While the vast majority of patients receive some combination of radiation and chemotherapy, the age, location of tumor (2), extent of spread, histology, and subgroup inform the intensity and duration of treatment. MRI, Magnetic Resonance Imaging; STR, Subtotal Resection; GTR, Gross Total Resection; LCA, Large Cell Anaplastic; DNV, Desmoplastic/Nodular; MBEN, medulloblastoma with extensive nodularity; LP, Lumbar Puncture *may be delayed to post-surgery.

Figure 2

A summary of the DNA damage response in medulloblastoma. Numerous DNA damage response signaling axes mediate DNA repair following medulloblastoma standard of care. Chemotherapies utilized, including temozolomide, cisplatin, lomustine, and cyclophosphamide (with cisplatin and cyclophosphamide constituting standard of care), bind directly to DNA to create bulky lesions repaired through a combination of either Base Excision Repair (BER), Nucleotide Excision Repair (NER), MGMT (Not shown), and HR or NHEJ, while IR mediated DSBs are repaired through NHEJ and HR. (Top) The deleterious effects of Ionizing radiation results primarily from DNA double strand breaks (DSBs) repaired either through Homologous Repair (HR) or Non-Homologous End Joining (NHEJ). While PARP, M/R/N (a complex of 3 proteins: MRE-11, RAD50, NBS), and ATM function universally to recognize strand breaks, cells in G0/G1 lack sister chromatids and will repair through a more error prone NHEJ, while those in S, Interphase, or G2 will repair through HR, utilizing the sister chromatid to replace the missing bases. Proteins directly involved in repair of double strand breaks include Ku70/80 and DNA-PKcs for NHEJ and M/R/N, BRCA1/2, and RAD50 for HR. Upstream effectors, ATM and ATR, will signal through Chk1 and 2 to arrest the cell and either repair or commit apoptosis ( Figure 4 ). (Middle) Single strand breaks resulting from radiation or excision of damaged base pairs by APE1 (such as those resulting from alkylation) are identified by PARP1 and repaired by XRCC1, DNA Ligase III (LIG3), and DNA Polymerase β. (Bottom) Bulky lesions resulting from platinum-based drugs such as cisplatin or carboplatin can be repaired either through transcription coupled NER (TC-NER) or global genomic NER (GG-NER). While it remains unclear what role PARP may play in TC-NER, it serves as a recognition and recruitment protein for GG-NER, functioning alongside DNA Ligase I or III (LIG1/3), RPA, and ERCC1 proteins. Bifunctional alkylating agents and chloro-ethylating agents, such as cyclophosphamide or lomustine, respectively, can methylate guanine. Methylated guanine can be repaired through MGMT (not shown), through direct removal of guanine methyl group, potentially resulting intrastrand cross-linking (ICL) requiring a combination of NER and HR or NHEJ to repair) or interstrand cross-linking, repaired through NER. Platinum drugs mediate interstrand cross-linking, repaired predominantly through NER. Finally, temozolomide can be repaired either through BER (requiring APE1) or MGMT (24).

Figure 3

The role of AKT in DNA damage. (A) Canonical AKT signaling begins with the activation of a receptor tyrosine kinase (e.g., VEGFR, EGFR, or IGFR). Receptor activation, by binding of a ligand, dimerization, and autophosphorylation, causes binding and activation of p85 and p110, the respective regulatory and catalytic subunits of PI3K. Activated PI3K will go on to catalyze the conversion of PIP2 to PIP3, a process which can be reversed by the tumor suppressor PTEN. If PIP3 is formed, it will recruit AKT to the membrane at which point AKT will be phosphorylated by PDK1 and mTORC2. This dual phosphorylation fully activates AKT, allowing it to affect several downstream processes such as cell survival, proliferation, growth, and the DNA damage response. (B) Following exposure to radiation, repair proteins (e.g. ATM, ATR, DNA-PKs) are recruited to the damaged site to facilitate repair. The activation and recruitment of these proteins correlates with an increase in phosphorylated AKT. p-AKT then inhibits apoptosis through Bad and Bax inhibition and Bcl-2 activation. p-AKT will also inhibit the downstream target of p53, p21, to inhibit cell cycle arrest. A feedback loop exists in that AKT can also phosphorylate and activate these proteins to aid in DNA repair. AKT facilitates NHEJ by phosphorylating DNA-PKs and inhibits HR by promoting cytoplasmic localization of BRCA1 and Rad51. (C) Irradiation decreases the population of tumor cells at the perivascular niche with the exception of stem cells which are largely radioresistant. Nestin-positive stem cells in the perivascular niche treated with radiation activate the PI3K/AKT pathway to undergo cell cycle arrest and re-enter the cycle 72 hours later, driving MB survival and recurrence. AKT inhibition prior to irradiation abrogates perivascular stem cell radiation resistance, leading to apoptosis of these cells.

Figure 4

Impact of p53 on survival and p53 signaling in the DNA damage response. P53 is a downstream effector following double strand DNA breaks and plays a crucial role in tumorigenesis. Following Damage detection by ATM/ATR or activation of DNA-PKcs, p53 is phosphorylated at Ser15 or Ser37. ATM and ATR will also activate Chk1 and 2 resulting in phosphorylation of p53 at Ser20. Together these phospho-sites inhibit p53 degradation by preventing MDM2 binding and facilitate p53 tetramerization, a crucial step in p53 activation. p53 will activate p21 and inhibit CDC25 A,B, and C to activate the cell cycle checkpoint and inhibit cycling. p53 will also transcriptionally upregulate pro-death proteins, BAX, PUMA, and NOXA. Upon overcoming inhibition by BCL-2 family apoptotic inhibitors (not shown), BAX, PUMA, and NOXA will activate apoptosis. Inhibition of the MDM2-p53 interaction with nutlin allows for p53 accumulation and apoptosis, ideally in cells heavily reliant on p53 suppression for survival.