BRAIN CANCER

Medulloblastomas are primary brain tumors that occur in the cerebellum of children and young adults. The name medulloblastoma was given by Bailey and Cushing in 1925; they suggested that these tumors arise from a hypothesized CNS precursor cell called a medulloblast [1]. As this cell type has not been identified, these tumors have subsequently (although somewhat controversially) been placed in a group of histologically similar CNS tumors, called primitive neuro-ectodermal tumors (PNET) [2]. Because these tumors arise in the cerebellum and frequently obstruct the 4th ventricle, patients present with symptoms of cerebellar dysfunction. Medulloblastomas are usually detected by magnetic resonance imaging and usually arise sporadically, although rarely occur as part of an inherited cancer syndrome such as Turcot or Gorlin syndrome. Although these tumors respond to therapy, in many cases there are recurrences with fatal results. The biology of the cells that survive therapy and give rise to tumor recurrence is not fully understood. We need a better understanding of the cells that are resistant to therapy and that repopulate the tumor post-therapy. In order to investigate the details of the radiation biology of medulloblastomas, we need a mouse model that recapitulates the biology of the tumors that can be used to specifically address the biology of therapeutic response in the cell populations that comprise these lesions.
Medulloblastomas in humans have been divided into 5 groups by gene expression analysis. One of these groups shows activation of the SHH signaling pathway and most available mouse models of medulloblastomas reflect this medulloblastoma subtype. We are able to generate medulloblastoma tumors by several methods including SHH gene transfer to nestin expressing cells in the developing cerebellum using the RCAS/tv-a system of somatic cell gene transfer developed by the Holland lab. These tumors are composed of three cell types, proliferating tumor bulk cells, post-mitotic neuroblastic cells, and perivascular cells with stem like character. The PI3K pathway affects these cell types differently: the tumor bulk has relatively low PI3K signaling, the neuroblastic cells are driven to their neuronal state by Akt activity specifically, and the perivascular cells proliferate in response to mTOR/pS6 activity. Radiation kills the tumor bulk while the neuroblastic cells with high Akt activity are spared, and the perivascular cells show a further induction of the PI3K pathway (PAkt and pS6) and arrest rather than undergo apoptosis. Concomitant with this arrest and activation of the PI3K pathway these perivascular cells increase the expression of the stem cell marker nestin.

Gliomas are classified into four grades with glioblastoma multiforme (GBM; WHO grade IV) being the most aggressive type. GBMs develop de novo (primary glioblastomas) or through progressive malignification of lower grade gliomas (secondary glioblastomas). Diffuse gliomas (WHO grade II) are categorized into astrocytomas, oligodendrogliomas, and mixed oligoastrocytomas based on the cells’ resemblance to astrocytes or oligodendrocytes. WHO grade III gliomas show intermediate aggressiveness and include anaplastic astrocytomas and anaplastic oligodendrogliomas. GBMs have several other findings that define the grade 4 character of this tumor subtype. First, the vasculature both within and surrounding these lesions is hyperproliferative referred to as “gliomeruloid”, a reference to the similarity to the renal glomerulus. Second, the nuclei show regional variability ranging from areas of cells with very small nuclei to others with giant and irregular nuclei. Finally, the most characteristic trait of the GBM subgroup is the presence of pseudopalisading necrosis -- regions of cell death surrounded by a dense wreath of tumor cells. The gross appearance of GBMs is that of heterogeneous masses with regions of hemorrhage and necrosis. Microscopically, these tumors show microvascular proliferation, pseudopalisading necrosis and nuclear pleomorphism as indicated [3,4]. Gliomas in humans are divided using proteomic and genomic analysis into three major subgroups including those with NF1 loss, those with EGFR activation and a third subset driven by signaling through the PDGF receptor. It is this third subset that is efficiently modeled in the mouse by the Holland lab using RCAS/tv-a by transferring expression of PDGF ligand to nestin expressing cells in either newborn or adult mice. These tumors contain more cell types than medulloblastomas. There are the endothelial cells, around them in a perivascular niche is a collection of GFAP expressing and recruited astrocytes, nestin-expressing stem like cells, SMA expressing pericytes, and macrophages. The tumor bulk contains Olig2 expressing cells intermixed with regions of necrosis ringed by cells expressing very undifferentiated markers. The perivascular niche also contains many cells expressing these classic stem cell markers as well. The PI3K pathway is substantially activated in the perivascular niche and to a lesser degree in the tumor bulk (but much more than in medulloblastomas). Upon radiation, these tumors show little cell killing, even at 10 Gy. The PI3K pathway is activated by radiation showing substantial elevation of pAkt and pS6 as well as the induction of nestin expression throughout the tumor.

The problems with these two tumor types are related but different. In the case of medulloblastomas, the tumors are relatively responsive to therapy. However, 30% of these children’s tumors relapse following therapy with fatal results. Furthermore, the children treated for this tumor are frequently significantly injured by the radiation and chemotherapy. A complete understanding of the biology of these tumors and why they recur following therapy could help improve responses with a reduction in toxicity. In the case of gliomas, the tumors are far more resistant to therapy and the clarity of the binary response of apoptosis vs. cell cycle arrest seen with medulloblastomas is not present. Glioma cells acquire greater degrees of stem cell character as the tumors progress and they acquire mutations. Understanding the role of stem cell character and the cellular origins of these tumors will be critical for our designing therapies targeting these cells preferentially over the adult brains normal CNS stem cells. Our PS-OC aims to address these issues.

LITERATURE CITED

1. Bailey PC, H (1926) A classification of the tumors of the glioma group on a histogenetic basis with a correlated study of prognosis. Philadelphia: JB. Lippincott Co.
2. Rorke LB (1983) The cerebellar medulloblastoma and its relationship to primitive neuroectodermal tumors. J Neuropathol Exp Neurol 42: 1-15.
3. Kleihues P, Louis DN, Scheithauer BW, Rorke LB, Reifenberger G, et al. (2002) The WHO classification of tumors of the nervous system. J Neuropathol Exp Neurol 61: 215-225; discussion 226-219.
4. Louis DN, Ohgaki H, Wiestler OD, Cavenee WK, Burger PC, et al. (2007) The 2007 WHO classification of tumours of the central nervous system. Acta Neuropathol 114: 97-109.