A Thesis submitted to The University of Arizona College of Medicine - Phoenix in partial fulfillment of the requirements for the Degree of Doctor of Medicine. / Background: Great strides have been made in childhood cancer treatment efficacy over the past two decades leading to improved survival rates, and now attention is being directed toward identifying and understanding complications that affect many of these patients as they reach adulthood. Obesity is a well‐recognized late effect that has many potential long‐term consequences some of which include cardiovascular disease, type II diabetes mellitus, dyslipidemia and even death. Materials/Methods: We conducted a retrospective chart review to determine the prevalence of obesity among survivors of pediatric brain tumors 5 years after the completion of therapy and compare this to the general pediatric population of the same age. We also sought to identify potential risk factors for the development of obesity among survivors of childhood brain tumors. Obesity was defined as a body mass index (BMI) greater than the 95th percentile for age and gender as defined by the most recent Center for Disease Control growth curves. Results: We identified 96 patients who met our inclusion criteria, however only 43 had follow‐up data at 5 years after the completion of therapy to be included in final analysis. Of 43 patients, 5 (11.63%) were obese 5 years after completion of therapy. The CDC sites general population obesity rates in three age groups: 2‐5 years (8.4% obesity rate), 6‐ 11 years (18% obesity rate), 12‐19 years (21% obesity rate). Using CDC guidelines, we found no significant difference between the obesity rate among the brain tumor survivor population for each age group and the general population, p‐values of 0.865, 0.865, and 0.249 respectively. Conclusion: Our small sample size was likely not adequate to find a significant difference between the two groups or identify risk factors associated with the development of obesity. Larger studies are needed to further examine the risk of obesity among pediatric brain tumor survivors and to identify risk factors associated with this late effect.

A Thesis submitted to The University of Arizona College of Medicine - Phoenix in partial fulfillment of the requirements for the Degree of Doctor of Medicine. / Significance: Contrast‐enhanced (CE) and Fluid attenuation inversion recovery (FLAIR) MRI are current standard of care tools for delineating radiation treatment targets in high‐grade glioma (HGG) patients. However, in the setting of retreatment, tumor regrowth and non‐tumor therapy‐related inflammation, known as post‐treatment radiation effect (PTRE), have identical MRI appearances. As a result, FLAIR MRI can be an unreliable tool for treatment planning. Surgical biopsy can definitively distinguish recurrent tumors from PTRE but has many disadvantages, namely operative risk and cost. Dynamic Susceptibility‐weighted Contrast‐ Enhanced (DSC) MRI Perfusion can non‐invasively detect distinct characteristics of tumor and PTRE through measurements of relative cerebral blood volume (rCBV). PTRE exhibits decreased microvascular density, whereas tumor recurrence displays angiogenesis and microvascular proliferation. Thus, DSC‐MRI affords the opportunity to better define tumor burden within and possibly outside of these nonspecific regions. Objective: To assess the extent with which rCBV maps correlate with re‐radiation treatment plans in patients with recurrent tumor in order to identify potential differences in treatment planning. Design: This study enrolled 8 previously treated HGG patients presenting for re‐irradiation of suspected recurrent tumor at a single hospital on an IRB‐approved trial. All patients underwent DSC‐MRI and routine MRI imaging prior to re‐irradiation treatment planning, and underwent treatment as per routine clinical protocol. Following therapy, rCBV and radiation dose maps were overlaid on conventional MR to delineate differences in identified tumor burden. Results: Of the 8 patients, four rCBV images showed evidence of tumor outside of the RT planning volumes, while the other 4 showed fully treated tumor but with large volumes of uninvolved brain receiving radiation. Conclusion: DSC‐MRI better identified unique regions of potential tumor burden in recurrent HGG patients compared to conventional MRI and could be used to improve radiation treatment planning in re‐radiated patients.

