Recent genome-wide analyses have changed the diagnosis of medulloblastoma (MB), the most common malignant brain tumor in children. It is now clear that MB is not a single disease of the cerebellum (CB), but encompasses a range of disease subgroups with diverse clinical presentations, histology, and pathway activation, cells of origin and cancer genetics. Given the complexity of MB, development of targeted therapies tailored to each patient will require deeper insight into the major pathways fueling the disease and further subdivision of the current 4 subgroups of patients. The sonic hedgehog (SHH) subgroup represents ~30% of MBs, has intermediate prognosis, and can arise from granule cell precursors (GCPs) that proliferate in response to SHH. The SHH subgroup of MB is diverse at all levels, and each factor is likely to influence prognosis and be critical for treatment. Loss-of function mutations are seen in genes that inhibit SHH signaling (PTCH1), activating mutations in the receptor SMO (e.g. SmoM2) and amplifications of the effector gene GLI2 and its targets associated with loss of TP53. Current HH inhibitors are only effective in a small percentage of MBs, and all patients develop resistance. We hypothesize that some of the intertumoral heterogeneity within SHH-MBs is due to their arising from different cell lineages, at different times, and in different anatomic regions of the CB. Additionally, that cellular and transcriptional states inherited from distinct cells of origin are maintained in the mature tumor, and could represent targets for therapy. Basic research approaches in mouse models are thus necessary as a foundation for translational studies. We have studied the SHH-GLI pathway in mammalian development for 20 years, with a focus on the CB. Recently we developed sporadic MB models with mutations in Ptch1 or Smo. Significantly, reproducible yet distinct histologies are seen during tumor progression in each model. Moreover, the majorities of lesions seen at early stages regresses and have different cellular phenotypes to lesions that progress to MB. We propose to:
Aim 1. Marker analysis and longitudinal Mn-Enhanced MRI (Turnbull) of Ptch1 and SmoM2 models initiated at 2 developmental stages and lineages, and identify candidate intrinsic genes that influence progression of SHH-MBs (RNA-seq profiling).
Aim 2. Define the microenvironment and test how tumor associated microglia and macrophages influence SHH-MB progression.
Aim 3. Determine whether select candidate genes identified in Aims 1/2, mTOR signaling or NR2F2 alter SHH- MB progression and compare the results to human SHH-MB data sets (Taylor) and tumor samples. Our experimental studies of MB progression in mouse models will synergize with human MB studies and should aid in stratification of SHH subgroup patients, wherein new therapies targeted to essential pathways driving MB progression can be tested, improving both prognosis and quality of life for survivors.
Medulloblastoma is the most common malignant pediatric brain tumor, and while long-term survival of patients is greater than 65 percent, most survivors suffer from major cognitive deficits and other neurological problems as a result of an aggressive chemotherapy and radiation therapy protocol that is not tailored to the emerging clear heterogeneity of medulloblastomas. In order to optimize targeted therapies, it is critical to furthr subdivide the existing four subgroups of medulloblastoma based on the biological effects of tumors arising from different cell types, as well as the timing, location and kinds of gene mutations that occur in each tumor. We propose to identify candidate pathways that drive regression or promote medulloblastoma formation in multiple mouse models, test their effect on tumor progression/regression in mouse models, and apply our results to human medulloblastoma data sets and tissue samples in order to subdivide the SHH subgroup of patients and provide a basis for testing new targeted therapies to improve quality of life.
| Tan, I-Li; Wojcinski, Alexandre; Rallapalli, Harikrishna et al. (2018) Lateral cerebellum is preferentially sensitive to high sonic hedgehog signaling and medulloblastoma formation. Proc Natl Acad Sci U S A 115:3392-3397 |