Glioblastoma (GBM) is the most common and aggressive brain tumor in adults which is characterized by extensive cellular and genetic heterogeneity. Even with surgery, chemotherapy with temozolomide, and radiation, tumor re-growth and patient relapse are inevitable. The extensive inter- and intra-tumoral heterogeneity (ITH) of recurrent GBM emerges from dysregulation at multiple -omic levels of the tumor. ITH exits at the cellular level due to a small subpopulation of chemo- and radio-resistant cells, called brain tumor initiating cells, which may drive GBM treatment resistance. Although a wealth of literature describes the biology of primary GBM (pGBM), we currently lack an understanding of how GBM evolves through therapy to become a very different tumor at recurrence, which may explain why therapies against primary GBM fail to work in recurrent GBM (rGBM). Thus, understanding the tumor evolution from a multi-omic perspective is critical for the development of effective therapeutic approaches. The current work focuses on identification and validation of novel predictive and prognostic biomarkers for rGBM using proteomics analysis on a large cohort of patientmatched pGBM-rGBM samples. This work allowed for detailed characterization of rGBM and its cognate TIME toward a better understanding of the molecular players driving recurrence which can be further used for instructing effective targeted and personalized therapies for the treatment of therapy-resistant GBM. In another part, we developed a novel immunotherapeutic modality called dual antigen T cell engager, to target Carbonic Anhydrase 9, a highly enriched hypoxia-inducible enzyme in GBM. We demonstrated that this immunotherapeutic strategy which allows for targeting tumor cells while recruiting and triggering T cells through simultaneously, is highly effective in eliminating tumor cells and can be a complementary component of combinatorial therapy for GBM patients. Altogether, this study provided key data for instructing novel and rational combinatorial polytherapeutic approaches for the treatment of therapy-resistant GBM. / Thesis / Doctor of Philosophy (PhD) / Cancer is the leading cause of death in Canada and Glioblastoma Multiforme (GBM) is the most common type of malignant adult brain tumor which is one of the difficult human cancers to treat. In spite of the multi-model therapy including surgery, chemotherapy, and radiation, tumor re-growth and patient relapse are inevitable. A wealth of literature describes the biology of treatment-naïve or primary GBM, but we currently lack an understanding of how GBM evolves through therapy to become a very different tumor at recurrence, which may explain why therapies against primary GBM fail to work in recurrent GBM (rGBM). Clinical trials have not shown significant survival advantages for GBM patients, due not only to the lack of biological characterization of the distinct landscape of GBM recurrence, but also due to our poor understanding of the tumor immune microenvironment (TIME), the immune cells surrounding the tumor that may somehow fail to attack it due to GBM cells’ ability to suppress the immune system and evade detection. To understand how GBM cells and the TIME evolve under therapeutic pressure, we performed proteomics analysis on a large cohort of primary-recurrent GBM patient samples, to further understand treatment failure and develop effective and empirical combinatorial poly-therapies for the treatment of therapy-resistant GBM. Besides, in the next part of this study we showed existence of few cells within the tumor, termed brain tumor initiating cells (BTICs) can alone drive tumor growth and cause therapy resistance. To be able to target these population of cells, we identified treatment resistant tumor associated markers (highly expressed cell surface proteins) on these cells and developed novel treatments using a new class of biologics, Dual Antigen T cell Engagers (DATEs), to target these tumor associated markers. DATEs act like a “molecular glue” that specifically binds the patient’s own naturally circulating T-cells (soldier cells of the immune system) to cancer cells. Once bound, the T-cells attack and kill the patient’s cancer cells. In this strategy, the abnormal expression of tumor surface proteins can be used as a handle to drive T cell-mediated cell death. We predict that the dual specific antibodies through this study could be used alone or in combination with existing drugs to treat recurrent GBM.

Despite that most vestibular schwannomas are successfully treated with radiotherapy, current follow-up protocols entail years of serial magnetic resonance imaging (MRI) scans to ensure cessation of growth. This pilot study sought to identify early predictors of radiation treatment response using a non-invasive multi-modality imaging approach. We hypothesized that by combining information acquired from dynamic contrast-enhanced MRI (DCE-MRI), diffusion tensor imaging (DTI), and L-¹¹C-methionine positron emission tomography (MET-PET) treatment response could be identified sooner than the current several year waiting period. This thesis presents the baseline MRI and MET-PET results of the pilot study acquired to-date with follow-up data to be acquired in the next six months. Baseline results suggest that DTI and DCE-MRI yield information that may be useful in identifying the response of vestibular schwannomas to radiotherapy. In particular, vestibular schwannomas display elevated mean diffusion coefficients relative to the contra-lateral cerebellum. Also, the novel use of arterial input functions derived from the anterior inferior cerebellar arteries has led to the successful implementation of DCE-MRI pharmaco-kinetic models which may be used to quantitatively monitor tumor response to radiotherapy. Furthermore, MET-PET has shown promise as a tool for evaluating response as all tumors exhibited enhancement under this modality as compared to the contra-lateral side of the brain. Single-voxel spectroscopy with 3T MRI has proven to be a poor technique with which to examine vestibular schwannomas since only two of eight spectra were acquired successfully. All of the techniques that have shown promise as investigatory tools of tumor response can potentially be implemented clinically in the near future.

Cancer has become the major factor causing death in Taiwan. Although brain tumor take a few ratio in cancer, the current therapy shown not effect well. In previous studies, EGCG will promote cancer cells to go apoptosis after treatment with EGCG. But the mechanism in brain tumor is unknown. So, we use EGCG from green tea extract to investigate cell death of human brain tumor cell lines GBM8401, U87MG and U251. The metabolic rates of GBM8401, U87 and U251 are decrease with EGCG treatment, and it is significant in GBM8401. We also examine some genes and proteins related apoptosis by RT-PCR and Western blotting. The accumulation of p21, p27 and p53 are increasing with concentration and time course treatments. EGCG can inhibit the ability of brain tumor to form foci in anchorage-independent growth assay. Our data show that EGCG can inhibit brain tumor cell lines GBM8401, U87 and U251. It is validated that EGCG has the potential of cancer therapy and prevention.

Despite that most vestibular schwannomas are successfully treated with radiotherapy, current follow-up protocols entail years of serial magnetic resonance imaging (MRI) scans to ensure cessation of growth. This pilot study sought to identify early predictors of radiation treatment response using a non-invasive multi-modality imaging approach. We hypothesized that by combining information acquired from dynamic contrast-enhanced MRI (DCE-MRI), diffusion tensor imaging (DTI), and L-¹¹C-methionine positron emission tomography (MET-PET) treatment response could be identified sooner than the current several year waiting period. This thesis presents the baseline MRI and MET-PET results of the pilot study acquired to-date with follow-up data to be acquired in the next six months. Baseline results suggest that DTI and DCE-MRI yield information that may be useful in identifying the response of vestibular schwannomas to radiotherapy. In particular, vestibular schwannomas display elevated mean diffusion coefficients relative to the contra-lateral cerebellum. Also, the novel use of arterial input functions derived from the anterior inferior cerebellar arteries has led to the successful implementation of DCE-MRI pharmaco-kinetic models which may be used to quantitatively monitor tumor response to radiotherapy. Furthermore, MET-PET has shown promise as a tool for evaluating response as all tumors exhibited enhancement under this modality as compared to the contra-lateral side of the brain. Single-voxel spectroscopy with 3T MRI has proven to be a poor technique with which to examine vestibular schwannomas since only two of eight spectra were acquired successfully. All of the techniques that have shown promise as investigatory tools of tumor response can potentially be implemented clinically in the near future.

Despite that most vestibular schwannomas are successfully treated with radiotherapy, current follow-up protocols entail years of serial magnetic resonance imaging (MRI) scans to ensure cessation of growth. This pilot study sought to identify early predictors of radiation treatment response using a non-invasive multi-modality imaging approach. We hypothesized that by combining information acquired from dynamic contrast-enhanced MRI (DCE-MRI), diffusion tensor imaging (DTI), and L-¹¹C-methionine positron emission tomography (MET-PET) treatment response could be identified sooner than the current several year waiting period. This thesis presents the baseline MRI and MET-PET results of the pilot study acquired to-date with follow-up data to be acquired in the next six months. Baseline results suggest that DTI and DCE-MRI yield information that may be useful in identifying the response of vestibular schwannomas to radiotherapy. In particular, vestibular schwannomas display elevated mean diffusion coefficients relative to the contra-lateral cerebellum. Also, the novel use of arterial input functions derived from the anterior inferior cerebellar arteries has led to the successful implementation of DCE-MRI pharmaco-kinetic models which may be used to quantitatively monitor tumor response to radiotherapy. Furthermore, MET-PET has shown promise as a tool for evaluating response as all tumors exhibited enhancement under this modality as compared to the contra-lateral side of the brain. Single-voxel spectroscopy with 3T MRI has proven to be a poor technique with which to examine vestibular schwannomas since only two of eight spectra were acquired successfully. All of the techniques that have shown promise as investigatory tools of tumor response can potentially be implemented clinically in the near future. / Science, Faculty of / Physics and Astronomy, Department of / Graduate

Gliomas are the most common human brain tumors and are divided into four stages by WHO classification scheme. Benign gliomas are defined as grades I (Pilocytic astrocytomas) and II (Astrocytomas), whereas grade III (Anaplastic Astrocytomas, AA) and grade IV (Glioblastoma Multiforme, GBM) are malignant. Although both grades III and IV are malignant, the prognoses for these tumors are quite different. The 2-year survival rate for grade III gliomas is 50%, and grade IV is < 20 %. Mechanisms of tumorigenesis are not exactly elucidated in brain tumor cells. The thesis is to study the role of p27 kip1 in human malignant brain tumors. The experimental methods include ribonuclease protection assay (RPA), western blotting, immunohistochemical staining and immunocytochemical staining. mRNAs of p130, p107, Rb, p53 and p27 kip1 in normal brain tissues and brain tumors were overexpressed in most case. The p27kip1 mRNA were expressed in all astrocytomas and GBM, and mRNA quantity of p27kip1 were more in brain tumors than in normal brain tissues. PI3K/Akt pathway regulates several cellular functions such as cell survival and cell proliferation. Active Akt can phosphorylate p27kip1 that may contribute cell cycle from G1 phase to S phase. Skp2 identifies phospho-p27kip1 and promotes p27kip1 degradation. We found p27kip1 overexpression and Akt activation in astrocytomas and GBM. The expression of p-Akt were found in 20 %, 87 % and 71 % in normal brain tissues, astrocytomas and GBM, respectively. Expression of p27kip1 and p-Akt has shown significant correlation in GBM (P = 0.0236). Overexpression of p27kip1 mRNA in brain tumors may be consequence of p-Akt and Skp2.

Cancer has become the first among the 10 major death causing factors in Taiwan. Glioblastoma multiforme (GBM) is the most common malignant tumor in adult human brain tumors. Previously, heat shock or all trans retinoic acid (ATRA) treatment has been shown to be effective in inducing cell apoptosis and cell cycle arrest in several cancer cell lines. In this study, human brain tumor cell line GBM8401 was exposed to 43¢J for 30 min followed by incubation with ATRA. The treatment resulted in up to 50% inhibition of cell growth rate and 50% reduction of cell survival rate . Analysis of cell apoptosis related gene expression and protein expression with RT-PCR and Western blot has showen that p21, p27, pro-caspase 3, phospho-JNK and phospho-p38 were overexpressed after treatment of tumor cells with 43¢J for 30 min followed by addition of ATRA for 15 min to 8 hr. The immunocytochemistry assay revealed that overexpression of phospho-p53 in the nuclei after tumor cells were treated with 43¢J for 30 min followed by addition of ATRA for 8 hr. Results from this study show that treating tumor cells with heat shock before incubation with ATRA will enhance cell apoptosis and inhibit cell growth.

Glioblastoma (GBM) remains the most aggressive primary brain tumor in adults. Since 2005, Standard of Care (SoC) consists of surgical resection followed by radiation and adjuvant chemotherapy with temozolomide. Treatment failure is attributed to intratumoral heterogeneity with populations capable of mechanisms to repair damaged DNA. Given the lack of progress to improve patient outcomes, the current work encompasses how multi-omic approaches can be utilized to uncover novel biology in GBM and develop precision medicines to exploit these cancer specific phenomena. Using patient derived GBM samples I first used the surface marker CD133 to interrogate glioblastoma stem cells, a subpopulation of cells identified to withstand conventional therapies and lead to tumor relapse. I used a genome-wide CRISPR-Cas9 library to conduct an unbiased loss-of-function phenotypic screen to identify regulators of CD133. I then validated SOX2 as a direct transcription factor to PROM1 encoding CD133. These findings further show the untapped potential of CRISPR to uncover novel biology to directly apply to broader fields of stem cells and cancer biology. Next, I combed GBM data sets at transcriptomic and proteomic levels to identify understudied proteins as potential targets for immunotherapies. Glycoprotein nonmetastatic melanoma protein B (GPNMB) has previously been identified as a clinically relevant target in GBM and shown to be active in the tumor immune microenvironment. I found GPNMB to be upregulated in recurrent GBM and macrophage populations which can be exploited in a more comprehensive manner to treat GBM. Through a series of models, I elucidated how GPNMB influences GBM biology, its effectiveness as a target for Chimeric Antigen Receptor T-cells, and how it can be paired with CD133 therapies to provide better coverage of tumor cells. Together, these studies highlight how advances in pre-clinical models and technologies can be leveraged to develop new therapies in a rational manner. / Thesis / Doctor of Science (PhD) / Glioblastoma (GBM) remains an aggressive and incurable brain cancer despite decades of intense research. Treatment failure is due to the untargeted approaches currently undertaken in the clinic. The current work uses multiples methods to interrogate how GBM grows and develops over time. Using GBM samples from consenting patients, I investigated an important population of the tumor using a surface marker CD133 and CRISPR to study which genes influenced it. I then successfully validated SOX2 as a direct regulator of CD133 expression. Next, I combed multiple data sets for a target to kill GBM cells without harming healthy tissue in patients. I found Glycoprotein Non-Metastatic Melanoma Protein B (GPNMB) to be exploitable and used several experimental methods to investigate its role in GBM progression. Finally, we used a novel immunotherapy to eliminate cells which express GPNMB. Together, these findings could apply to the broader field of stem cell biology and be used for a more targeted method to eliminate the cancer entirely.